Python Bindings for CUPTI C APIs

The cupti.cupti module exposes 1:1 Python bindings for the CUPTI C API. For most users the higher-level cupti.pm_sampling and cupti.profiler_host modules are the recommended entry points; the bindings on this page are intended for the Activity and Callback APIs (which do not yet have a pythonic wrapper), and as raw-access fallbacks for PM Sampling and Profiler Host.

Known limitations

• The members of the inner struct/union classes (_py_anon_pod*) are not adequately documented. Refer to the CUPTI C documentation for member-level details.

• The kind member of Python classes has type int. Use cupti.cupti.ActivityKind to interpret its value.

• Some parts of the API still cite C enums / data structures by their C names instead of mapping them to their Python counterparts.

Activity API

Functions

cupti.cupti.activity_configure_unified_memory_counter(config, count: int)

Set Unified Memory Counter configuration.

Parameters:

• config (intptr_t) – A pointer to CUpti_ActivityUnifiedMemoryCounterConfig structures containing Unified Memory counter configuration.

• count (uint32_t) – Number of Unified Memory counter configuration structures.

See also

cuptiActivityConfigureUnifiedMemoryCounter

cupti.cupti.activity_disable(kind: int)

Disable collection of a specific kind of activity record.

Parameters:

kind (ActivityKind) – The kind of activity record to stop collecting.

See also

cuptiActivityDisable

cupti.cupti.activity_disable_context(context: int, kind: int)

Disable collection of a specific kind of activity record for a context.

Parameters:

• context (intptr_t) – The context for which activity is to be disabled.

• kind (ActivityKind) – The kind of activity record to stop collecting.

See also

cuptiActivityDisableContext

cupti.cupti.activity_disable_v2(subscriber: int, kind: int, p_activity_config)

Disable collection of a specific kind of activity record for the subscriber.

Parameters:

• subscriber (intptr_t) – The subscriber handle.

• kind (ActivityKind) – The kind of activity record to stop collecting.

• p_activity_config (intptr_t) – The activity config. It can be NULL.

See also

cuptiActivityDisable_v2

cupti.cupti.activity_enable(kind: int)

Enable collection of a specific kind of activity record.

Parameters:

kind (ActivityKind) – The kind of activity record to collect.

See also

cuptiActivityEnable

cupti.cupti.activity_enable_all_sync_records(enable: int)

Enables collecting records for all synchronization operations.

Parameters:

enable (uint8_t) – is a boolean, denoting whether to enable or disable the collection of all CUDA event query and stream query records.

See also

cuptiActivityEnableAllSyncRecords

cupti.cupti.activity_enable_allocation_source(enable: int)

Enables tracking the source library for memory allocation requests.

Parameters:

enable (uint8_t) – is a boolean, denoting whether the source library of the memory allocation request needs to be tracked.

See also

cuptiActivityEnableAllocationSource

cupti.cupti.activity_enable_and_dump(kind: int)

Enable collection of a specific kind of activity record. For certain activity kinds it dumps existing records.

Parameters:

kind (ActivityKind) – The kind of activity record to collect.

See also

cuptiActivityEnableAndDump

cupti.cupti.activity_enable_and_dump_v2(

subscriber: int,

kind: int,

activity_config,

)

Enable collection of a specific kind of activity record. For certain activity kinds it dumps existing records.

Parameters:

• subscriber (intptr_t) – The subscriber handle.

• kind (ActivityKind) – The kind of activity record to collect.

• activity_config (intptr_t) – The activity config. It can be NULL.

See also

cuptiActivityEnableAndDump_v2

cupti.cupti.activity_enable_context(context: int, kind: int)

Enable collection of a specific kind of activity record for a context.

Parameters:

• context (intptr_t) – The context for which activity is to be enabled.

• kind (ActivityKind) – The kind of activity record to collect.

See also

cuptiActivityEnableContext

cupti.cupti.activity_enable_cuda_event_device_timestamps(enable: int)

Enable/Disable collecting device timestamp for CUPTI_ACTIVITY_KIND_CUDA_EVENT record.

Parameters:

enable (uint8_t) – is a boolean, denoting whether to enable or disable the collection of CUDA event device timestamps.

See also

cuptiActivityEnableCudaEventDeviceTimestamps

cupti.cupti.activity_enable_device_graph(enable: int)

Controls the collection of records for device launched graphs.

Parameters:

enable (uint8_t) – is a boolean, denoting whether these records should be collected.

See also

cuptiActivityEnableDeviceGraph

cupti.cupti.activity_enable_driver_api(cbid: int, enable: int)

Controls the collection of activity records for specific CUDA Driver APIs.

Parameters:

• cbid (uint32_t) – callback id of the CUDA Driver API. This can be found in the header cupti_driver_cbid.h.

• enable (uint8_t) – is a boolean, denoting whether to enable or disable the collection.

See also

cuptiActivityEnableDriverApi

cupti.cupti.activity_enable_driver_api_v2(subscriber: int, cbid: int, enable: int)

Controls the collection of activity records for specific CUDA Driver APIs for a given subscriber.

Parameters:

• subscriber (intptr_t) – Handle to the subscriber. Must have been previously created via cuptiSubscribe_v2.

• cbid (uint32_t) – callback id of the CUDA Driver API.

• enable (uint8_t) – is a boolean, denoting whether to enable or disable the collection.

See also

cuptiActivityEnableDriverApi_v2

cupti.cupti.activity_enable_hw_trace(enable: int)

Enables CUDA kernel timestamp collection via Hardware Event System (HES).

Parameters:

enable (uint8_t) – is a boolean flag to enable (true) HES-based timestamp collection.

See also

cuptiActivityEnableHWTrace

cupti.cupti.activity_enable_latency_timestamps(enable: int)

Controls the collection of queued and submitted timestamps for kernels.

Parameters:

enable (uint8_t) – is a boolean, denoting whether these timestamps should be collected.

See also

cuptiActivityEnableLatencyTimestamps

cupti.cupti.activity_enable_launch_attributes(enable: int)

Controls the collection of launch attributes for kernels.

Parameters:

enable (uint8_t) – is a boolean denoting whether these launch attributes should be collected.

See also

cuptiActivityEnableLaunchAttributes

cupti.cupti.activity_enable_runtime_api(cbid: int, enable: int)

Controls the collection of activity records for specific CUDA Runtime APIs.

Parameters:

• cbid (uint32_t) – callback id of the CUDA Runtime API. This can be found in the header cupti_runtime_cbid.h.

• enable (uint8_t) – is a boolean, denoting whether to enable or disable the collection.

See also

cuptiActivityEnableRuntimeApi

cupti.cupti.activity_enable_runtime_api_v2(

subscriber: int,

cbid: int,

enable: int,

)

Controls the collection of activity records for specific CUDA Runtime APIs for a given subscriber.

Parameters:

• subscriber (intptr_t) – Handle to the subscriber. Must have been previously created via cuptiSubscribe_v2.

• cbid (uint32_t) – callback id of the CUDA Runtime API.

• enable (uint8_t) – is a boolean, denoting whether to enable or disable the collection.

See also

cuptiActivityEnableRuntimeApi_v2

cupti.cupti.activity_enable_v2(subscriber: int, kind: int, p_activity_config)

Enable collection of a specific kind of activity record for the subscriber.

Parameters:

• subscriber (intptr_t) – The subscriber handle.

• kind (ActivityKind) – The kind of activity record to collect.

• p_activity_config (intptr_t) – The activity config. It can be NULL.

See also

cuptiActivityEnable_v2

cupti.cupti.activity_flush_all(flag: int)

Request to deliver activity records via the buffer completion callback.

Parameters:

flag (uint32_t) – The flag can be set to indicate a forced flush. See CUpti_ActivityFlag.

See also

cuptiActivityFlushAll

cupti.cupti.activity_flush_period(time: int)

Sets the flush period for the worker thread.

Parameters:

time (uint32_t) – flush period in milliseconds (ms).

See also

cuptiActivityFlushPeriod

cupti.cupti.activity_get_attribute(attr: int, value_size: int, value: int)

Read an activity API attribute.

Parameters:

• attr (ActivityAttribute) – The attribute to read.

• value_size (intptr_t) – Size of buffer pointed by the value, and returns the number of bytes written to value.

• value (intptr_t) – Returns the value of the attribute.

See also

cuptiActivityGetAttribute

cupti.cupti.activity_get_attribute_v2(subscriber: int, attr: int)

Read a per-subscriber activity attribute as a plain Python value.

Parameters:

• subscriber (intptr_t) – The subscriber handle.

• attr (ActivityAttribute) – The attribute to read.

Returns:

For ATTR_TIMESTAMP_CALLBACK, the Python callable previously registered for this subscriber (or None if none is set). For every other attribute, the value as an int.

See also

cuptiActivityGetAttribute_v2

cupti.cupti.activity_get_enabled_kinds(subscriber: int)

Get the activity kinds currently enabled for a subscriber.

Parameters:

subscriber (intptr_t) – The subscriber handle. Pass 0 (NULL) to get the union of activity kinds enabled across all subscribers.

Returns:

the enabled activity kinds (empty list if none are enabled).

Return type:

list[ActivityKind]

See also

cuptiActivityGetEnabledKinds

cupti.cupti.activity_get_num_dropped_records(

context: int,

stream_id: int,

dropped: int,

)

Get the number of activity records that were dropped of insufficient buffer space.

Parameters:

• context (intptr_t) – The context, or NULL to get dropped count from global queue.

• stream_id (uint32_t) – The stream ID.

• dropped (intptr_t) – The number of records that were dropped since the last call to this function.

See also

cuptiActivityGetNumDroppedRecords

cupti.cupti.activity_get_num_dropped_records_v2(

subscriber: int,

context: int,

stream_id: int,

) → int

Get the number of dropped activity records for a specific subscriber.

Parameters:

• subscriber (intptr_t) – The subscriber handle.

• context (intptr_t) – CUDA context (reserved for future use).

• stream_id (uint32_t) – Stream ID (reserved for future use).

Returns:

Returns the number of dropped records.

Return type:

size_t

See also

cuptiActivityGetNumDroppedRecords_v2

cupti.cupti.activity_pop_external_correlation_id(kind: int) → int

Pop an external correlation id for the calling thread.

Parameters:

kind (ExternalCorrelationKind) – The kind of external API activities should be correlated with.

Returns:

If the function returns successful, contains the last

external correlation id for this kind, can be NULL.

Return type:

uint64_t

See also

cuptiActivityPopExternalCorrelationId

cupti.cupti.activity_pop_external_correlation_id_v2(

subscriber: int,

kind: int,

) → int

Pop an external correlation id for a specific subscriber.

Parameters:

• subscriber (intptr_t) – The subscriber handle.

• kind (ExternalCorrelationKind) – The kind of external API to pop correlation for.

Returns:

Returns the last external correlation id popped, can be

NULL.

Return type:

uint64_t

See also

cuptiActivityPopExternalCorrelationId_v2

cupti.cupti.activity_push_external_correlation_id(kind: int, id: int)

Push an external correlation id for the calling thread.

Parameters:

• kind (ExternalCorrelationKind) – The kind of external API activities should be correlated with.

• id (uint64_t) – External correlation id.

See also

cuptiActivityPushExternalCorrelationId

cupti.cupti.activity_push_external_correlation_id_v2(

subscriber: int,

kind: int,

id: int,

)

Push an external correlation id for a specific subscriber.

Parameters:

• subscriber (intptr_t) – The subscriber handle.

• kind (ExternalCorrelationKind) – The kind of external API activities should be correlated with.

• id (uint64_t) – External correlation id.

See also

cuptiActivityPushExternalCorrelationId_v2

cupti.cupti.activity_register_callbacks(

func_buffer_requested,

func_buffer_completed,

)

Registers callback functions with CUPTI for activity buffer handling.

Parameters:

• func_buffer_requested (object) – callback which is invoked when an empty buffer is requested by CUPTI.

• func_buffer_completed (object) – callback which is invoked when a buffer containing activity records is available from CUPTI.

See also

cuptiActivityRegisterCallbacks

cupti.cupti.activity_register_callbacks_v2(

subscriber: int,

func_buffer_requested,

func_buffer_completed,

)

Registers callback functions with CUPTI for activity buffer handling for the subscriber.

Parameters:

• subscriber (intptr_t) – The subscriber handle.

• func_buffer_requested (object) – callback which is invoked when an empty buffer is requested by CUPTI.

• func_buffer_completed (object) – callback which is invoked when a buffer containing activity records is available from CUPTI.

See also

cuptiActivityRegisterCallbacks_v2

cupti.cupti.activity_register_timestamp_callback(func_timestamp)

Registers callback function with CUPTI for providing timestamp.

Parameters:

func_timestamp (object) – callback which is invoked when a timestamp is needed by CUPTI.

See also

cuptiActivityRegisterTimestampCallback

cupti.cupti.activity_set_attribute(attr: int, value_size: int, value: int)

Write an activity API attribute.

Parameters:

• attr (ActivityAttribute) – The attribute to write.

• value_size (intptr_t) – The size, in bytes, of the value.

• value (intptr_t) – The attribute value to write.

See also

cuptiActivitySetAttribute

cupti.cupti.activity_set_attribute_v2(subscriber: int, attr: int, value)

Write a per-subscriber activity attribute from a plain Python value.

Parameters:

• subscriber (intptr_t) – The subscriber handle.

• attr (ActivityAttribute) – The attribute to write.

• value – For ATTR_TIMESTAMP_CALLBACK, a callable returning the current time as an int (nanoseconds), or None to clear it. For every other attribute, an integer value.

See also

cuptiActivitySetAttribute_v2

cupti.cupti.compute_capability_supported(major: int, minor: int) → int

Check support for a compute capability.

Parameters:

• major (int) – The major revision number of the compute capability.

• minor (int) – The minor revision number of the compute capability.

Returns:

Pointer to an integer to return the support status.

Return type:

int

See also

cuptiComputeCapabilitySupported

cupti.cupti.device_supported(dev: int) → int

Check support for a compute device.

Parameters:

dev (int) – The device handle returned by CUDA Driver API cuDeviceGet.

Returns:

Pointer to an integer to return the support status.

Return type:

int

See also

cuptiDeviceSupported

cupti.cupti.device_virtualization_mode(dev: int) → int

Query the virtualization mode of the device.

Parameters:

dev (int) – The device handle returned by CUDA Driver API cuDeviceGet.

Returns:

Pointer to an CUpti_DeviceVirtualizationMode to return

the virtualization mode.

Return type:

int

See also

cuptiDeviceVirtualizationMode

cupti.cupti.finalize()

Detach CUPTI from the running process.

See also

cuptiFinalize

cupti.cupti.get_auto_boost_state(context: int, state)

Get auto boost state.

Parameters:

• context (intptr_t) – A valid CUcontext.

• state (intptr_t) – A pointer to CUpti_ActivityAutoBoostState structure which contains the current state and the id of the process that has requested the current state.

See also

cuptiGetAutoBoostState

cupti.cupti.get_context_id(context: int) → int

Get the ID of a context.

Parameters:

context (intptr_t) – The context.

Returns:

Returns a process-unique ID for the context.

Return type:

uint32_t

See also

cuptiGetContextId

cupti.cupti.get_cuda_event_id(event: int) → int

Get the unique ID of a CUDA event.

Parameters:

event (intptr_t) – The CUDA event.

Returns:

Returns the unique ID of the event.

Return type:

uint32_t

See also

cuptiGetCudaEventId

cupti.cupti.get_device_id(context: int) → int

Get the ID of a device.

Parameters:

context (intptr_t) – The context, or NULL to indicate the current context.

Returns:

Returns the ID of the device that is current for the

calling thread.

Return type:

uint32_t

See also

cuptiGetDeviceId

cupti.cupti.get_graph_exec_id(graph_exec: int) → int

Get the unique ID of executable graph.

Parameters:

graph_exec (intptr_t) – The executable graph.

Returns:

Returns the unique ID of the executable graph.

Return type:

uint32_t

See also

cuptiGetGraphExecId

cupti.cupti.get_graph_id(graph: int) → int

Get the unique ID of graph.

Parameters:

graph (intptr_t) – The graph.

Returns:

Returns the unique ID of the graph.

Return type:

uint32_t

See also

cuptiGetGraphId

cupti.cupti.get_graph_node_id(node: int) → int

Get the unique ID of a graph node.

Parameters:

node (intptr_t) – The graph node.

Returns:

Returns the unique ID of the node.

Return type:

uint64_t

See also

cuptiGetGraphNodeId

cupti.cupti.get_last_error() → int

Returns the last error from a cupti call or callback.

See also

cuptiGetLastError

cupti.cupti.get_stream_id_ex(

context: int,

stream: int,

per_thread_stream: int,

) → int

Get the ID of a stream.

Parameters:

• context (intptr_t) – If non-NULL then the stream is checked to ensure that it belongs to this context. Typically this parameter should be null.

• stream (intptr_t) – The stream.

• per_thread_stream (uint8_t) – Flag to indicate if program is compiled for per-thread streams.

Returns:

Returns a context-unique ID for the stream.

Return type:

uint32_t

See also

cuptiGetStreamIdEx

cupti.cupti.get_thread_id_type() → int

Get the thread-id type.

Returns:

Returns the thread-id type currently used by CUPTI. See enum

CUpti_ActivityThreadIdType for the list of thread-id types.

Return type:

int

See also

cuptiGetThreadIdType

cupti.cupti.get_timestamp() → int

Get the CUPTI timestamp.

Returns:

Returns the CUPTI timestamp.

Return type:

uint64_t

See also

cuptiGetTimestamp

cupti.cupti.get_timestamp_v2(subscriber: int) → int

Get the CUPTI timestamp. If a subscriber registers a particular timestamp callback, this API will use that callback to get the timestamp. In CUDA 13.3, this API falls back to get_timestamp(). In CUDA 13.4 and later, this API provides the timestamp callback registered by the requesting subscriber.

Parameters:

subscriber (intptr_t) – The subscriber handle.

Returns:

Returns the CUPTI timestamp.

Return type:

uint64_t

See also

cuptiGetTimestamp_v2

cupti.cupti.get_version() → int

Get the CUPTI API version.

Returns:

Returns the version.

Return type:

uint32_t

See also

cuptiGetVersion

cupti.cupti.is_tracing_session_running() → int

Check whether a CUPTI tracing session is still running. This API returns true if a CUPTI library is already loaded by another CUPTI user by using any CUPTI API. This API returns false when CUPTI is finalized and there is no CUPTI currently loaded and active. Can be used to determine if it is safe to unload your CUPTI based tool. Note that this API itself does not load the CUPTI library, it merely checks if a CUPTI library is already loaded by another CUPTI user.

Returns:

Returns whether the tracing session is still running.

Return type:

uint8_t

See also

cuptiIsTracingSessionRunning

cupti.cupti.set_thread_id_type(type: int)

Set the thread-id type.

Parameters:

type (ActivityThreadIdType) – The thread-id type to set. See enum CUpti_ActivityThreadIdType for the list of thread-id types.

See also

cuptiSetThreadIdType

Enums

class cupti.cupti.ActivityAttribute(value)

Bases: IntEnum

Activity attributes. These attributes are used to control the behavior of the activity API.

See CUpti_ActivityAttribute.

ATTR_CIG_MODE = 2147483649

ATTR_COUNT = 25

ATTR_DEVICE_BUFFER_FORCE_INT = 2147483647

ATTR_DEVICE_BUFFER_POOL_LIMIT = 2

ATTR_DEVICE_BUFFER_PRE_ALLOCATE_VALUE = 6

ATTR_DEVICE_BUFFER_SIZE = 0

ATTR_DEVICE_BUFFER_SIZE_CDP = 1

ATTR_DEVICE_BUFFER_SIZE_DEVICE_GRAPHS = 10

ATTR_ENABLE_ALLOCATION_SOURCE_TRACKING = 14

ATTR_ENABLE_ALL_SYNC_RECORDS = 16

ATTR_ENABLE_CIG_MODE = 23

ATTR_ENABLE_CUDA_EVENT_DEVICE_TIMESTAMPS = 17

ATTR_ENABLE_DEVICE_GRAPH_TRACE = 19

ATTR_ENABLE_HES = 13

ATTR_ENABLE_KERNEL_LATENCY_TIMESTAMPS = 15

ATTR_ENABLE_KERNEL_LAUNCH_ATTRIBUTES = 18

ATTR_ENABLE_MULTI_SUBSCRIBER_GRAPH_LEVEL_TRACE = 2147483648

ATTR_ENABLE_MULTI_SUBSCRIBER_GRAPH_TRACE = 20

ATTR_HES_BUFFER_SIZE = 24

ATTR_MEM_ALLOCATION_TYPE_HOST_PINNED = 8

ATTR_MULTIPLE_SUBSCRIBER_STATE = 12

ATTR_PER_THREAD_BUFFER = 9

ATTR_PROFILING_SEMAPHORE_POOL_LIMIT = 4

ATTR_PROFILING_SEMAPHORE_POOL_SIZE = 3

ATTR_PROFILING_SEMAPHORE_PRE_ALLOCATE_VALUE = 7

ATTR_THREAD_ID_TYPE = 21

ATTR_TIMESTAMP_CALLBACK = 22

ATTR_USER_DEFINED_RECORDS = 11

ATTR_ZEROED_OUT_BUFFER = 5

class cupti.cupti.ActivityComputeApiKind(value)

Bases: IntEnum

The kind of a compute API.

See CUpti_ActivityComputeApiKind.

CUDA = 1

CUDA_MPS = 2

FORCE_INT = 2147483647

UNKNOWN = 0

class cupti.cupti.ActivityEnvironmentKind(value)

Bases: IntEnum

The kind of environment data. Used to indicate what type of data is being reported by an environment activity record.

See CUpti_ActivityEnvironmentKind.

COOLING = 4

COUNT = 5

FORCE_INT = 2147483647

POWER = 3

SPEED = 1

TEMPERATURE = 2

UNKNOWN = 0

class cupti.cupti.ActivityFlag(value)

Bases: IntEnum

Flags associated with activity records. Activity record flags. Flags can be combined by bitwise OR to associated multiple flags with an activity record. Each flag is specific to a certain activity kind, as noted below.

See CUpti_ActivityFlag.

DEVICE_ATTRIBUTE_CUDEVICE = 1

DEVICE_CONCURRENT_KERNELS = 1

FLUSH_FORCED = 1

FORCE_INT = 2147483647

GLOBAL_ACCESS_KIND_CACHED = 512

GLOBAL_ACCESS_KIND_LOAD = 256

GLOBAL_ACCESS_KIND_SIZE_MASK = 255

INSTRUCTION_CLASS_MASK = 510

INSTRUCTION_VALUE_INVALID = 1

MARKER_COLOR_ARGB = 2

MARKER_COLOR_NONE = 1

MARKER_INSTANTANEOUS = 1

MARKER_START = 2

MARKER_SYNC_ACQUIRE = 8

MARKER_SYNC_ACQUIRE_FAILED = 32

MARKER_SYNC_ACQUIRE_SUCCESS = 16

MARKER_SYNC_RELEASE = 64

MEMCPY_ASYNC = 1

MEMSET_ASYNC = 1

METRIC_OVERFLOWED = 1

METRIC_VALUE_INVALID = 2

NONE = 0

SHARED_ACCESS_KIND_LOAD = 256

SHARED_ACCESS_KIND_SIZE_MASK = 255

THRASHING_IN_CPU = 1

THROTTLING_IN_CPU = 1

class cupti.cupti.ActivityInstructionClass(value)

Bases: IntEnum

SASS instruction classification. The sass instruction are broadly divided into different class. Each enum represents a classification.

See CUpti_ActivityInstructionClass.

BARRIER = 17

BIT_CONVERSION = 4

CONSTANT = 11

CONTROL_FLOW = 5

FP_16 = 19

FP_32 = 1

FP_64 = 2

GENERIC = 9

GLOBAL = 6

GLOBAL_ATOMIC = 13

INTEGER = 3

INTER_THREAD_COMMUNICATION = 16

KIND_FORCE_INT = 2147483647

LOCAL = 8

MISCELLANEOUS = 18

SHARED = 7

SHARED_ATOMIC = 14

SURFACE = 10

SURFACE_ATOMIC = 15

TEXTURE = 12

UNIFORM = 20

UNKNOWN = 0

class cupti.cupti.ActivityJitEntryType(value)

Bases: IntEnum

The types of JIT entry. To be used in CUpti_ActivityJit.

See CUpti_ActivityJitEntryType.

FORCE_INT = 2147483647

INVALID = 0

NVVM_IR_TO_PTX = 2

PTX_TO_CUBIN = 1

class cupti.cupti.ActivityJitOperationType(value)

Bases: IntEnum

The types of JIT compilation operations. To be used in CUpti_ActivityJit.

See CUpti_ActivityJitOperationType.

CACHE_LOAD = 1

CACHE_STORE = 2

COMPILE = 3

FORCE_INT = 2147483647

INVALID = 0

class cupti.cupti.ActivityKind(value)

Bases: IntEnum

The kinds of activity records. Each activity record kind represents information about a GPU or an activity occurring on a CPU or GPU. Each kind is associated with a activity record structure that holds the information associated with the kind.

See CUpti_ActivityKind.

BRANCH = 16

CDP_KERNEL = 18

COMPUTE_ENGINE_CTX_SWITCH = 57

CONCURRENT_KERNEL = 10

CONTEXT = 9

COUNT = 60

CUDA_EVENT = 36

DEVICE = 8

DEVICE_ATTRIBUTE = 28

DEVICE_GRAPH_TRACE = 53

DRIVER = 4

ENVIRONMENT = 20

EVENT = 6

EVENT_INSTANCE = 21

EXTERNAL_CORRELATION = 39

FORCE_INT = 2147483647

FUNCTION = 26

GLOBAL_ACCESS = 15

GRAPH_HOST_NODE = 56

GRAPH_TRACE = 51

GREEN_CONTEXT = 59

HOST_LAUNCH = 58

INSTANTANEOUS_EVENT = 41

INSTANTANEOUS_EVENT_INSTANCE = 42

INSTANTANEOUS_METRIC = 43

INSTANTANEOUS_METRIC_INSTANCE = 44

INSTRUCTION_CORRELATION = 32

INSTRUCTION_EXECUTION = 24

INTERNAL_LAUNCH_API = 48

INVALID = 0

JIT = 52

KERNEL = 3

MARKER = 12

MARKER_DATA = 13

MEMCPY = 1

MEMCPY2 = 22

MEMORY = 45

MEMORY2 = 49

MEMORY_POOL = 50

MEMSET = 2

MEM_DECOMPRESS = 54

METRIC = 7

METRIC_INSTANCE = 23

MODULE = 27

NAME = 11

NVLINK = 40

OPENACC_DATA = 33

OPENACC_LAUNCH = 34

OPENACC_OTHER = 35

OPENMP = 47

OVERHEAD = 17

PCIE = 46

PC_SAMPLING = 30

PC_SAMPLING_RECORD_INFO = 31

PREEMPTION = 19

ROTATION = 55

RUNTIME = 5

SHARED_ACCESS = 29

SOURCE_LOCATOR = 14

STREAM = 37

SYNCHRONIZATION = 38

UNIFIED_MEMORY_COUNTER = 25

class cupti.cupti.ActivityLaunchType(value)

Bases: IntEnum

The type of the CUDA kernel launch.

See CUpti_ActivityLaunchType.

CBL_COMMANDLIST = 3

COOPERATIVE_MULTI_DEVICE = 2

COOPERATIVE_SINGLE_DEVICE = 1

REGULAR = 0

class cupti.cupti.ActivityMemcpyKind(value)

Bases: IntEnum

The kind of a memory copy, indicating the source and destination targets of the copy. Each kind represents the source and destination targets of a memory copy. Targets are host, device, and array.

See CUpti_ActivityMemcpyKind.

ATOA = 5

ATOD = 6

ATOH = 4

DTOA = 7

DTOD = 8

DTOH = 2

FORCE_INT = 2147483647

HTOA = 3

HTOD = 1

HTOH = 9

PTOP = 10

UNKNOWN = 0

class cupti.cupti.ActivityMemoryKind(value)

Bases: IntEnum

The kinds of memory accessed by a memory operation/copy. Each kind represents the type of the memory accessed by a memory operation/copy.

See CUpti_ActivityMemoryKind.

ARRAY = 4

DEVICE = 3

DEVICE_STATIC = 6

FORCE_INT = 2147483647

MANAGED = 5

MANAGED_STATIC = 7

PAGEABLE = 1

PINNED = 2

UNKNOWN = 0

class cupti.cupti.ActivityMemoryOperationType(value)

Bases: IntEnum

Memory operation types. Describes the type of memory operation, to be used with CUpti_ActivityMemory4.

See CUpti_ActivityMemoryOperationType.

ALLOCATION = 1

FORCE_INT = 2147483647

INVALID = 0

RELEASE = 2

class cupti.cupti.ActivityMemoryPoolOperationType(value)

Bases: IntEnum

Memory pool operation types. Describes the type of memory pool operation, to be used with CUpti_ActivityMemoryPool2.

See CUpti_ActivityMemoryPoolOperationType.

CREATED = 1

DESTROYED = 2

FORCE_INT = 2147483647

INVALID = 0

TRIMMED = 3

class cupti.cupti.ActivityMemoryPoolType(value)

Bases: IntEnum

Memory pool types. Describes the type of memory pool, to be used with CUpti_ActivityMemory4.

See CUpti_ActivityMemoryPoolType.

FORCE_INT = 2147483647

IMPORTED = 2

INVALID = 0

LOCAL = 1

class cupti.cupti.ActivityObjectKind(value)

Bases: IntEnum

The kinds of activity objects.

See CUpti_ActivityObjectKind.

CONTEXT = 4

DEVICE = 3

FORCE_INT = 2147483647

PROCESS = 1

STREAM = 5

THREAD = 2

UNKNOWN = 0

class cupti.cupti.ActivityOverheadKind(value)

Bases: IntEnum

The kinds of activity overhead.

See CUpti_ActivityOverheadKind.

ACTIVITY_BUFFER_REQUEST = 458752

COMMAND_BUFFER_FULL = 393216

CUPTI_BUFFER_FLUSH = 65536

CUPTI_INSTRUMENTATION = 131072

CUPTI_RESOURCE = 196608

DRIVER_COMPILER = 1

FORCE_INT = 2147483647

LAZY_FUNCTION_LOADING = 327680

RUNTIME_TRIGGERED_MODULE_LOADING = 262144

UNKNOWN = 0

UVM_ACTIVITY_INIT = 524288

class cupti.cupti.ActivityPCSamplingPeriod(value)

Bases: IntEnum

Sampling period for PC sampling method. Sampling period can be set using cuptiActivityConfigurePCSampling

See CUpti_ActivityPCSamplingPeriod.

FORCE_INT = 2147483647

HIGH = 4

INVALID = 0

LOW = 2

MAX = 5

MID = 3

MIN = 1

class cupti.cupti.ActivityPCSamplingStallReason(value)

Bases: IntEnum

The stall reason for PC sampling activity.

See CUpti_ActivityPCSamplingStallReason.

CONSTANT_MEMORY_DEPENDENCY = 7

EXEC_DEPENDENCY = 3

FORCE_INT = 2147483647

INST_FETCH = 2

INVALID = 0

MEMORY_DEPENDENCY = 4

MEMORY_THROTTLE = 9

NONE = 1

NOT_SELECTED = 10

OTHER = 11

PIPE_BUSY = 8

SLEEPING = 12

SYNC = 6

TEXTURE = 5

class cupti.cupti.ActivityPartitionedGlobalCacheConfig(value)

Bases: IntEnum

Partitioned global caching option.

See CUpti_ActivityPartitionedGlobalCacheConfig.

FORCE_INT = 2147483647

NOT_SUPPORTED = 1

OFF = 2

ON = 3

UNKNOWN = 0

class cupti.cupti.ActivityPreemptionKind(value)

Bases: IntEnum

The kind of a preemption activity.

See CUpti_ActivityPreemptionKind.

FORCE_INT = 2147483647

RESTORE = 2

SAVE = 1

UNKNOWN = 0

class cupti.cupti.ActivityStreamFlag(value)

Bases: IntEnum

stream type. The types of stream to be used with CUpti_ActivityStream.

See CUpti_ActivityStreamFlag.

FLAG_DEFAULT = 1

FLAG_FORCE_INT = 2147483647

FLAG_NON_BLOCKING = 2

FLAG_NULL = 3

FLAG_UNKNOWN = 0

MASK = 65535

class cupti.cupti.ActivitySynchronizationType(value)

Bases: IntEnum

Synchronization type. The types of synchronization to be used with CUpti_ActivitySynchronization2.

See CUpti_ActivitySynchronizationType.

CONTEXT_SYNCHRONIZE = 4

EVENT_SYNCHRONIZE = 1

FORCE_INT = 2147483647

STREAM_SYNCHRONIZE = 3

STREAM_WAIT_EVENT = 2

UNKNOWN = 0

class cupti.cupti.ActivityThreadIdType(value)

Bases: IntEnum

Thread-Id types. CUPTI uses different methods to obtain the thread-id depending on the support and the underlying platform. This enum documents these methods for each type. APIs cuptiSetThreadIdType and cuptiGetThreadIdType can be used to set and get the thread-id type.

See CUpti_ActivityThreadIdType.

DEFAULT = 0

FORCE_INT = 2147483647

SIZE = 2

SYSTEM = 1

class cupti.cupti.ActivityUnifiedMemoryAccessType(value)

Bases: IntEnum

Memory access type for unified memory page faults. This is valid for CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_GPU_PAGE_FAULT and CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_CPU_PAGE_FAULT_COUNT

See CUpti_ActivityUnifiedMemoryAccessType.

ATOMIC = 3

PREFETCH = 4

READ = 1

UNKNOWN = 0

WRITE = 2

class cupti.cupti.ActivityUnifiedMemoryCounterKind(value)

Bases: IntEnum

Kind of the Unified Memory counter. Many activities are associated with Unified Memory mechanism; among them are transfers from host to device, device to host, page fault at host side.

See CUpti_ActivityUnifiedMemoryCounterKind.

BYTES_TRANSFER_DTOD = 8

BYTES_TRANSFER_DTOH = 2

BYTES_TRANSFER_HTOD = 1

COUNT = 9

CPU_PAGE_FAULT_COUNT = 3

FORCE_INT = 2147483647

GPU_PAGE_FAULT = 4

REMOTE_MAP = 7

THRASHING = 5

THROTTLING = 6

UNKNOWN = 0

class cupti.cupti.ActivityUnifiedMemoryCounterScope(value)

Bases: IntEnum

Scope of the unified memory counter (deprecated in CUDA 7.0).

See CUpti_ActivityUnifiedMemoryCounterScope.

COUNT = 3

FORCE_INT = 2147483647

PROCESS_ALL_DEVICES = 2

PROCESS_SINGLE_DEVICE = 1

UNKNOWN = 0

class cupti.cupti.ActivityUnifiedMemoryMigrationCause(value)

Bases: IntEnum

Migration cause of the Unified Memory counter. This is valid for CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_BYTES_TRANSFER_HTOD and CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_BYTES_TRANSFER_DTOH and CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_BYTES_TRANSFER_DTOD

See CUpti_ActivityUnifiedMemoryMigrationCause.

ACCESS_COUNTERS = 5

COHERENCE = 2

EVICTION = 4

PREFETCH = 3

UNKNOWN = 0

USER = 1

class cupti.cupti.ActivityUnifiedMemoryRemoteMapCause(value)

Bases: IntEnum

Remote memory map cause of the Unified Memory counter. This is valid for CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_REMOTE_MAP

See CUpti_ActivityUnifiedMemoryRemoteMapCause.

COHERENCE = 1

EVICTION = 5

OUT_OF_MEMORY = 4

POLICY = 3

THRASHING = 2

UNKNOWN = 0

class cupti.cupti.ChannelType(value)

Bases: IntEnum

See CUpti_ChannelType.

ASYNC_MEMCPY = 2

COMPUTE = 1

DECOMP = 3

FORCE_INT = 2147483647

INVALID = 0

class cupti.cupti.ComputeEngineCtxSwitchOperationType(value)

Bases: IntEnum

The operation type of CUDA context switch event records.

See CUpti_ComputeEngineCtxSwitchOperationType.

COUNT = 2147483647

INVALID = 0

START = 1

class cupti.cupti.ConfidentialComputeRotationEventType(value)

Bases: IntEnum

See CUpti_ConfidentialComputeRotationEventType.

EVENT_TYPE_FORCE_INT = 2147483647

INVALID_ROTATION_EVENT = 0

KEY_ROTATION_ACKNOWLEGED = 2

KEY_ROTATION_CHANNEL_BLOCKED = 1

KEY_ROTATION_CHANNEL_DRAINED = 2

KEY_ROTATION_CHANNEL_UNBLOCKED = 3

KEY_ROTATION_COMPLETED = 4

KEY_ROTATION_REQUESTED = 1

KEY_ROTATION_STARTED = 3

class cupti.cupti.ContextCigMode(value)

Bases: IntEnum

CIG (CUDA in Graphics) Modes. Describes the CIG modes associated with the CUDA context.

See CUpti_ContextCigMode.

CIG = 1

CIG_FALLBACK = 2

FORCE_INT = 2147483647

NONE = 0

class cupti.cupti.DevType(value)

Bases: IntEnum

The device type for device connected to NVLink.

See CUpti_DevType.

FORCE_INT = 2147483647

GPU = 1

INVALID = 0

NPU = 2

class cupti.cupti.DeviceAttribute(value)

Bases: IntEnum

Device attributes. CUPTI device attributes. These attributes can be read using cuptiDeviceGetAttribute.

See CUpti_DeviceAttribute.

ATTR_CLASS = 10

ATTR_FLOP_DP_PER_CYCLE = 12

ATTR_FLOP_HP_PER_CYCLE = 17

ATTR_FLOP_SP_PER_CYCLE = 11

ATTR_FORCE_INT = 2147483647

ATTR_GLOBAL_MEMORY_BANDWIDTH = 3

ATTR_GPU_CPU_NVLINK_BW = 19

ATTR_INSTRUCTION_PER_CYCLE = 4

ATTR_INSTRUCTION_THROUGHPUT_SINGLE_PRECISION = 5

ATTR_MAX_EVENT_DOMAIN_ID = 2

ATTR_MAX_EVENT_ID = 1

ATTR_MAX_FRAME_BUFFERS = 6

ATTR_MAX_L2_UNITS = 13

ATTR_MAX_SHARED_MEMORY_CACHE_CONFIG_PREFER_EQUAL = 16

ATTR_MAX_SHARED_MEMORY_CACHE_CONFIG_PREFER_L1 = 15

ATTR_MAX_SHARED_MEMORY_CACHE_CONFIG_PREFER_SHARED = 14

ATTR_NVLINK_PRESENT = 18

ATTR_NVSWITCH_PRESENT = 20

ATTR_PCIE_GEN = 9

ATTR_PCIE_LINK_RATE = 7

ATTR_PCIE_LINK_WIDTH = 8

class cupti.cupti.DeviceVirtualizationMode(value)

Bases: IntEnum

This indicates the virtualization mode in which CUDA device is running

See CUpti_DeviceVirtualizationMode.

FORCE_INT = 2147483647

NONE = 0

PASS_THROUGH = 1

VIRTUAL_GPU = 2

class cupti.cupti.EnvironmentClocksThrottleReason(value)

Bases: IntEnum

Reasons for clock throttling. The possible reasons that a clock can be throttled. There can be more than one reason that a clock is being throttled so these types can be combined by bitwise OR. These are used in the clocksThrottleReason field in the Environment Activity Record.

See CUpti_EnvironmentClocksThrottleReason.

FORCE_INT = 2147483647

GPU_IDLE = 1

HW_SLOWDOWN = 8

NONE = 0

SW_POWER_CAP = 4

UNKNOWN = 2147483648

UNSUPPORTED = 1073741824

USER_DEFINED_CLOCKS = 2

class cupti.cupti.ExternalCorrelationKind(value)

Bases: IntEnum

The kind of external APIs supported for correlation. Custom correlation kinds are reserved for usage in external tools.

See CUpti_ExternalCorrelationKind.

CUSTOM0 = 3

CUSTOM1 = 4

CUSTOM2 = 5

FORCE_INT = 2147483647

INVALID = 0

OPENACC = 2

SIZE = 6

UNKNOWN = 1

class cupti.cupti.FuncExecutionModel(value)

Bases: IntEnum

The execution model of a kernel function. This should be used to set executionModel field in kernel records.

See CUpti_FuncExecutionModel.

FORCE_INT = 2147483647

HYBRID = 3

SIMT = 1

SIZE = 4

TILE = 2

UNKNOWN = 0

class cupti.cupti.FuncShmemLimitConfig(value)

Bases: IntEnum

The shared memory limit per block config for a kernel This should be used to set ‘cudaOccFuncShmemConfig’ field in occupancy calculator API.

See CUpti_FuncShmemLimitConfig.

DEFAULT = 0

FORCE_INT = 2147483647

OPTIN = 1

class cupti.cupti.OpenAccConstructKind(value)

Bases: IntEnum

The OpenAcc parent construct kind for OpenAcc activity records.

See CUpti_OpenAccConstructKind.

ATOMIC = 8

DATA = 4

DECLARE = 9

ENTER_DATA = 5

EXIT_DATA = 6

FORCE_INT = 2147483647

HOST_DATA = 7

INIT = 10

KERNELS = 2

LOOP = 3

PARALLEL = 1

ROUTINE = 14

RUNTIME_API = 16

SET = 12

SHUTDOWN = 11

UNKNOWN = 0

UPDATE = 13

WAIT = 15

class cupti.cupti.OpenAccEventKind(value)

Bases: IntEnum

The OpenAcc event kind for OpenAcc activity records.

See CUpti_OpenAccEventKind.

ALLOC = 15

COMPUTE_CONSTRUCT = 9

CREATE = 13

DELETE = 14

DEVICE_INIT = 1

DEVICE_SHUTDOWN = 2

ENQUEUE_DOWNLOAD = 6

ENQUEUE_LAUNCH = 4

ENQUEUE_UPLOAD = 5

ENTER_DATA = 11

EXIT_DATA = 12

FORCE_INT = 2147483647

FREE = 16

IMPLICIT_WAIT = 8

INVALID = 0

RUNTIME_SHUTDOWN = 3

UPDATE = 10

WAIT = 7

class cupti.cupti.OpenMpEventKind(value)

Bases: IntEnum

See CUpti_OpenMpEventKind.

FORCE_INT = 2147483647

IDLE = 4

INVALID = 0

PARALLEL = 1

TASK = 2

THREAD = 3

WAIT_BARRIER = 5

WAIT_TASKWAIT = 6

class cupti.cupti.PcieDeviceType(value)

Bases: IntEnum

Field to differentiate whether PCIE Activity record is of a GPU or a PCI Bridge

See CUpti_PcieDeviceType.

BRIDGE = 1

FORCE_INT = 2147483647

GPU = 0

Classes

class cupti.cupti.ActivityAPI

Bases: object

Empty-initialize an instance of CUpti_ActivityAPI.

See also

CUpti_ActivityAPI

cbid

The ID of the driver or runtime function.

Type:

int

correlation_id

The correlation ID of the driver or runtime CUDA function. Each function invocation is assigned a unique correlation ID that is identical to the correlation ID in the memcpy, memset, or kernel activity record that is associated with this function.

Type:

int

end

The end timestamp for the function, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the function.

Type:

int

static from_buffer(buffer)

Create an ActivityAPI instance with the memory from the given buffer.

static from_data(data)

Create an ActivityAPI instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_api_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityAPI instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_DRIVER, CUPTI_ACTIVITY_KIND_RUNTIME, or CUPTI_ACTIVITY_KIND_INTERNAL_LAUNCH_API.

Type:

int

process_id

The ID of the process where the driver or runtime CUDA function is executing.

Type:

int

ptr

Get the pointer address to the data as Python int.

return_value

The return value for the function. For a CUDA driver function with will be a CUresult value, and for a CUDA runtime function this will be a cudaError_t value.

Type:

int

start

The start timestamp for the function, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the function.

Type:

int

thread_id

The ID of the thread where the driver or runtime CUDA function is executing.

Type:

int

class cupti.cupti.ActivityApiCbidOptions

Bases: object

Empty-initialize an instance of CUpti_ActivityApiCbidOptions.

See also

CUpti_ActivityApiCbidOptions

enable

1 = allowlist (record only the listed cbids); 0 = denylist (suppress the listed cbids).

Type:

int

static from_buffer(buffer)

Create an ActivityApiCbidOptions instance with the memory from the given buffer.

static from_data(data)

Create an ActivityApiCbidOptions instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_api_cbid_options_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityApiCbidOptions instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_cbids

Array of callback IDs to allow or deny. Must not be NULL when numCbids > 0.

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

Size of the data structure. CUPTI client should set the size of the structure. It will be used in CUPTI to check what fields are available in the structure. Used to preserve backward compatibility.

Type:

int

class cupti.cupti.ActivityApiOptions

Bases: object

Empty-initialize an instance of CUpti_ActivityApiOptions.

See also

CUpti_ActivityApiOptions

static from_buffer(buffer)

Create an ActivityApiOptions instance with the memory from the given buffer.

static from_data(data)

Create an ActivityApiOptions instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_api_options_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityApiOptions instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_cbid_options

Pointer to per-cbid filter options. If NULL, all cbids for the kind are enabled.

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

Size of the data structure. CUPTI client should set the size of the structure. It will be used in CUPTI to check what fields are available in the structure. Used to preserve backward compatibility.

Type:

int

class cupti.cupti.ActivityAutoBoostState

Bases: object

Empty-initialize an instance of CUpti_ActivityAutoBoostState.

See also

CUpti_ActivityAutoBoostState

enabled

Returned auto boost state. 1 is returned in case auto boost is enabled, 0 otherwise

Type:

int

static from_buffer(buffer)

Create an ActivityAutoBoostState instance with the memory from the given buffer.

static from_data(data)

Create an ActivityAutoBoostState instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_auto_boost_state_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityAutoBoostState instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pid

Id of process that has set the current boost state. The value will be CUPTI_AUTO_BOOST_INVALID_CLIENT_PID if the user does not have the permission to query process ids or there is an error in querying the process id.

Type:

int

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.ActivityCdpKernel

Bases: object

Empty-initialize an instance of CUpti_ActivityCdpKernel.

See also

CUpti_ActivityCdpKernel

block_x

The X-dimension block size for the kernel.

Type:

int

block_y

The Y-dimension block size for the kernel.

Type:

int

block_z

The Z-dimension grid size for the kernel.

Type:

int

cache_config

_py_anon_pod9:

completed

The timestamp when kernel is marked as completed, in ns. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the completion time is unknown.

Type:

int

context_id

The ID of the context where the kernel is executing.

Type:

int

correlation_id

The correlation ID of the kernel. Each kernel execution is assigned a unique correlation ID that is identical to the correlation ID in the driver API activity record that launched the kernel.

Type:

int

device_id

The ID of the device where the kernel is executing.

Type:

int

dynamic_shared_memory

The dynamic shared memory reserved for the kernel, in bytes.

Type:

int

end

The end timestamp for the kernel execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the kernel.

Type:

int

static from_buffer(buffer)

Create an ActivityCdpKernel instance with the memory from the given buffer.

static from_data(data)

Create an ActivityCdpKernel instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_cdp_kernel_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityCdpKernel instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

grid_id

The grid ID of the kernel. Each kernel execution is assigned a unique grid ID.

Type:

int

grid_x

The X-dimension grid size for the kernel.

Type:

int

grid_y

The Y-dimension grid size for the kernel.

Type:

int

grid_z

The Z-dimension grid size for the kernel.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_CDP_KERNEL

Type:

int

local_memory_per_thread

The amount of local memory reserved for each thread, in bytes.

Type:

int

local_memory_total

The total amount of local memory reserved for the kernel, in bytes.

Type:

int

name

The name of the kernel. This name is shared across all activity records representing the same kernel, and so should not be modified.

Type:

str

pad

Undefined. Reserved for internal use.

Type:

int

parent_block_x

The X-dimension of the parent block.

Type:

int

parent_block_y

The Y-dimension of the parent block.

Type:

int

parent_block_z

The Z-dimension of the parent block.

Type:

int

parent_grid_id

The grid ID of the parent kernel.

Type:

int

ptr

Get the pointer address to the data as Python int.

queued

The timestamp when kernel is queued up, in ns. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the queued time is unknown.

Type:

int

registers_per_thread

The number of registers required for each thread executing the kernel.

Type:

int

shared_memory_config

The shared memory configuration used for the kernel. The value is one of the CUsharedconfig enumeration values from cuda.h.

Type:

int

start

The start timestamp for the kernel execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the kernel.

Type:

int

static_shared_memory

The static shared memory allocated for the kernel, in bytes.

Type:

int

stream_id

The ID of the stream where the kernel is executing.

Type:

int

submitted

The timestamp when kernel is submitted to the gpu, in ns. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the submission time is unknown.

Type:

int

class cupti.cupti.ActivityComputeEngineCtxSwitch

Bases: object

Empty-initialize an instance of CUpti_ActivityComputeEngineCtxSwitch.

See also

CUpti_ActivityComputeEngineCtxSwitch

context_id

The ID of the CUDA context.

Type:

int

static from_buffer(buffer)

Create an ActivityComputeEngineCtxSwitch instance with the memory from the given buffer.

static from_data(data)

Create an ActivityComputeEngineCtxSwitch instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_compute_engine_ctx_switch_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityComputeEngineCtxSwitch instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_COMPUTE_ENGINE_CTX_SWITCH

Type:

int

operation_type

The type of the Compute Engine Context switch operation. CUPTI_COMPUTE_ENGINE_CTX_SWITCH_OPERATION_START indicates the start of the context switch operation. CUPTI_COMPUTE_ENGINE_CTX_SWITCH_OPERATION_END indicates the end of the context switch operation.

Type:

int

ptr

Get the pointer address to the data as Python int.

timestamp

The timestamp at which the CUpti_ComputeEngineCtxSwitchOperationType occurs.

Type:

int

class cupti.cupti.ActivityConfidentialComputeRotation

Bases: object

Empty-initialize an instance of CUpti_ActivityConfidentialComputeRotation.

See also

CUpti_ActivityConfidentialComputeRotation

channel_id

Channel ID

Type:

int

channel_type

Channel Type CUpti_ChannelType

Type:

int

context_id

Context ID

Type:

int

device_id

Device ID

Type:

int

event_type

Type of event CUpti_ConfidentialComputeRotationEventType

Type:

int

static from_buffer(buffer)

Create an ActivityConfidentialComputeRotation instance with the memory from the given buffer.

static from_data(data)

Create an ActivityConfidentialComputeRotation instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_confidential_compute_rotation_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityConfidentialComputeRotation instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_CONFIDENTIAL_COMPUTE_ROTATION.

Type:

int

ptr

Get the pointer address to the data as Python int.

timestamp

Timestamp in ns

Type:

int

class cupti.cupti.ActivityConfig

Bases: object

Empty-initialize an instance of CUpti_ActivityConfig.

See also

CUpti_ActivityConfig

static from_buffer(buffer)

Create an ActivityConfig instance with the memory from the given buffer.

static from_data(data)

Create an ActivityConfig instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_config_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityConfig instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_kind_options

Optional pointer to kind-specific options. For CUPTI_ACTIVITY_KIND_RUNTIME and CUPTI_ACTIVITY_KIND_DRIVER, this may point to a CUpti_ActivityApiOptions struct to enable allowlist or denylist filtering of individual cbids. NULL disables filtering (all cbids enabled). Ignored for all other activity kinds. Available since CUPTI_API_VERSION 130400.

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

Size of the data structure. CUPTI client should set the size of the structure. It will be used in CUPTI to check what fields are available in the structure. Used to preserve backward compatibility.

Type:

int

class cupti.cupti.ActivityContext4

Bases: object

Empty-initialize an instance of CUpti_ActivityContext4.

See also

CUpti_ActivityContext4

cig_mode

This field indicates the CIG mode

Type:

int

compute_api_kind

The compute API kind. CUpti_ActivityComputeApiKind

Type:

int

context_id

The context ID.

Type:

int

device_id

The device ID.

Type:

int

static from_buffer(buffer)

Create an ActivityContext4 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityContext4 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_context4_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityContext4 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

is_green_context

This field indicates whether the context is a green context

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_CONTEXT.

Type:

int

null_stream_id

The ID for the NULL stream in this context

Type:

int

num_multiprocessors

Number of multiprocessors assigned to the green context Invalid if the field ‘isGreenContext’ is 0

Type:

int

parent_context_id

The ID of the parent context. It would be 0 if context does not have parent

Type:

int

process_id

The ID of the process associated with the context.

Type:

int

ptr

Get the pointer address to the data as Python int.

Deprecated since version 13.2.0.

class cupti.cupti.ActivityContext3

Bases: object

Empty-initialize an instance of CUpti_ActivityContext3.

See also

CUpti_ActivityContext3

cig_mode

This field indicates the CIG mode

Type:

int

compute_api_kind

The compute API kind. CUpti_ActivityComputeApiKind

Type:

int

context_id

The context ID.

Type:

int

device_id

The device ID.

Type:

int

static from_buffer(buffer)

Create an ActivityContext3 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityContext3 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_context3_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityContext3 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

is_green_context

This field indicates whether the context is a green context

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_CONTEXT.

Type:

int

null_stream_id

The ID for the NULL stream in this context

Type:

int

num_multiprocessors

Number of multiprocessors assigned to the green context Invalid if the field ‘isGreenContext’ is 0

Type:

int

parent_context_id

The ID of the parent context. It would be 0 if context does not have parent

Type:

int

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.ActivityCudaEvent3

Bases: object

Empty-initialize an instance of CUpti_ActivityCudaEvent3.

See also

CUpti_ActivityCudaEvent3

context_id

The ID of the context where the event was recorded.

Type:

int

correlation_id

The correlation ID of the API to which this result is associated.

Type:

int

cuda_event_sync_id

A unique ID to associate event synchronization records with the latest CUDA Event record. Similar field is added in CUpti_ActivitySynchronization2 to associate CUDA Event record to the synchronization record. The same CUDA event can be used multiple times, so the event id will not be unique to correlate the synchronization record with the latest CUDA Event record. This field will be unique and can be used to do the required correlation.

Type:

int

device_id

The ID of the device where the event was recorded.

Type:

int

device_timestamp

The device-side timestamp on CUDA event record. Timestamp is in nanoseconds. Collection of this field is disabled by default. It can be enabled by calling CUPTI API cuptiActivityEnableCudaEventDeviceTimestamps

Type:

int

event_id

A unique event ID to identify the event record.

Type:

int

static from_buffer(buffer)

Create an ActivityCudaEvent3 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityCudaEvent3 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_cuda_event3_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityCudaEvent3 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

The unique ID of the graph that recorded this event through graph launch. This field will be 0 if the event is not recorded by a graph event record node.

Type:

int

graph_node_id

The unique ID of the graph node that recorded this event through graph launch. This field will be 0 if the event is not recorded by a graph event record node.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_CUDA_EVENT.

Type:

int

pad

Undefined. Reserved for internal use.

Type:

int

pad2

Undefined. Reserved for internal use.

Type:

int

ptr

Get the pointer address to the data as Python int.

source_graph_id

The unique ID of the source graph from which the executable graph node that recorded this event was instantiated or last updated. This field will be 0 if the event is not recorded by a graph event record node.

Type:

int

source_graph_node_id

The unique ID of the source graph node from which the executable graph node that recorded this event was instantiated or last updated. This field will be 0 if the event is not recorded by a graph event record node.

Type:

int

stream_id

The compute stream where the event was recorded.

Type:

int

Deprecated since version 13.5.0.

class cupti.cupti.ActivityCudaEvent2

Bases: object

Empty-initialize an instance of CUpti_ActivityCudaEvent2.

See also

CUpti_ActivityCudaEvent2

context_id

The ID of the context where the event was recorded.

Type:

int

correlation_id

The correlation ID of the API to which this result is associated.

Type:

int

cuda_event_sync_id

A unique ID to associate event synchronization records with the latest CUDA Event record.

Type:

int

device_id

The ID of the device where the event was recorded.

Type:

int

device_timestamp

The device-side timestamp on CUDA event record. Timestamp is in nanoseconds. Collection of this field is disabled by default. It can be enabled by calling CUPTI API cuptiActivityEnableCudaEventDeviceTimestamps

Type:

int

event_id

A unique event ID to identify the event record.

Type:

int

static from_buffer(buffer)

Create an ActivityCudaEvent2 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityCudaEvent2 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_cuda_event2_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityCudaEvent2 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_CUDA_EVENT.

Type:

int

pad

Undefined. Reserved for internal use.

Type:

int

pad2

Undefined. Reserved for internal use.

Type:

int

ptr

Get the pointer address to the data as Python int.

stream_id

The compute stream where the event was recorded.

Type:

int

Deprecated since version 12.8.0.

class cupti.cupti.ActivityCudaEvent

Bases: object

Empty-initialize an instance of CUpti_ActivityCudaEvent.

See also

CUpti_ActivityCudaEvent

context_id

int:

correlation_id

int:

event_id

int:

static from_buffer(buffer)

Create an ActivityCudaEvent instance with the memory from the given buffer.

static from_data(data)

Create an ActivityCudaEvent instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_cuda_event_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityCudaEvent instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

int:

pad

int:

ptr

Get the pointer address to the data as Python int.

stream_id

int:

class cupti.cupti.ActivityDevice6

Bases: object

Empty-initialize an instance of CUpti_ActivityDevice6.

See also

CUpti_ActivityDevice6

compute_capability_major

Compute capability for the device, major number.

Type:

int

compute_capability_minor

Compute capability for the device, minor number.

Type:

int

compute_instance_id

Compute Instance id for MIG enabled devices. If mig mode is disabled value is set to UINT32_MAX

Type:

int

constant_memory_size

The amount of constant memory on the device, in bytes.

Type:

int

core_clock_rate

The core clock rate of the device, in kHz.

Type:

int

ecc_enabled

ECC enabled flag for device

Type:

int

flags_

The flags associated with the device. CUpti_ActivityFlag

Type:

int

static from_buffer(buffer)

Create an ActivityDevice6 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityDevice6 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_device6_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityDevice6 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

global_memory_bandwidth

The global memory bandwidth available on the device, in kBytes/sec.

Type:

int

global_memory_size

The amount of global memory on the device, in bytes.

Type:

int

gpu_instance_id

GPU Instance id for MIG enabled devices. If mig mode is disabled value is set to UINT32_MAX

Type:

int

id

The device ID.

Type:

int

is_cuda_visible

Flag to indicate whether the device is visible to CUDA. Users can set the device visibility using CUDA_VISIBLE_DEVICES environment

Type:

int

is_mig_enabled

MIG enabled flag for device

Type:

int

is_numa_node

Numa (Non-uniform memory access) information for device GPU is a NUMA node or not

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_DEVICE.

Type:

int

l2cache_size

The size of the L2 cache on the device, in bytes.

Type:

int

max_block_dim_x

Maximum allowed X dimension for a block.

Type:

int

max_block_dim_y

Maximum allowed Y dimension for a block.

Type:

int

max_block_dim_z

Maximum allowed Z dimension for a block.

Type:

int

max_blocks_per_multiprocessor

Maximum number of blocks that can be present on a multiprocessor at any given time.

Type:

int

max_grid_dim_x

Maximum allowed X dimension for a grid.

Type:

int

max_grid_dim_y

Maximum allowed Y dimension for a grid.

Type:

int

max_grid_dim_z

Maximum allowed Z dimension for a grid.

Type:

int

max_ipc

The maximum “instructions per cycle” possible on each device multiprocessor.

Type:

int

max_registers_per_block

Maximum number of registers that can be allocated to a block.

Type:

int

max_registers_per_multiprocessor

Maximum number of 32-bit registers available per multiprocessor.

Type:

int

max_shared_memory_per_block

Maximum amount of shared memory that can be assigned to a block, in bytes.

Type:

int

max_shared_memory_per_multiprocessor

Maximum amount of shared memory available per multiprocessor, in bytes.

Type:

int

max_threads_per_block

Maximum number of threads allowed in a block.

Type:

int

max_warps_per_multiprocessor

Maximum number of warps that can be present on a multiprocessor at any given time.

Type:

int

mig_uuid

The MIG UUID. This value is the globally unique immutable alphanumeric identifier of the device.

Type:

cuda.bindings.driver.CUuuid

name

The device name. Client is responsible for freeing this memory using the free function when done.

Type:

str

num_memcpy_engines

Number of memory copy engines on the device.

Type:

int

num_multiprocessors

Number of multiprocessors on the device.

Type:

int

num_threads_per_warp

The number of threads per warp on the device.

Type:

int

num_tpcs

Number of TPCs on the device.

Type:

int

numa_id

Numa (Non-uniform memory access) information for device NUMA node ID of the GPU memory if GPU is not a NUMA node, it returns invalidNumaId

Type:

int

ptr

Get the pointer address to the data as Python int.

uuid

The device UUID. This value is the globally unique immutable alphanumeric identifier of the device.

Type:

cuda.bindings.driver.CUuuid

Deprecated since version 13.2.0.

class cupti.cupti.ActivityDevice5

Bases: object

Empty-initialize an instance of CUpti_ActivityDevice5.

See also

CUpti_ActivityDevice5

compute_capability_major

Compute capability for the device, major number.

Type:

int

compute_capability_minor

Compute capability for the device, minor number.

Type:

int

compute_instance_id

Compute Instance id for MIG enabled devices. If mig mode is disabled value is set to UINT32_MAX

Type:

int

constant_memory_size

The amount of constant memory on the device, in bytes.

Type:

int

core_clock_rate

The core clock rate of the device, in kHz.

Type:

int

ecc_enabled

ECC enabled flag for device

Type:

int

flags_

The flags associated with the device. CUpti_ActivityFlag

Type:

int

static from_buffer(buffer)

Create an ActivityDevice5 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityDevice5 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_device5_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityDevice5 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

global_memory_bandwidth

The global memory bandwidth available on the device, in kBytes/sec.

Type:

int

global_memory_size

The amount of global memory on the device, in bytes.

Type:

int

gpu_instance_id

GPU Instance id for MIG enabled devices. If mig mode is disabled value is set to UINT32_MAX

Type:

int

id

The device ID.

Type:

int

is_cuda_visible

Flag to indicate whether the device is visible to CUDA. Users can set the device visibility using CUDA_VISIBLE_DEVICES environment

Type:

int

is_mig_enabled

MIG enabled flag for device

Type:

int

is_numa_node

Numa (Non-uniform memory access) information for device GPU is a NUMA node or not

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_DEVICE.

Type:

int

l2cache_size

The size of the L2 cache on the device, in bytes.

Type:

int

max_block_dim_x

Maximum allowed X dimension for a block.

Type:

int

max_block_dim_y

Maximum allowed Y dimension for a block.

Type:

int

max_block_dim_z

Maximum allowed Z dimension for a block.

Type:

int

max_blocks_per_multiprocessor

Maximum number of blocks that can be present on a multiprocessor at any given time.

Type:

int

max_grid_dim_x

Maximum allowed X dimension for a grid.

Type:

int

max_grid_dim_y

Maximum allowed Y dimension for a grid.

Type:

int

max_grid_dim_z

Maximum allowed Z dimension for a grid.

Type:

int

max_ipc

The maximum “instructions per cycle” possible on each device multiprocessor.

Type:

int

max_registers_per_block

Maximum number of registers that can be allocated to a block.

Type:

int

max_registers_per_multiprocessor

Maximum number of 32-bit registers available per multiprocessor.

Type:

int

max_shared_memory_per_block

Maximum amount of shared memory that can be assigned to a block, in bytes.

Type:

int

max_shared_memory_per_multiprocessor

Maximum amount of shared memory available per multiprocessor, in bytes.

Type:

int

max_threads_per_block

Maximum number of threads allowed in a block.

Type:

int

max_warps_per_multiprocessor

Maximum number of warps that can be present on a multiprocessor at any given time.

Type:

int

mig_uuid

The MIG UUID. This value is the globally unique immutable alphanumeric identifier of the device.

Type:

cuda.bindings.driver.CUuuid

name

The device name. Client is responsible for freeing this memory using the free function when done.

Type:

str

num_memcpy_engines

Number of memory copy engines on the device.

Type:

int

num_multiprocessors

Number of multiprocessors on the device.

Type:

int

num_threads_per_warp

The number of threads per warp on the device.

Type:

int

numa_id

Numa (Non-uniform memory access) information for device NUMA node ID of the GPU memory if GPU is not a NUMA node, it returns invalidNumaId

Type:

int

ptr

Get the pointer address to the data as Python int.

uuid

The device UUID. This value is the globally unique immutable alphanumeric identifier of the device.

Type:

cuda.bindings.driver.CUuuid

class cupti.cupti.ActivityDeviceAttribute

Bases: object

Empty-initialize an instance of CUpti_ActivityDeviceAttribute.

See also

CUpti_ActivityDeviceAttribute

attribute

The attribute, either a CUpti_DeviceAttribute or CUdevice_attribute. Flag CUPTI_ACTIVITY_FLAG_DEVICE_ATTRIBUTE_CUDEVICE is used to indicate what kind of attribute this is. If CUPTI_ACTIVITY_FLAG_DEVICE_ATTRIBUTE_CUDEVICE is 1 then CUdevice_attribute field is value, otherwise CUpti_DeviceAttribute field is valid.

Type:

_py_anon_pod11

device_id

The ID of the device that this attribute applies to.

Type:

int

flags_

The flags associated with the device. CUpti_ActivityFlag

Type:

int

static from_buffer(buffer)

Create an ActivityDeviceAttribute instance with the memory from the given buffer.

static from_data(data)

Create an ActivityDeviceAttribute instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_device_attribute_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityDeviceAttribute instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_DEVICE_ATTRIBUTE.

Type:

int

ptr

Get the pointer address to the data as Python int.

value

The value for the attribute. See CUpti_DeviceAttribute and CUdevice_attribute for the type of the value for a given attribute.

Type:

_py_anon_pod12

class cupti.cupti.ActivityDeviceGraphTrace

Bases: object

Empty-initialize an instance of CUpti_ActivityDeviceGraphTrace.

See also

CUpti_ActivityDeviceGraphTrace

context_id

The ID of the context where the first node of the graph is executed.

Type:

int

device_id

The ID of the device where the first node of the graph is executed.

Type:

int

device_launch_mode

The type of launch. See CUpti_DeviceGraphLaunchMode This value is not valid if Hardware Event System (HES) is turned ON.

Type:

int

end

The end timestamp for the graph execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the graph.

Type:

int

static from_buffer(buffer)

Create an ActivityDeviceGraphTrace instance with the memory from the given buffer.

static from_data(data)

Create an ActivityDeviceGraphTrace instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_device_graph_trace_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityDeviceGraphTrace instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

The unique ID of the graph that is launched.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_DEVICE_GRAPH_TRACE

Type:

int

launcher_graph_id

The unique ID of the graph that has launched this graph. This value is not valid if Hardware Event System (HES) is turned ON.

Type:

int

ptr

Get the pointer address to the data as Python int.

start

The start timestamp for the graph execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the graph.

Type:

int

stream_id

The ID of the stream where the graph is being launched. This value is not valid if Hardware Event System (HES) is turned ON.

Type:

int

class cupti.cupti.ActivityEnvironment

Bases: object

Empty-initialize an instance of CUpti_ActivityEnvironment.

See also

CUpti_ActivityEnvironment

data_

_py_anon_pod13:

device_id

The ID of the device

Type:

int

environment_kind

The kind of data reported in this record.

Type:

int

static from_buffer(buffer)

Create an ActivityEnvironment instance with the memory from the given buffer.

static from_data(data)

Create an ActivityEnvironment instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_environment_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityEnvironment instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_ENVIRONMENT.

Type:

int

ptr

Get the pointer address to the data as Python int.

timestamp

The timestamp when this sample was retrieved, in ns. A value of 0 indicates that timestamp information could not be collected for the marker.

Type:

int

class cupti.cupti.ActivityEnvironmentCooling

Bases: object

Empty-initialize an instance of CUpti_ActivityEnvironmentCooling.

See also

CUpti_ActivityEnvironmentCooling

fan_speed

The fan speed as percentage of maximum.

Type:

int

static from_buffer(buffer)

Create an ActivityEnvironmentCooling instance with the memory from the given buffer.

static from_data(data)

Create an ActivityEnvironmentCooling instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_environment_cooling_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityEnvironmentCooling instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.ActivityEnvironmentPower

Bases: object

Empty-initialize an instance of CUpti_ActivityEnvironmentPower.

See also

CUpti_ActivityEnvironmentPower

static from_buffer(buffer)

Create an ActivityEnvironmentPower instance with the memory from the given buffer.

static from_data(data)

Create an ActivityEnvironmentPower instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_environment_power_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityEnvironmentPower instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

power

int:

power_limit

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.ActivityEnvironmentSpeed

Bases: object

Empty-initialize an instance of CUpti_ActivityEnvironmentSpeed.

See also

CUpti_ActivityEnvironmentSpeed

clocks_throttle_reasons

The clocks throttle reasons.

Type:

int

static from_buffer(buffer)

Create an ActivityEnvironmentSpeed instance with the memory from the given buffer.

static from_data(data)

Create an ActivityEnvironmentSpeed instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_environment_speed_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityEnvironmentSpeed instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

memory_clock

The memory frequency in MHz

Type:

int

pcie_link_gen

The PCIe link generation.

Type:

int

pcie_link_width

The PCIe link width.

Type:

int

ptr

Get the pointer address to the data as Python int.

sm_clock

The SM frequency in MHz

Type:

int

class cupti.cupti.ActivityEnvironmentTemperature

Bases: object

Empty-initialize an instance of CUpti_ActivityEnvironmentTemperature.

See also

CUpti_ActivityEnvironmentTemperature

static from_buffer(buffer)

Create an ActivityEnvironmentTemperature instance with the memory from the given buffer.

static from_data(data)

Create an ActivityEnvironmentTemperature instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_environment_temperature_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityEnvironmentTemperature instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

gpu_temperature

The GPU temperature in degrees C.

Type:

int

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.ActivityExternalCorrelation

Bases: object

Empty-initialize an instance of CUpti_ActivityExternalCorrelation.

See also

CUpti_ActivityExternalCorrelation

correlation_id

The correlation ID of the associated CUDA driver or runtime API record.

Type:

int

external_id

The correlation ID of the associated non-CUDA API record. The exact field in the associated external record depends on that record’s activity kind (externalKind).

Type:

int

external_kind

The kind of external API this record correlated to.

Type:

int

static from_buffer(buffer)

Create an ActivityExternalCorrelation instance with the memory from the given buffer.

static from_data(data)

Create an ActivityExternalCorrelation instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_external_correlation_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityExternalCorrelation instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The kind of this activity.

Type:

int

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.ActivityFieldSelection

Bases: object

Empty-initialize an instance of CUpti_ActivityFieldSelection.

See also

CUpti_ActivityFieldSelection

static from_buffer(buffer)

Create an ActivityFieldSelection instance with the memory from the given buffer.

static from_data(data)

Create an ActivityFieldSelection instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_field_selection_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityFieldSelection instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_field_ids

int:

ptr

Get the pointer address to the data as Python int.

struct_size

int:

class cupti.cupti.ActivityFunction

Bases: object

Empty-initialize an instance of CUpti_ActivityFunction.

See also

CUpti_ActivityFunction

context_id

The ID of the context where the function is launched.

Type:

int

static from_buffer(buffer)

Create an ActivityFunction instance with the memory from the given buffer.

static from_data(data)

Create an ActivityFunction instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_function_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityFunction instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

function_index

The function’s unique symbol index in the module.

Type:

int

id

ID to uniquely identify the record

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_FUNCTION.

Type:

int

module_id

The module ID in which this global/device function is present.

Type:

int

name

The name of the function. This name is shared across all activity records representing the same kernel, and so should not be modified.

Type:

str

pad

Undefined. Reserved for internal use.

Type:

int

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.ActivityGraphHostNode

Bases: object

Empty-initialize an instance of CUpti_ActivityGraphHostNode.

See also

CUpti_ActivityGraphHostNode

context_id

The ID of the CUDA context to which the waiting CUDA stream belongs.

Type:

int

correlation_id

The correlation ID of the graph host node operation. Each operation is assigned a unique correlation ID that is identical to the correlation ID in the driver API activity record that launched the operation.

Type:

int

device_id

The ID of the CUDA device to which the CUDA context and stream belong. The graph host node is executing on the CPU, but it is associated with the CUDA device through the CUDA context and stream.

Type:

int

end

The end timestamp. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the start time is unknown.

Type:

int

static from_buffer(buffer)

Create an ActivityGraphHostNode instance with the memory from the given buffer.

static from_data(data)

Create an ActivityGraphHostNode instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_graph_host_node_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityGraphHostNode instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

The unique ID of the graph that executed this host node through graph launch.

Type:

int

graph_node_id

The unique ID of the graph node that executed this host node through graph launch.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_GRAPH_HOST_NODE

Type:

int

process_id

The ID of the process where the host node is executing.

Type:

int

ptr

Get the pointer address to the data as Python int.

start

The start timestamp. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the start time is unknown.

Type:

int

stream_id

The ID of the stream that waits for the graph host node to finish.

Type:

int

thread_id

The ID of the thread where the host node is executing.

Type:

int

class cupti.cupti.ActivityGraphTrace2

Bases: object

Empty-initialize an instance of CUpti_ActivityGraphTrace2.

See also

CUpti_ActivityGraphTrace2

context_id

The ID of the context where the first node of the graph is executed. If this is INT_MAX, then the start is on the host.

Type:

int

correlation_id

The correlation ID of the graph launch. Each graph launch is assigned a unique correlation ID that is identical to the correlation ID in the driver API activity record that launched the graph.

Type:

int

device_id

The ID of the device where the first node of the graph is executed. If this is INT_MAX, then the start is on the host.

Type:

int

end

The end timestamp for the graph execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the graph.

Type:

int

end_context_id

The ID of the context where the last node of the graph is executed.

Type:

int

end_device_id

The ID of the device where last node of the graph is executed

Type:

int

static from_buffer(buffer)

Create an ActivityGraphTrace2 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityGraphTrace2 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_graph_trace2_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityGraphTrace2 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

The unique ID of the graph that is launched.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_GRAPH_TRACE

Type:

int

ptr

Get the pointer address to the data as Python int.

start

The start timestamp for the graph execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the graph.

Type:

int

stream_id

The ID of the stream where the graph is being launched.

Type:

int

class cupti.cupti.ActivityGreenContext3

Bases: object

Empty-initialize an instance of CUpti_ActivityGreenContext3.

See also

CUpti_ActivityGreenContext3

context_id

The context ID of the green context.

Type:

int

device_id

The device ID associated with the green context.

Type:

int

static from_buffer(buffer)

Create an ActivityGreenContext3 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityGreenContext3 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_green_context3_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityGreenContext3 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_GREEN_CONTEXT.

Type:

int

logical_tpc_mask

Interpreting the TPC Mask** For a green context with numTpcs=3, logicalTpcMaskSize=8, and logicalTpcMask={66,8,0,0,0,0,0,0}: - Word 0: Value 66 (binary: …01000010, showing lowest 8 bits of 32-bit word) indicates TPCs 1 and 6 - Word 1: Value 8 (binary: …00001000) indicates TPC 35 (bit 3 of word 1, i.e., global TPC ID 32+3) - Each word is 32 bits covering TPCs 0-31 (word 0), 32-63 (word 1), etc. - To check if TPC N is allocated: (logicalTpcMask[N/32] & (1U << (N%32))) - logicalTpcMaskSize indicates how many words in the array contain valid data

Type:

uint32

Type:

(array of length 32).Bitset of allocated TPC IDs represented as 32-bit words. Valid words are specified by logicalTpcMaskSize; unused entries are zeroed. The array supports up to 1024 TPCs (32 words × 32 bits). Each bit represents a logical TPC ID assigned to the green context. For example, if bit k is set, logical TPC ID k is assigned to the context. **Example

logical_tpc_mask_size

The size (in 32-bit words) of the logical TPC mask stored in logicalTpcMask[].

Type:

int

num_multiprocessors

The number of multiprocessors (SMs) allocated to the green context.

Type:

int

num_tpcs

The number of TPCs allocated to the green context.

Type:

int

parent_context_id

The ID of the parent context.

Type:

int

ptr

Get the pointer address to the data as Python int.

workqueue_channel_count

The number of HW channels assigned to this green context’s work queue. This reflects the actual channel assignment at green context creation time. A value of 0 indicates that channel information is not available.

Type:

int

workqueue_channel_ids

(array of length 32).Array of HW channel IDs assigned to this green context’s work queue. Valid entries are specified by workqueueChannelCount; unused entries are zeroed. The array supports up to 32 channel IDs.

Type:

uint32

workqueue_concurrency_limit

The concurrency limit for the work queue associated with the green context. This defines the maximum number of concurrent operations allowed on this work queue.

Type:

int

workqueue_resource_id

The work queue resource ID associated with the green context. This is a unique identifier for the work queue resource used by this green context.

Type:

int

workqueue_sharing_scope

The sharing scope for the work queue associated with the green context. This defines how the work queue can be shared across different contexts or processes. Refer CUdevWorkqueueConfigScope in cuda.h for possible values.

Type:

int

Deprecated since version 13.4.0.

class cupti.cupti.ActivityGreenContext2

Bases: object

Empty-initialize an instance of CUpti_ActivityGreenContext2.

See also

CUpti_ActivityGreenContext2

context_id

The context ID of the green context.

Type:

int

device_id

The device ID associated with the green context.

Type:

int

static from_buffer(buffer)

Create an ActivityGreenContext2 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityGreenContext2 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_green_context2_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityGreenContext2 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_GREEN_CONTEXT.

Type:

int

logical_tpc_mask

Interpreting the TPC Mask** For a green context with numTpcs=3, logicalTpcMaskSize=8, and logicalTpcMask={66,8,0,0,0,0,0,0}: - Word 0: Value 66 (binary: …01000010, showing lowest 8 bits of 32-bit word) indicates TPCs 1 and 6 - Word 1: Value 8 (binary: …00001000) indicates TPC 35 (bit 3 of word 1, i.e., global TPC ID 32+3) - Each word is 32 bits covering TPCs 0-31 (word 0), 32-63 (word 1), etc. - To check if TPC N is allocated: (logicalTpcMask[N/32] & (1U << (N%32))) - logicalTpcMaskSize indicates how many words in the array contain valid data

Type:

uint32

Type:

(array of length 32).Bitset of allocated TPC IDs represented as 32-bit words. Valid words are specified by logicalTpcMaskSize; unused entries are zeroed. The array supports up to 1024 TPCs (32 words × 32 bits). Each bit represents a logical TPC ID assigned to the green context. For example, if bit k is set, logical TPC ID k is assigned to the context. **Example

logical_tpc_mask_size

The size (in 32-bit words) of the logical TPC mask stored in logicalTpcMask[].

Type:

int

num_multiprocessors

The number of multiprocessors (SMs) allocated to the green context.

Type:

int

num_tpcs

The number of TPCs allocated to the green context.

Type:

int

parent_context_id

The ID of the parent context.

Type:

int

ptr

Get the pointer address to the data as Python int.

workqueue_concurrency_limit

The concurrency limit for the work queue associated with the green context. This defines the maximum number of concurrent operations allowed on this work queue.

Type:

int

workqueue_resource_id

The work queue resource ID associated with the green context. This is a unique identifier for the work queue resource used by this green context.

Type:

int

workqueue_sharing_scope

The sharing scope for the work queue associated with the green context. This defines how the work queue can be shared across different contexts or processes. Refer CUdevWorkqueueConfigScope in cuda.h for possible values.

Type:

int

Deprecated since version 13.3.0.

class cupti.cupti.ActivityGreenContext

Bases: object

Empty-initialize an instance of CUpti_ActivityGreenContext.

See also

CUpti_ActivityGreenContext

context_id

The context ID of the green context.

Type:

int

device_id

The device ID associated with the green context.

Type:

int

static from_buffer(buffer)

Create an ActivityGreenContext instance with the memory from the given buffer.

static from_data(data)

Create an ActivityGreenContext instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_green_context_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityGreenContext instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_GREEN_CONTEXT.

Type:

int

logical_tpc_mask

Interpreting the TPC Mask** For a green context with numTpcs=3, logicalTpcMaskSize=8, and logicalTpcMask={66,8,0,0,0,0,0,0}: - Word 0: Value 66 (binary: …01000010, showing lowest 8 bits of 32-bit word) indicates TPCs 1 and 6 - Word 1: Value 8 (binary: …00001000) indicates TPC 35 (bit 3 of word 1, i.e., global TPC ID 32+3) - Each word is 32 bits covering TPCs 0-31 (word 0), 32-63 (word 1), etc. - To check if TPC N is allocated: (logicalTpcMask[N/32] & (1U << (N%32))) - logicalTpcMaskSize indicates how many words in the array contain valid data

Type:

uint32

Type:

(array of length 32).Bitset of allocated TPC IDs represented as 32-bit words. Valid words are specified by logicalTpcMaskSize; unused entries are zeroed. The array supports up to 1024 TPCs (32 words × 32 bits). Each bit represents a logical TPC ID assigned to the green context. For example, if bit k is set, logical TPC ID k is assigned to the context. **Example

logical_tpc_mask_size

The size (in 32-bit words) of the logical TPC mask stored in logicalTpcMask[].

Type:

int

num_multiprocessors

The number of multiprocessors (SMs) allocated to the green context.

Type:

int

num_tpcs

The number of TPCs allocated to the green context.

Type:

int

parent_context_id

The ID of the parent context.

Type:

int

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.ActivityHostLaunch

Bases: object

Empty-initialize an instance of CUpti_ActivityHostLaunch.

See also

CUpti_ActivityHostLaunch

context_id

The ID of the CUDA context to which the waiting CUDA stream belongs.

Type:

int

correlation_id

The correlation ID of the host launch operation. Each operation is assigned a unique correlation ID that is identical to the correlation ID in the driver API activity record that launched the operation.

Type:

int

device_id

The ID of the CUDA device to which the CUDA context and stream belong. The host function is executing on the CPU, but it is associated with the CUDA device through the CUDA context and stream.

Type:

int

end

The end timestamp. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the start time is unknown.

Type:

int

static from_buffer(buffer)

Create an ActivityHostLaunch instance with the memory from the given buffer.

static from_data(data)

Create an ActivityHostLaunch instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_host_launch_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityHostLaunch instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_HOST_LAUNCH

Type:

int

process_id

The ID of the process where the host function is executing.

Type:

int

ptr

Get the pointer address to the data as Python int.

start

The start timestamp. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the start time is unknown.

Type:

int

stream_id

The ID of the CUDA context to which the waiting CUDA stream belongs.

Type:

int

thread_id

The ID of the thread where the host function is executing.

Type:

int

class cupti.cupti.ActivityJit2

Bases: object

Empty-initialize an instance of CUpti_ActivityJit2.

See also

CUpti_ActivityJit2

cache_path

The path where the fat binary is cached.

Type:

str

cache_size

The size of compute cache.

Type:

int

correlation_id

The correlation ID of the JIT operation to which records belong to. Each JIT operation is assigned a unique correlation ID that is identical to the correlation ID in the driver or runtime API activity record that launched the JIT operation.

Type:

int

device_id

The device ID.

Type:

int

end

The end timestamp for the JIT operation, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the JIT operation.

Type:

int

static from_buffer(buffer)

Create an ActivityJit2 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityJit2 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_jit2_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityJit2 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

jit_entry_type

The JIT entry type.

Type:

int

jit_operation_correlation_id

The correlation ID to correlate JIT compilation, load and store operations. Each JIT compilation unit is assigned a unique correlation ID at the time of the JIT compilation. This correlation id can be used to find the matching JIT cache load/store records.

Type:

int

jit_operation_type

The JIT operation type.

Type:

int

kind

The activity record kind must be CUPTI_ACTIVITY_KIND_JIT.

Type:

int

process_id

The ID of the process where the JIT operation is executing.

Type:

int

ptr

Get the pointer address to the data as Python int.

start

The start timestamp for the JIT operation, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the JIT operation.

Type:

int

thread_id

The ID of the thread where the JIT operation is executing.

Type:

int

class cupti.cupti.ActivityKernel13

Bases: object

Empty-initialize an instance of CUpti_ActivityKernel13.

See also

CUpti_ActivityKernel13

block_x

The X-dimension block size for the kernel.

Type:

int

block_y

The Y-dimension block size for the kernel.

Type:

int

block_z

The Z-dimension grid size for the kernel.

Type:

int

cache_config

For devices with compute capability 7.5+ cacheConfig values are not updated in case field isSharedMemoryCarveoutRequested is set

Type:

_py_anon_pod58

channel_id

The ID of the HW channel on which the kernel is launched. This field would be CUPTI_INVALID_CHANNEL_ID if the isDeviceLaunched is set to 1.

Type:

int

channel_type

The type of the channel

Type:

int

cluster_scheduling_policy

The cluster scheduling policy for the kernel. Refer CUclusterSchedulingPolicy Field is valid for devices with compute capability 9.0 and higher

Type:

int

cluster_x

The X-dimension cluster size for the kernel. Field is valid for devices with compute capability 9.0 and higher

Type:

int

cluster_y

The Y-dimension cluster size for the kernel. Field is valid for devices with compute capability 9.0 and higher

Type:

int

cluster_z

The Z-dimension cluster size for the kernel. Field is valid for devices with compute capability 9.0 and higher

Type:

int

completed

The completed timestamp for the kernel execution, in ns. It represents the completion of all it’s child kernels and the kernel itself. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the completion time is unknown.

Type:

int

context_id

The ID of the context where the kernel is executing.

Type:

int

correlation_id

The correlation ID of the kernel. Each kernel execution is assigned a unique correlation ID that is identical to the correlation ID in the driver or runtime API activity record that launched the kernel.

Type:

int

device_id

The ID of the device where the kernel is executing.

Type:

int

dynamic_shared_memory

The dynamic shared memory reserved for the kernel, in bytes.

Type:

int

end

The end timestamp for the kernel execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the kernel.

Type:

int

execution_model

The execution model of the kernel function. CUpti_FuncExecutionModel

Type:

int

static from_buffer(buffer)

Create an ActivityKernel13 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityKernel13 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_kernel13_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityKernel13 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

The unique ID of the graph that launched this kernel through graph launch APIs. This field will be 0 if the kernel is not launched through graph launch APIs.

Type:

int

graph_node_id

The unique ID of the graph node that launched this kernel through graph launch APIs. This field will be 0 if the kernel is not launched through graph launch APIs.

Type:

int

grid_id

The grid ID of the kernel. Each kernel is assigned a unique grid ID at runtime.

Type:

int

grid_x

The X-dimension grid size for the kernel.

Type:

int

grid_y

The Y-dimension grid size for the kernel.

Type:

int

grid_z

The Z-dimension grid size for the kernel.

Type:

int

is_device_launched

This field is set to 1 if the kernel is part of a device launched graph.

Type:

int

is_shared_memory_carveout_requested

This indicates if CU_FUNC_ATTRIBUTE_PREFERRED_SHARED_MEMORY_CARVEOUT was updated for the kernel launch

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_KERNEL or CUPTI_ACTIVITY_KIND_CONCURRENT_KERNEL.

Type:

int

launch_type

The indicates if the kernel was executed via a regular launch or via a single/multi device cooperative launch. CUpti_ActivityLaunchType

Type:

int

local_memory_per_thread

The amount of local memory reserved for each thread, in bytes.

Type:

int

local_memory_total

The total amount of local memory reserved for the kernel, in bytes (deprecated in CUDA 11.8). Refer field localMemoryTotal_v2

Type:

int

local_memory_total_v2

The total amount of local memory reserved for the kernel, in bytes.

Type:

int

max_active_clusters

The maximum clusters that could co-exist on the target device for the kernel

Type:

int

max_potential_cluster_size

The maximum cluster size for the kernel

Type:

int

name

The name of the kernel. This name is shared across all activity records representing the same kernel, and so should not be modified.

Type:

str

p_access_policy_window

The pointer to the access policy window. The structure CUaccessPolicyWindow is defined in cuda.h.

Type:

int

partitioned_global_cache_executed

The partitioned global caching executed for the kernel. Partitioned global caching is required to enable caching on certain chips, such as devices with compute capability 5.2. Partitioned global caching can be automatically disabled if the occupancy requirement of the launch cannot support caching.

Type:

int

partitioned_global_cache_requested

The partitioned global caching requested for the kernel. Partitioned global caching is required to enable caching on certain chips, such as devices with compute capability 5.2.

Type:

int

priority

The launch priority of the kernel.

Type:

int

ptr

Get the pointer address to the data as Python int.

queued

The timestamp when the kernel is queued up in the command buffer, in ns. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the queued time could not be collected for the kernel. This timestamp is not collected by default. Use API cuptiActivityEnableLatencyTimestamps() to enable collection. Command buffer is a buffer written by CUDA driver to send commands like kernel launch, memory copy etc to the GPU. All launches of CUDA kernels are asynchronous with respect to the host, the host requests the launch by writing commands into the command buffer, then returns without checking the GPU’s progress.

Type:

int

registers_per_thread

The number of registers required for each thread executing the kernel.

Type:

int

shared_memory_carveout_requested

Shared memory carveout value requested for the function in percentage of the total resource. The value will be updated only if field isSharedMemoryCarveoutRequested is set.

Type:

int

shared_memory_config

The shared memory configuration used for the kernel. The value is one of the CUsharedconfig enumeration values from cuda.h.

Type:

int

shared_memory_executed

Shared memory size set by the driver.

Type:

int

shmem_limit_config

The shared memory limit config for the kernel. This field shows whether user has opted for a higher per block limit of dynamic shared memory.

Type:

int

source_graph_id

The unique graph ID of the node from which the kernel node was instantiated or last updated. This field will be 0 if the kernel is not launched through graph launch APIs.

Type:

int

source_graph_node_id

The unique graph node ID of the node from which the kernel node was instantiated or last updated. This field will be 0 if the kernel is not launched through graph launch APIs.

Type:

int

start

The start timestamp for the kernel execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the kernel.

Type:

int

static_shared_memory

The static shared memory allocated for the kernel, in bytes.

Type:

int

stream_id

The ID of the stream where the kernel is executing.

Type:

int

submitted

The timestamp when the command buffer containing the kernel launch is submitted to the GPU, in ns. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the submitted time could not be collected for the kernel. This timestamp is not collected by default. Use API cuptiActivityEnableLatencyTimestamps() to enable collection.

Type:

int

Deprecated since version 13.4.0.

class cupti.cupti.ActivityKernel12

Bases: object

Empty-initialize an instance of CUpti_ActivityKernel12.

See also

CUpti_ActivityKernel12

block_x

The X-dimension block size for the kernel.

Type:

int

block_y

The Y-dimension block size for the kernel.

Type:

int

block_z

The Z-dimension grid size for the kernel.

Type:

int

cache_config

For devices with compute capability 7.5+ cacheConfig values are not updated in case field isSharedMemoryCarveoutRequested is set

Type:

_py_anon_pod56

channel_id

The ID of the HW channel on which the kernel is launched.

Type:

int

channel_type

The type of the channel

Type:

int

cluster_scheduling_policy

The cluster scheduling policy for the kernel. Refer CUclusterSchedulingPolicy Field is valid for devices with compute capability 9.0 and higher

Type:

int

cluster_x

The X-dimension cluster size for the kernel. Field is valid for devices with compute capability 9.0 and higher

Type:

int

cluster_y

The Y-dimension cluster size for the kernel. Field is valid for devices with compute capability 9.0 and higher

Type:

int

cluster_z

The Z-dimension cluster size for the kernel. Field is valid for devices with compute capability 9.0 and higher

Type:

int

completed

The completed timestamp for the kernel execution, in ns. It represents the completion of all it’s child kernels and the kernel itself. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the completion time is unknown.

Type:

int

context_id

The ID of the context where the kernel is executing.

Type:

int

correlation_id

The correlation ID of the kernel. Each kernel execution is assigned a unique correlation ID that is identical to the correlation ID in the driver or runtime API activity record that launched the kernel.

Type:

int

device_id

The ID of the device where the kernel is executing.

Type:

int

dynamic_shared_memory

The dynamic shared memory reserved for the kernel, in bytes.

Type:

int

end

The end timestamp for the kernel execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the kernel.

Type:

int

execution_model

The execution model of the kernel function. CUpti_FuncExecutionModel

Type:

int

static from_buffer(buffer)

Create an ActivityKernel12 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityKernel12 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_kernel12_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityKernel12 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

The unique ID of the graph that launched this kernel through graph launch APIs. This field will be 0 if the kernel is not launched through graph launch APIs.

Type:

int

graph_node_id

The unique ID of the graph node that launched this kernel through graph launch APIs. This field will be 0 if the kernel is not launched through graph launch APIs.

Type:

int

grid_id

The grid ID of the kernel. Each kernel is assigned a unique grid ID at runtime.

Type:

int

grid_x

The X-dimension grid size for the kernel.

Type:

int

grid_y

The Y-dimension grid size for the kernel.

Type:

int

grid_z

The Z-dimension grid size for the kernel.

Type:

int

is_device_launched

This field is set to 1 if the kernel is part of a device launched graph.

Type:

int

is_shared_memory_carveout_requested

This indicates if CU_FUNC_ATTRIBUTE_PREFERRED_SHARED_MEMORY_CARVEOUT was updated for the kernel launch

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_KERNEL or CUPTI_ACTIVITY_KIND_CONCURRENT_KERNEL.

Type:

int

launch_type

The indicates if the kernel was executed via a regular launch or via a single/multi device cooperative launch. CUpti_ActivityLaunchType

Type:

int

local_memory_per_thread

The amount of local memory reserved for each thread, in bytes.

Type:

int

local_memory_total

The total amount of local memory reserved for the kernel, in bytes (deprecated in CUDA 11.8). Refer field localMemoryTotal_v2

Type:

int

local_memory_total_v2

The total amount of local memory reserved for the kernel, in bytes.

Type:

int

max_active_clusters

The maximum clusters that could co-exist on the target device for the kernel

Type:

int

max_potential_cluster_size

The maximum cluster size for the kernel

Type:

int

name

The name of the kernel. This name is shared across all activity records representing the same kernel, and so should not be modified.

Type:

str

p_access_policy_window

The pointer to the access policy window. The structure CUaccessPolicyWindow is defined in cuda.h.

Type:

int

partitioned_global_cache_executed

The partitioned global caching executed for the kernel. Partitioned global caching is required to enable caching on certain chips, such as devices with compute capability 5.2. Partitioned global caching can be automatically disabled if the occupancy requirement of the launch cannot support caching.

Type:

int

partitioned_global_cache_requested

The partitioned global caching requested for the kernel. Partitioned global caching is required to enable caching on certain chips, such as devices with compute capability 5.2.

Type:

int

priority

The launch priority of the kernel.

Type:

int

ptr

Get the pointer address to the data as Python int.

queued

The timestamp when the kernel is queued up in the command buffer, in ns. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the queued time could not be collected for the kernel. This timestamp is not collected by default. Use API cuptiActivityEnableLatencyTimestamps() to enable collection. Command buffer is a buffer written by CUDA driver to send commands like kernel launch, memory copy etc to the GPU. All launches of CUDA kernels are asynchronous with respect to the host, the host requests the launch by writing commands into the command buffer, then returns without checking the GPU’s progress.

Type:

int

registers_per_thread

The number of registers required for each thread executing the kernel.

Type:

int

shared_memory_carveout_requested

Shared memory carveout value requested for the function in percentage of the total resource. The value will be updated only if field isSharedMemoryCarveoutRequested is set.

Type:

int

shared_memory_config

The shared memory configuration used for the kernel. The value is one of the CUsharedconfig enumeration values from cuda.h.

Type:

int

shared_memory_executed

Shared memory size set by the driver.

Type:

int

shmem_limit_config

The shared memory limit config for the kernel. This field shows whether user has opted for a higher per block limit of dynamic shared memory.

Type:

int

start

The start timestamp for the kernel execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the kernel.

Type:

int

static_shared_memory

The static shared memory allocated for the kernel, in bytes.

Type:

int

stream_id

The ID of the stream where the kernel is executing.

Type:

int

submitted

The timestamp when the command buffer containing the kernel launch is submitted to the GPU, in ns. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the submitted time could not be collected for the kernel. This timestamp is not collected by default. Use API cuptiActivityEnableLatencyTimestamps() to enable collection.

Type:

int

Deprecated since version 13.3.0.

class cupti.cupti.ActivityKernel11

Bases: object

Empty-initialize an instance of CUpti_ActivityKernel11.

See also

CUpti_ActivityKernel11

block_x

The X-dimension block size for the kernel.

Type:

int

block_y

The Y-dimension block size for the kernel.

Type:

int

block_z

The Z-dimension grid size for the kernel.

Type:

int

cache_config

For devices with compute capability 7.5+ cacheConfig values are not updated in case field isSharedMemoryCarveoutRequested is set

Type:

_py_anon_pod52

channel_id

The ID of the HW channel on which the kernel is launched.

Type:

int

channel_type

The type of the channel

Type:

int

cluster_scheduling_policy

The cluster scheduling policy for the kernel. Refer CUclusterSchedulingPolicy Field is valid for devices with compute capability 9.0 and higher

Type:

int

cluster_x

The X-dimension cluster size for the kernel. Field is valid for devices with compute capability 9.0 and higher

Type:

int

cluster_y

The Y-dimension cluster size for the kernel. Field is valid for devices with compute capability 9.0 and higher

Type:

int

cluster_z

The Z-dimension cluster size for the kernel. Field is valid for devices with compute capability 9.0 and higher

Type:

int

completed

The completed timestamp for the kernel execution, in ns. It represents the completion of all it’s child kernels and the kernel itself. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the completion time is unknown.

Type:

int

context_id

The ID of the context where the kernel is executing.

Type:

int

correlation_id

The correlation ID of the kernel. Each kernel execution is assigned a unique correlation ID that is identical to the correlation ID in the driver or runtime API activity record that launched the kernel.

Type:

int

device_id

The ID of the device where the kernel is executing.

Type:

int

dynamic_shared_memory

The dynamic shared memory reserved for the kernel, in bytes.

Type:

int

end

The end timestamp for the kernel execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the kernel.

Type:

int

static from_buffer(buffer)

Create an ActivityKernel11 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityKernel11 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_kernel11_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityKernel11 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

The unique ID of the graph that launched this kernel through graph launch APIs. This field will be 0 if the kernel is not launched through graph launch APIs.

Type:

int

graph_node_id

The unique ID of the graph node that launched this kernel through graph launch APIs. This field will be 0 if the kernel is not launched through graph launch APIs.

Type:

int

grid_id

The grid ID of the kernel. Each kernel is assigned a unique grid ID at runtime.

Type:

int

grid_x

The X-dimension grid size for the kernel.

Type:

int

grid_y

The Y-dimension grid size for the kernel.

Type:

int

grid_z

The Z-dimension grid size for the kernel.

Type:

int

is_device_launched

This field is set to 1 if the kernel is part of a device launched graph.

Type:

int

is_shared_memory_carveout_requested

This indicates if CU_FUNC_ATTRIBUTE_PREFERRED_SHARED_MEMORY_CARVEOUT was updated for the kernel launch

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_KERNEL or CUPTI_ACTIVITY_KIND_CONCURRENT_KERNEL.

Type:

int

launch_type

The indicates if the kernel was executed via a regular launch or via a single/multi device cooperative launch. CUpti_ActivityLaunchType

Type:

int

local_memory_per_thread

The amount of local memory reserved for each thread, in bytes.

Type:

int

local_memory_total

The total amount of local memory reserved for the kernel, in bytes (deprecated in CUDA 11.8). Refer field localMemoryTotal_v2

Type:

int

local_memory_total_v2

The total amount of local memory reserved for the kernel, in bytes.

Type:

int

max_active_clusters

The maximum clusters that could co-exist on the target device for the kernel

Type:

int

max_potential_cluster_size

The maximum cluster size for the kernel

Type:

int

name

The name of the kernel. This name is shared across all activity records representing the same kernel, and so should not be modified.

Type:

str

p_access_policy_window

The pointer to the access policy window. The structure CUaccessPolicyWindow is defined in cuda.h.

Type:

int

partitioned_global_cache_executed

The partitioned global caching executed for the kernel. Partitioned global caching is required to enable caching on certain chips, such as devices with compute capability 5.2. Partitioned global caching can be automatically disabled if the occupancy requirement of the launch cannot support caching.

Type:

int

partitioned_global_cache_requested

The partitioned global caching requested for the kernel. Partitioned global caching is required to enable caching on certain chips, such as devices with compute capability 5.2.

Type:

int

priority

The launch priority of the kernel.

Type:

int

ptr

Get the pointer address to the data as Python int.

queued

The timestamp when the kernel is queued up in the command buffer, in ns. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the queued time could not be collected for the kernel. This timestamp is not collected by default. Use API cuptiActivityEnableLatencyTimestamps() to enable collection. Command buffer is a buffer written by CUDA driver to send commands like kernel launch, memory copy etc to the GPU. All launches of CUDA kernels are asynchronous with respect to the host, the host requests the launch by writing commands into the command buffer, then returns without checking the GPU’s progress.

Type:

int

registers_per_thread

The number of registers required for each thread executing the kernel.

Type:

int

shared_memory_carveout_requested

Shared memory carveout value requested for the function in percentage of the total resource. The value will be updated only if field isSharedMemoryCarveoutRequested is set.

Type:

int

shared_memory_config

The shared memory configuration used for the kernel. The value is one of the CUsharedconfig enumeration values from cuda.h.

Type:

int

shared_memory_executed

Shared memory size set by the driver.

Type:

int

shmem_limit_config

The shared memory limit config for the kernel. This field shows whether user has opted for a higher per block limit of dynamic shared memory.

Type:

int

start

The start timestamp for the kernel execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the kernel.

Type:

int

static_shared_memory

The static shared memory allocated for the kernel, in bytes.

Type:

int

stream_id

The ID of the stream where the kernel is executing.

Type:

int

submitted

The timestamp when the command buffer containing the kernel launch is submitted to the GPU, in ns. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the submitted time could not be collected for the kernel. This timestamp is not collected by default. Use API cuptiActivityEnableLatencyTimestamps() to enable collection.

Type:

int

Deprecated since version 13.2.0.

class cupti.cupti.ActivityKernel10

Bases: object

Empty-initialize an instance of CUpti_ActivityKernel10.

See also

CUpti_ActivityKernel10

block_x

The X-dimension block size for the kernel.

Type:

int

block_y

The Y-dimension block size for the kernel.

Type:

int

block_z

The Z-dimension grid size for the kernel.

Type:

int

cache_config

For devices with compute capability 7.5+ cacheConfig values are not updated in case field isSharedMemoryCarveoutRequested is set

Type:

_py_anon_pod50

channel_id

The ID of the HW channel on which the kernel is launched.

Type:

int

channel_type

The type of the channel

Type:

int

cluster_scheduling_policy

The cluster scheduling policy for the kernel. Refer CUclusterSchedulingPolicy Field is valid for devices with compute capability 9.0 and higher

Type:

int

cluster_x

The X-dimension cluster size for the kernel. Field is valid for devices with compute capability 9.0 and higher

Type:

int

cluster_y

The Y-dimension cluster size for the kernel. Field is valid for devices with compute capability 9.0 and higher

Type:

int

cluster_z

The Z-dimension cluster size for the kernel. Field is valid for devices with compute capability 9.0 and higher

Type:

int

completed

The completed timestamp for the kernel execution, in ns. It represents the completion of all it’s child kernels and the kernel itself. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the completion time is unknown.

Type:

int

context_id

The ID of the context where the kernel is executing.

Type:

int

correlation_id

The correlation ID of the kernel. Each kernel execution is assigned a unique correlation ID that is identical to the correlation ID in the driver or runtime API activity record that launched the kernel.

Type:

int

device_id

The ID of the device where the kernel is executing.

Type:

int

dynamic_shared_memory

The dynamic shared memory reserved for the kernel, in bytes.

Type:

int

end

The end timestamp for the kernel execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the kernel.

Type:

int

static from_buffer(buffer)

Create an ActivityKernel10 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityKernel10 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_kernel10_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityKernel10 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

The unique ID of the graph that launched this kernel through graph launch APIs. This field will be 0 if the kernel is not launched through graph launch APIs.

Type:

int

graph_node_id

The unique ID of the graph node that launched this kernel through graph launch APIs. This field will be 0 if the kernel is not launched through graph launch APIs.

Type:

int

grid_id

The grid ID of the kernel. Each kernel is assigned a unique grid ID at runtime.

Type:

int

grid_x

The X-dimension grid size for the kernel.

Type:

int

grid_y

The Y-dimension grid size for the kernel.

Type:

int

grid_z

The Z-dimension grid size for the kernel.

Type:

int

is_device_launched

This field is set to 1 if the kernel is part of a device launched graph.

Type:

int

is_shared_memory_carveout_requested

This indicates if CU_FUNC_ATTRIBUTE_PREFERRED_SHARED_MEMORY_CARVEOUT was updated for the kernel launch

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_KERNEL or CUPTI_ACTIVITY_KIND_CONCURRENT_KERNEL.

Type:

int

launch_type

The indicates if the kernel was executed via a regular launch or via a single/multi device cooperative launch. CUpti_ActivityLaunchType

Type:

int

local_memory_per_thread

The amount of local memory reserved for each thread, in bytes.

Type:

int

local_memory_total

The total amount of local memory reserved for the kernel, in bytes (deprecated in CUDA 11.8). Refer field localMemoryTotal_v2

Type:

int

local_memory_total_v2

The total amount of local memory reserved for the kernel, in bytes.

Type:

int

max_active_clusters

The maximum clusters that could co-exist on the target device for the kernel

Type:

int

max_potential_cluster_size

The maximum cluster size for the kernel

Type:

int

name

The name of the kernel. This name is shared across all activity records representing the same kernel, and so should not be modified.

Type:

str

p_access_policy_window

The pointer to the access policy window. The structure CUaccessPolicyWindow is defined in cuda.h.

Type:

int

partitioned_global_cache_executed

The partitioned global caching executed for the kernel. Partitioned global caching is required to enable caching on certain chips, such as devices with compute capability 5.2. Partitioned global caching can be automatically disabled if the occupancy requirement of the launch cannot support caching.

Type:

int

partitioned_global_cache_requested

The partitioned global caching requested for the kernel. Partitioned global caching is required to enable caching on certain chips, such as devices with compute capability 5.2.

Type:

int

ptr

Get the pointer address to the data as Python int.

queued

The timestamp when the kernel is queued up in the command buffer, in ns. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the queued time could not be collected for the kernel. This timestamp is not collected by default. Use API cuptiActivityEnableLatencyTimestamps() to enable collection. Command buffer is a buffer written by CUDA driver to send commands like kernel launch, memory copy etc to the GPU. All launches of CUDA kernels are asynchronous with respect to the host, the host requests the launch by writing commands into the command buffer, then returns without checking the GPU’s progress.

Type:

int

registers_per_thread

The number of registers required for each thread executing the kernel.

Type:

int

shared_memory_carveout_requested

Shared memory carveout value requested for the function in percentage of the total resource. The value will be updated only if field isSharedMemoryCarveoutRequested is set.

Type:

int

shared_memory_config

The shared memory configuration used for the kernel. The value is one of the CUsharedconfig enumeration values from cuda.h.

Type:

int

shared_memory_executed

Shared memory size set by the driver.

Type:

int

shmem_limit_config

The shared memory limit config for the kernel. This field shows whether user has opted for a higher per block limit of dynamic shared memory.

Type:

int

start

The start timestamp for the kernel execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the kernel.

Type:

int

static_shared_memory

The static shared memory allocated for the kernel, in bytes.

Type:

int

stream_id

The ID of the stream where the kernel is executing.

Type:

int

submitted

The timestamp when the command buffer containing the kernel launch is submitted to the GPU, in ns. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the submitted time could not be collected for the kernel. This timestamp is not collected by default. Use API cuptiActivityEnableLatencyTimestamps() to enable collection.

Type:

int

Deprecated since version 13.0.0.

class cupti.cupti.ActivityKernel9

Bases: object

Empty-initialize an instance of CUpti_ActivityKernel9.

See also

CUpti_ActivityKernel9

block_x

The X-dimension block size for the kernel.

Type:

int

block_y

The Y-dimension block size for the kernel.

Type:

int

block_z

The Z-dimension grid size for the kernel.

Type:

int

cache_config

For devices with compute capability 7.0+ cacheConfig values are not updated in case field isSharedMemoryCarveoutRequested is set

Type:

_py_anon_pod7

channel_id

The ID of the HW channel on which the kernel is launched.

Type:

int

channel_type

The type of the channel

Type:

int

cluster_scheduling_policy

The cluster scheduling policy for the kernel. Refer CUclusterSchedulingPolicy Field is valid for devices with compute capability 9.0 and higher

Type:

int

cluster_x

The X-dimension cluster size for the kernel. Field is valid for devices with compute capability 9.0 and higher

Type:

int

cluster_y

The Y-dimension cluster size for the kernel. Field is valid for devices with compute capability 9.0 and higher

Type:

int

cluster_z

The Z-dimension cluster size for the kernel. Field is valid for devices with compute capability 9.0 and higher

Type:

int

completed

The completed timestamp for the kernel execution, in ns. It represents the completion of all it’s child kernels and the kernel itself. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the completion time is unknown.

Type:

int

context_id

The ID of the context where the kernel is executing.

Type:

int

correlation_id

The correlation ID of the kernel. Each kernel execution is assigned a unique correlation ID that is identical to the correlation ID in the driver or runtime API activity record that launched the kernel.

Type:

int

device_id

The ID of the device where the kernel is executing.

Type:

int

dynamic_shared_memory

The dynamic shared memory reserved for the kernel, in bytes.

Type:

int

end

The end timestamp for the kernel execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the kernel.

Type:

int

static from_buffer(buffer)

Create an ActivityKernel9 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityKernel9 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_kernel9_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityKernel9 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

The unique ID of the graph that launched this kernel through graph launch APIs. This field will be 0 if the kernel is not launched through graph launch APIs.

Type:

int

graph_node_id

The unique ID of the graph node that launched this kernel through graph launch APIs. This field will be 0 if the kernel is not launched through graph launch APIs.

Type:

int

grid_id

The grid ID of the kernel. Each kernel is assigned a unique grid ID at runtime.

Type:

int

grid_x

The X-dimension grid size for the kernel.

Type:

int

grid_y

The Y-dimension grid size for the kernel.

Type:

int

grid_z

The Z-dimension grid size for the kernel.

Type:

int

is_shared_memory_carveout_requested

This indicates if CU_FUNC_ATTRIBUTE_PREFERRED_SHARED_MEMORY_CARVEOUT was updated for the kernel launch

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_KERNEL or CUPTI_ACTIVITY_KIND_CONCURRENT_KERNEL.

Type:

int

launch_type

The indicates if the kernel was executed via a regular launch or via a single/multi device cooperative launch. CUpti_ActivityLaunchType

Type:

int

local_memory_per_thread

The amount of local memory reserved for each thread, in bytes.

Type:

int

local_memory_total

The total amount of local memory reserved for the kernel, in bytes (deprecated in CUDA 11.8). Refer field localMemoryTotal_v2

Type:

int

local_memory_total_v2

The total amount of local memory reserved for the kernel, in bytes.

Type:

int

max_active_clusters

The maximum clusters that could co-exist on the target device for the kernel

Type:

int

max_potential_cluster_size

The maximum cluster size for the kernel

Type:

int

name

The name of the kernel. This name is shared across all activity records representing the same kernel, and so should not be modified.

Type:

str

p_access_policy_window

The pointer to the access policy window. The structure CUaccessPolicyWindow is defined in cuda.h.

Type:

int

partitioned_global_cache_executed

The partitioned global caching executed for the kernel. Partitioned global caching is required to enable caching on certain chips, such as devices with compute capability 5.2. Partitioned global caching can be automatically disabled if the occupancy requirement of the launch cannot support caching.

Type:

int

partitioned_global_cache_requested

The partitioned global caching requested for the kernel. Partitioned global caching is required to enable caching on certain chips, such as devices with compute capability 5.2.

Type:

int

ptr

Get the pointer address to the data as Python int.

queued

The timestamp when the kernel is queued up in the command buffer, in ns. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the queued time could not be collected for the kernel. This timestamp is not collected by default. Use API cuptiActivityEnableLatencyTimestamps() to enable collection. Command buffer is a buffer written by CUDA driver to send commands like kernel launch, memory copy etc to the GPU. All launches of CUDA kernels are asynchronous with respect to the host, the host requests the launch by writing commands into the command buffer, then returns without checking the GPU’s progress.

Type:

int

registers_per_thread

The number of registers required for each thread executing the kernel.

Type:

int

shared_memory_carveout_requested

Shared memory carveout value requested for the function in percentage of the total resource. The value will be updated only if field isSharedMemoryCarveoutRequested is set.

Type:

int

shared_memory_config

The shared memory configuration used for the kernel. The value is one of the CUsharedconfig enumeration values from cuda.h.

Type:

int

shared_memory_executed

Shared memory size set by the driver.

Type:

int

shmem_limit_config

The shared memory limit config for the kernel. This field shows whether user has opted for a higher per block limit of dynamic shared memory.

Type:

int

start

The start timestamp for the kernel execution, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the kernel.

Type:

int

static_shared_memory

The static shared memory allocated for the kernel, in bytes.

Type:

int

stream_id

The ID of the stream where the kernel is executing.

Type:

int

submitted

The timestamp when the command buffer containing the kernel launch is submitted to the GPU, in ns. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the submitted time could not be collected for the kernel. This timestamp is not collected by default. Use API cuptiActivityEnableLatencyTimestamps() to enable collection.

Type:

int

class cupti.cupti.ActivityMarker2

Bases: object

Empty-initialize an instance of CUpti_ActivityMarker2.

See also

CUpti_ActivityMarker2

domain

The name of the domain to which this marker belongs to. This will be NULL for default domain.

Type:

str

flags_

The flags associated with the marker. CUpti_ActivityFlag

Type:

int

static from_buffer(buffer)

Create an ActivityMarker2 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityMarker2 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_marker2_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityMarker2 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

id

The marker ID.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_MARKER.

Type:

int

name

The marker name for an instantaneous or start marker. This will be NULL for an end marker.

Type:

str

object_id

The identifier for the activity object associated with this marker. ‘objectKind’ indicates which ID is valid for this record.

Type:

ActivityObjectKindId

object_kind

The kind of activity object associated with this marker.

Type:

int

pad

Undefined. Reserved for internal use.

Type:

int

ptr

Get the pointer address to the data as Python int.

timestamp

The timestamp for the marker, in ns. A value of 0 indicates that timestamp information could not be collected for the marker.

Type:

int

class cupti.cupti.ActivityMarkerData2

Bases: object

Empty-initialize an instance of CUpti_ActivityMarkerData2.

See also

CUpti_ActivityMarkerData2

category

The category for the marker.

Type:

int

color

The color for the marker.

Type:

int

cupti_domain_id

CUPTI maintained domain id required for NVTX extended payloads. To parse the payload correctly, the domain id must be used to identify the payload attributes as they are domain specific.

Type:

int

flags_

The flags associated with the marker. CUpti_ActivityFlag

Type:

int

static from_buffer(buffer)

Create an ActivityMarkerData2 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityMarkerData2 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_marker_data2_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityMarkerData2 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

id

The marker ID.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_MARKER_DATA.

Type:

int

payload

The payload value.

Type:

MetricValue

payload_kind

Defines the payload format for the value associated with the marker.

Type:

int

ptr

Get the pointer address to the data as Python int.

Deprecated since version 13.2.0.

class cupti.cupti.ActivityMarkerData

Bases: object

Empty-initialize an instance of CUpti_ActivityMarkerData.

See also

CUpti_ActivityMarkerData

category

The category for the marker.

Type:

int

color

The payload value. The color for the marker.

Type:

int

flags_

The flags associated with the marker. CUpti_ActivityFlag

Type:

int

static from_buffer(buffer)

Create an ActivityMarkerData instance with the memory from the given buffer.

static from_data(data)

Create an ActivityMarkerData instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_marker_data_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityMarkerData instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

id

The marker ID.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_MARKER_DATA.

Type:

int

payload

MetricValue:

payload_kind

Defines the payload format for the value associated with the marker.

Type:

int

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.ActivityMemDecompress

Bases: object

Empty-initialize an instance of CUpti_ActivityMemDecompress.

See also

CUpti_ActivityMemDecompress

channel_id

The ID of the HW channel on which the memory copy is occurring.

Type:

int

channel_type

The type of the channel

Type:

int

context_id

The ID of the context.

Type:

int

correlation_id

The correlation ID of the decompression operations. Each operation is assigned a unique correlation ID that is identical to the correlation ID in the driver API activity record that launched the operation.

Type:

int

device_id

The ID of the device.

Type:

int

end

The end timestamp. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the start time is unknown.

Type:

int

static from_buffer(buffer)

Create an ActivityMemDecompress instance with the memory from the given buffer.

static from_data(data)

Create an ActivityMemDecompress instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_mem_decompress_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityMemDecompress instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_MEM_DECOMPRESS

Type:

int

number_of_operations

The number of operations in the batch.

Type:

int

ptr

Get the pointer address to the data as Python int.

source_bytes

The number of bytes to be read and decompressed in the batch operation.

Type:

int

start

The start timestamp. A value of CUPTI_TIMESTAMP_UNKNOWN indicates that the start time is unknown.

Type:

int

stream_id

The ID of the stream.

Type:

int

class cupti.cupti.ActivityMemcpy7

Bases: object

Empty-initialize an instance of CUpti_ActivityMemcpy7.

See also

CUpti_ActivityMemcpy7

bytes

The number of bytes transferred by the memory copy.

Type:

int

channel_id

The ID of the HW channel on which the memory copy is occurring. This field would be CUPTI_INVALID_CHANNEL_ID if the isDeviceLaunched is set to 1.

Type:

int

channel_type

The type of the channel

Type:

int

context_id

The ID of the context where the memory copy is occurring.

Type:

int

copy_count

The total number of memcopy operations traced in this record. This field is valid for memcpy operations happening using MemcpyBatchAsync APIs in CUDA. In MemcpyBatchAsync APIs, multiple memcpy operations are batched together for optimization purposes based on certain heuristics. For other memcpy operations, this field will be 1.

Type:

int

copy_kind

The kind of the memory copy, stored as a byte to reduce record size. CUpti_ActivityMemcpyKind

Type:

int

correlation_id

The correlation ID of the memory copy. Each memory copy is assigned a unique correlation ID that is identical to the correlation ID in the driver API activity record that launched the memory copy.

Type:

int

device_id

The ID of the device where the memory copy is occurring.

Type:

int

dst_kind

The destination memory kind read by the memory copy, stored as a byte to reduce record size. CUpti_ActivityMemoryKind

Type:

int

end

The end timestamp for the memory copy, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the memory copy.

Type:

int

flags_

The flags associated with the memory copy. CUpti_ActivityFlag

Type:

int

static from_buffer(buffer)

Create an ActivityMemcpy7 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityMemcpy7 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_memcpy7_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityMemcpy7 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

The unique ID of the graph that executed this memcpy through graph launch. This field will be 0 if the memcpy is not done through graph launch.

Type:

int

graph_node_id

The unique ID of the graph node that executed this memcpy through graph launch. This field will be 0 if the memcpy is not done through graph launch.

Type:

int

is_device_launched

This field is used to indicate if the memcpy operation is part of a device graph launch.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_MEMCPY.

Type:

int

pad

Undefined. Reserved for internal use.

Type:

int

pad2

(array of length 3).Reserved for internal use.

Type:

uint8

pad3

Reserved for internal use.

Type:

int

ptr

Get the pointer address to the data as Python int.

runtime_correlation_id

The runtime correlation ID of the memory copy. Each memory copy is assigned a unique runtime correlation ID that is identical to the correlation ID in the runtime API activity record that launched the memory copy.

Type:

int

source_graph_id

The unique graph ID of the node from which the memcpy node was instantiated or last updated. This field will be 0 if the memcpy is not launched through graph launch APIs.

Type:

int

source_graph_node_id

The unique graph node ID of the node from which the memcpy node was instantiated or last updated. This field will be 0 if the memcpy is not launched through graph launch APIs.

Type:

int

src_kind

The source memory kind read by the memory copy, stored as a byte to reduce record size. CUpti_ActivityMemoryKind

Type:

int

start

The start timestamp for the memory copy, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the memory copy.

Type:

int

stream_id

The ID of the stream where the memory copy is occurring.

Type:

int

Deprecated since version 13.4.0.

class cupti.cupti.ActivityMemcpy6

Bases: object

Empty-initialize an instance of CUpti_ActivityMemcpy6.

See also

CUpti_ActivityMemcpy6

bytes

The number of bytes transferred by the memory copy.

Type:

int

channel_id

The ID of the HW channel on which the memory copy is occurring.

Type:

int

channel_type

The type of the channel

Type:

int

context_id

The ID of the context where the memory copy is occurring.

Type:

int

copy_count

The total number of memcopy operations traced in this record. This field is valid for memcpy operations happening using MemcpyBatchAsync APIs in CUDA. In MemcpyBatchAsync APIs, multiple memcpy operations are batched together for optimization purposes based on certain heuristics. For other memcpy operations, this field will be 1.

Type:

int

copy_kind

The kind of the memory copy, stored as a byte to reduce record size. CUpti_ActivityMemcpyKind

Type:

int

correlation_id

The correlation ID of the memory copy. Each memory copy is assigned a unique correlation ID that is identical to the correlation ID in the driver API activity record that launched the memory copy.

Type:

int

device_id

The ID of the device where the memory copy is occurring.

Type:

int

dst_kind

The destination memory kind read by the memory copy, stored as a byte to reduce record size. CUpti_ActivityMemoryKind

Type:

int

end

The end timestamp for the memory copy, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the memory copy.

Type:

int

flags_

The flags associated with the memory copy. CUpti_ActivityFlag

Type:

int

static from_buffer(buffer)

Create an ActivityMemcpy6 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityMemcpy6 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_memcpy6_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityMemcpy6 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

The unique ID of the graph that executed this memcpy through graph launch. This field will be 0 if the memcpy is not done through graph launch.

Type:

int

graph_node_id

The unique ID of the graph node that executed this memcpy through graph launch. This field will be 0 if the memcpy is not done through graph launch.

Type:

int

is_device_launched

This field is used to indicate if the memcpy operation is part of a device graph launch.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_MEMCPY.

Type:

int

pad

Undefined. Reserved for internal use.

Type:

int

pad2

(array of length 3).Reserved for internal use.

Type:

uint8

ptr

Get the pointer address to the data as Python int.

runtime_correlation_id

The runtime correlation ID of the memory copy. Each memory copy is assigned a unique runtime correlation ID that is identical to the correlation ID in the runtime API activity record that launched the memory copy.

Type:

int

src_kind

The source memory kind read by the memory copy, stored as a byte to reduce record size. CUpti_ActivityMemoryKind

Type:

int

start

The start timestamp for the memory copy, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the memory copy.

Type:

int

stream_id

The ID of the stream where the memory copy is occurring.

Type:

int

Deprecated since version 12.8.0.

class cupti.cupti.ActivityMemcpy5

Bases: object

Empty-initialize an instance of CUpti_ActivityMemcpy5.

See also

CUpti_ActivityMemcpy5

bytes

int:

channel_id

int:

channel_type

int:

context_id

int:

copy_kind

int:

correlation_id

int:

device_id

int:

dst_kind

int:

end

int:

flags_

int:

static from_buffer(buffer)

Create an ActivityMemcpy5 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityMemcpy5 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_memcpy5_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityMemcpy5 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

int:

graph_node_id

int:

kind

int:

pad

int:

pad2

int:

ptr

Get the pointer address to the data as Python int.

runtime_correlation_id

int:

src_kind

int:

start

int:

stream_id

int:

class cupti.cupti.ActivityMemcpyPtoP4

Bases: object

Empty-initialize an instance of CUpti_ActivityMemcpyPtoP4.

See also

CUpti_ActivityMemcpyPtoP4

bytes

The number of bytes transferred by the memory copy.

Type:

int

channel_id

The ID of the HW channel on which the memory copy is occurring.

Type:

int

channel_type

The type of the channel

Type:

int

context_id

The ID of the context where the memory copy is occurring.

Type:

int

copy_kind

The kind of the memory copy, stored as a byte to reduce record size. CUpti_ActivityMemcpyKind

Type:

int

correlation_id

The correlation ID of the memory copy. Each memory copy is assigned a unique correlation ID that is identical to the correlation ID in the driver and runtime API activity record that launched the memory copy.

Type:

int

device_id

The ID of the device where the memory copy is occurring.

Type:

int

dst_context_id

The ID of the context owning the memory being copied to.

Type:

int

dst_device_id

The ID of the device where memory is being copied to.

Type:

int

dst_kind

The destination memory kind read by the memory copy, stored as a byte to reduce record size. CUpti_ActivityMemoryKind

Type:

int

end

The end timestamp for the memory copy, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the memory copy.

Type:

int

flags_

The flags associated with the memory copy. CUpti_ActivityFlag

Type:

int

static from_buffer(buffer)

Create an ActivityMemcpyPtoP4 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityMemcpyPtoP4 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_memcpy_pto_p4_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityMemcpyPtoP4 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

The unique ID of the graph that executed this memcpy through graph launch. This field will be 0 if the memcpy is not done through graph launch.

Type:

int

graph_node_id

The unique ID of the graph node that executed the memcpy through graph launch. This field will be 0 if memcpy is not done using graph launch.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_MEMCPY2.

Type:

int

ptr

Get the pointer address to the data as Python int.

src_context_id

The ID of the context owning the memory being copied from.

Type:

int

src_device_id

The ID of the device where memory is being copied from.

Type:

int

src_kind

The source memory kind read by the memory copy, stored as a byte to reduce record size. CUpti_ActivityMemoryKind

Type:

int

start

The start timestamp for the memory copy, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the memory copy.

Type:

int

stream_id

The ID of the stream where the memory copy is occurring.

Type:

int

class cupti.cupti.ActivityMemory

Bases: object

Empty-initialize an instance of CUpti_ActivityMemory.

See also

CUpti_ActivityMemory

address

The virtual address of the allocation

Type:

int

alloc_pc

The program counter of the allocation of memory

Type:

int

bytes

The number of bytes of memory allocated.

Type:

int

context_id

The ID of the context. If context is NULL, contextId is set to CUPTI_INVALID_CONTEXT_ID.

Type:

int

device_id

The ID of the device where the memory allocation is taking place.

Type:

int

end

The end timestamp for the memory operation, i.e. the time when memory was freed, in ns. This will be 0 if memory is not freed in the application

Type:

int

free_pc

The program counter of the freeing of memory. This will be 0 if memory is not freed in the application

Type:

int

static from_buffer(buffer)

Create an ActivityMemory instance with the memory from the given buffer.

static from_data(data)

Create an ActivityMemory instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_memory_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityMemory instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_MEMORY

Type:

int

memory_kind

The memory kind requested by the user

Type:

int

name

Variable name. This name is shared across all activity records representing the same symbol, and so should not be modified.

Type:

str

pad

Undefined. Reserved for internal use.

Type:

int

process_id

The ID of the process to which this record belongs to.

Type:

int

ptr

Get the pointer address to the data as Python int.

start

The start timestamp for the memory operation, i.e. the time when memory was allocated, in ns.

Type:

int

class cupti.cupti.ActivityMemory4

Bases: object

Empty-initialize an instance of CUpti_ActivityMemory4.

See also

CUpti_ActivityMemory4

address

The virtual address of the allocation. The base address of the memory pool.

Type:

int

bytes

The number of bytes of memory allocated.

Type:

int

context_id

The ID of the context. If context is NULL, contextId is set to CUPTI_INVALID_CONTEXT_ID.

Type:

int

correlation_id

The correlation ID of the memory operation. Each memory operation is assigned a unique correlation ID that is identical to the correlation ID in the driver and runtime API activity record that launched the memory operation.

Type:

int

device_id

The ID of the device where the memory operation is taking place.

Type:

int

static from_buffer(buffer)

Create an ActivityMemory4 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityMemory4 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_memory4_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityMemory4 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

is_async

isAsync is set if memory operation happens through async memory APIs.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_MEMORY2

Type:

int

memory_kind

The memory kind requested by the user, CUpti_ActivityMemoryKind.

Type:

int

memory_operation_type

The memory operation requested by the user, CUpti_ActivityMemoryOperationType.

Type:

int

memory_pool_config

The memory pool configuration used for the memory operations.

Type:

_py_anon_pod5

name

Variable name. This name is shared across all activity records representing the same symbol, and so should not be modified.

Type:

str

pad1

Undefined. Reserved for internal use.

Type:

int

pc

The program counter of the memory operation.

Type:

int

process_id

int:

ptr

Get the pointer address to the data as Python int.

source

The shared object or binary that the memory allocation request comes from.

Type:

str

stream_id

The ID of the stream. If memory operation is not async, streamId is set to CUPTI_INVALID_STREAM_ID.

Type:

int

timestamp

The start timestamp for the memory operation, in ns.

Type:

int

class cupti.cupti.ActivityMemoryPool3

Bases: object

Empty-initialize an instance of CUpti_ActivityMemoryPool3.

See also

CUpti_ActivityMemoryPool3

address

The virtual address of the allocation.

Type:

int

correlation_id

The correlation ID of the memory pool operation. Each memory pool operation is assigned a unique correlation ID that is identical to the correlation ID in the driver and runtime API activity record that launched the memory operation.

Type:

int

device_id

The ID of the device where the memory pool is created.

Type:

int

static from_buffer(buffer)

Create an ActivityMemoryPool3 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityMemoryPool3 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_memory_pool3_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityMemoryPool3 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

is_managed_pool

Whether the pool is of managed memory allocation or pinned memory allocation. If it is 0, it is pinned and if it is 1, the memory pool allocation is of managed memory type.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_MEMORY_POOL

Type:

int

memory_pool_operation_type

The memory operation requested by the user, CUpti_ActivityMemoryPoolOperationType.

Type:

int

memory_pool_type

The type of the memory pool, CUpti_ActivityMemoryPoolType

Type:

int

min_bytes_to_keep

The minimum bytes to keep of the memory pool. minBytesToKeep is valid for CUPTI_ACTIVITY_MEMORY_POOL_OPERATION_TYPE_TRIMMED, CUpti_ActivityMemoryPoolOperationType

Type:

int

pad2

(array of length 7).Undefined. Reserved for internal use.

Type:

uint8

process_id

The ID of the process to which this record belongs to.

Type:

int

ptr

Get the pointer address to the data as Python int.

release_threshold

The release threshold of the memory pool. releaseThreshold is valid for CUPTI_ACTIVITY_MEMORY_POOL_TYPE_LOCAL, CUpti_ActivityMemoryPoolType.

Type:

int

size_

The size of the memory pool operation in bytes. size is valid for CUPTI_ACTIVITY_MEMORY_POOL_TYPE_LOCAL, CUpti_ActivityMemoryPoolType.

Type:

int

timestamp

The start timestamp for the memory operation, in ns.

Type:

int

utilized_size

The utilized size of the memory pool. utilizedSize is valid for CUPTI_ACTIVITY_MEMORY_POOL_TYPE_LOCAL, CUpti_ActivityMemoryPoolType.

Type:

int

Deprecated since version 13.0.0.

class cupti.cupti.ActivityMemoryPool2

Bases: object

Empty-initialize an instance of CUpti_ActivityMemoryPool2.

See also

CUpti_ActivityMemoryPool2

address

The virtual address of the allocation.

Type:

int

correlation_id

The correlation ID of the memory pool operation. Each memory pool operation is assigned a unique correlation ID that is identical to the correlation ID in the driver and runtime API activity record that launched the memory operation.

Type:

int

device_id

The ID of the device where the memory pool is created.

Type:

int

static from_buffer(buffer)

Create an ActivityMemoryPool2 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityMemoryPool2 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_memory_pool2_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityMemoryPool2 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_MEMORY_POOL

Type:

int

memory_pool_operation_type

The memory operation requested by the user, CUpti_ActivityMemoryPoolOperationType.

Type:

int

memory_pool_type

The type of the memory pool, CUpti_ActivityMemoryPoolType

Type:

int

min_bytes_to_keep

The minimum bytes to keep of the memory pool. minBytesToKeep is valid for CUPTI_ACTIVITY_MEMORY_POOL_OPERATION_TYPE_TRIMMED, CUpti_ActivityMemoryPoolOperationType

Type:

int

process_id

The ID of the process to which this record belongs to.

Type:

int

ptr

Get the pointer address to the data as Python int.

release_threshold

The release threshold of the memory pool. releaseThreshold is valid for CUPTI_ACTIVITY_MEMORY_POOL_TYPE_LOCAL, CUpti_ActivityMemoryPoolType.

Type:

int

size_

The size of the memory pool operation in bytes. size is valid for CUPTI_ACTIVITY_MEMORY_POOL_TYPE_LOCAL, CUpti_ActivityMemoryPoolType.

Type:

int

timestamp

The start timestamp for the memory operation, in ns.

Type:

int

utilized_size

The utilized size of the memory pool. utilizedSize is valid for CUPTI_ACTIVITY_MEMORY_POOL_TYPE_LOCAL, CUpti_ActivityMemoryPoolType.

Type:

int

class cupti.cupti.ActivityMemset5

Bases: object

Empty-initialize an instance of CUpti_ActivityMemset5.

See also

CUpti_ActivityMemset5

bytes

The number of bytes being set by the memory set.

Type:

int

channel_id

The ID of the HW channel on which the memory set is occurring. This field would be CUPTI_INVALID_CHANNEL_ID if the isDeviceLaunched is set to 1.

Type:

int

channel_type

The type of the channel

Type:

int

context_id

The ID of the context where the memory set is occurring.

Type:

int

correlation_id

The correlation ID of the memory set. Each memory set is assigned a unique correlation ID that is identical to the correlation ID in the driver API activity record that launched the memory set.

Type:

int

device_id

The ID of the device where the memory set is occurring.

Type:

int

end

The end timestamp for the memory set, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the memory set.

Type:

int

flags_

The flags associated with the memset. CUpti_ActivityFlag

Type:

int

static from_buffer(buffer)

Create an ActivityMemset5 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityMemset5 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_memset5_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityMemset5 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

The unique ID of the graph that executed this memset through graph launch. This field will be 0 if the memset is not executed through graph launch.

Type:

int

graph_node_id

The unique ID of the graph node that executed this memset through graph launch. This field will be 0 if the memset is not executed through graph launch.

Type:

int

is_device_launched

This field is used to indicate if the memset operation is part of a device graph launch.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_MEMSET.

Type:

int

memory_kind

The memory kind of the memory set CUpti_ActivityMemoryKind

Type:

int

pad

Undefined. Reserved for internal use.

Type:

int

pad2

(array of length 3).Undefined. Reserved for internal use

Type:

uint8

pad3

Undefined. Reserved for internal use

Type:

int

ptr

Get the pointer address to the data as Python int.

source_graph_id

The unique graph ID of the node from which the memset node was instantiated or last updated. This field will be 0 if the memset is not launched through graph launch APIs.

Type:

int

source_graph_node_id

The unique graph node ID of the node from which the memset node was instantiated or last updated. This field will be 0 if the memset is not launched through graph launch APIs.

Type:

int

start

The start timestamp for the memory set, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the memory set.

Type:

int

stream_id

The ID of the stream where the memory set is occurring.

Type:

int

value

The value being assigned to memory by the memory set.

Type:

int

Deprecated since version 13.4.0.

class cupti.cupti.ActivityMemset4

Bases: object

Empty-initialize an instance of CUpti_ActivityMemset4.

See also

CUpti_ActivityMemset4

bytes

The number of bytes being set by the memory set.

Type:

int

channel_id

The ID of the HW channel on which the memory set is occurring.

Type:

int

channel_type

The type of the channel

Type:

int

context_id

The ID of the context where the memory set is occurring.

Type:

int

correlation_id

The correlation ID of the memory set. Each memory set is assigned a unique correlation ID that is identical to the correlation ID in the driver API activity record that launched the memory set.

Type:

int

device_id

The ID of the device where the memory set is occurring.

Type:

int

end

The end timestamp for the memory set, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the memory set.

Type:

int

flags_

The flags associated with the memset. CUpti_ActivityFlag

Type:

int

static from_buffer(buffer)

Create an ActivityMemset4 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityMemset4 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_memset4_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityMemset4 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

The unique ID of the graph that executed this memset through graph launch. This field will be 0 if the memset is not executed through graph launch.

Type:

int

graph_node_id

The unique ID of the graph node that executed this memset through graph launch. This field will be 0 if the memset is not executed through graph launch.

Type:

int

is_device_launched

This field is used to indicate if the memset operation is part of a device graph launch.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_MEMSET.

Type:

int

memory_kind

The memory kind of the memory set CUpti_ActivityMemoryKind

Type:

int

pad

Undefined. Reserved for internal use.

Type:

int

pad2

(array of length 3).Undefined. Reserved for internal use

Type:

uint8

ptr

Get the pointer address to the data as Python int.

start

The start timestamp for the memory set, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the memory set.

Type:

int

stream_id

The ID of the stream where the memory set is occurring.

Type:

int

value

The value being assigned to memory by the memory set.

Type:

int

class cupti.cupti.ActivityModule

Bases: object

Empty-initialize an instance of CUpti_ActivityModule.

See also

CUpti_ActivityModule

context_id

The ID of the context where the module is loaded.

Type:

int

cubin

The pointer to cubin.

Type:

int

cubin_size

The cubin size.

Type:

int

static from_buffer(buffer)

Create an ActivityModule instance with the memory from the given buffer.

static from_data(data)

Create an ActivityModule instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_module_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityModule instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

id

The module ID.

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_MODULE.

Type:

int

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.ActivityName

Bases: object

Empty-initialize an instance of CUpti_ActivityName.

See also

CUpti_ActivityName

static from_buffer(buffer)

Create an ActivityName instance with the memory from the given buffer.

static from_data(data)

Create an ActivityName instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_name_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityName instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_NAME.

Type:

int

name

The name.

Type:

str

object_id

The identifier for the activity object. ‘objectKind’ indicates which ID is valid for this record.

Type:

ActivityObjectKindId

object_kind

The kind of activity object being named.

Type:

int

pad

Undefined. Reserved for internal use.

Type:

int

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.ActivityObjectKindId

Bases: object

Empty-initialize an instance of CUpti_ActivityObjectKindId.

See also

CUpti_ActivityObjectKindId

dcs

A device object requires that we identify the device ID. A context object requires that we identify both the device and context ID. A stream object requires that we identify device, context, and stream ID.

Type:

_py_anon_pod4

static from_buffer(buffer)

Create an ActivityObjectKindId instance with the memory from the given buffer.

static from_data(data)

Create an ActivityObjectKindId instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_object_kind_id_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityObjectKindId instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pt

A process object requires that we identify the process ID. A thread object requires that we identify both the process and thread ID.

Type:

_py_anon_pod3

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.ActivityOpenAccData

Bases: object

Empty-initialize an instance of CUpti_ActivityOpenAccData.

See also

CUpti_ActivityOpenAccData

async_

int:

async_map

int:

bytes

Number of bytes

Type:

int

cu_context_id

CUDA context id Valid only if deviceType is acc_device_nvidia.

Type:

int

cu_device_id

CUDA device id Valid only if deviceType is acc_device_nvidia.

Type:

int

cu_process_id

The ID of the process where the OpenACC activity is executing.

Type:

int

cu_stream_id

CUDA stream id Valid only if deviceType is acc_device_nvidia.

Type:

int

cu_thread_id

The ID of the thread where the OpenACC activity is executing.

Type:

int

device_number

int:

device_ptr

Device pointer if available

Type:

int

device_type

int:

end

CUPTI end timestamp

Type:

int

end_line_no

int:

event_kind

CUPTI OpenACC event kind (CUpti_OpenAccEventKind)

Type:

int

external_id

The OpenACC correlation ID. Valid only if deviceType is acc_device_nvidia. If not 0, it uniquely identifies this record. It is identical to the externalId in the preceding external correlation record of type CUPTI_EXTERNAL_CORRELATION_KIND_OPENACC.

Type:

int

static from_buffer(buffer)

Create an ActivityOpenAccData instance with the memory from the given buffer.

static from_data(data)

Create an ActivityOpenAccData instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_open_acc_data_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityOpenAccData instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func_end_line_no

int:

func_line_no

int:

func_name

str:

host_ptr

Host pointer if available

Type:

int

implicit

int:

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_OPENACC_DATA.

Type:

int

line_no

int:

parent_construct

int:

ptr

Get the pointer address to the data as Python int.

src_file

str:

start

CUPTI start timestamp

Type:

int

thread_id

ThreadId

Type:

int

var_name

str:

version

int:

class cupti.cupti.ActivityOpenAccLaunch

Bases: object

Empty-initialize an instance of CUpti_ActivityOpenAccLaunch.

See also

CUpti_ActivityOpenAccLaunch

async_

Value of async() clause of the corresponding directive

Type:

int

async_map

Internal asynchronous queue number used

Type:

int

cu_context_id

CUDA context id Valid only if deviceType is acc_device_nvidia.

Type:

int

cu_device_id

CUDA device id Valid only if deviceType is acc_device_nvidia.

Type:

int

cu_process_id

The ID of the process where the OpenACC activity is executing.

Type:

int

cu_stream_id

CUDA stream id Valid only if deviceType is acc_device_nvidia.

Type:

int

cu_thread_id

The ID of the thread where the OpenACC activity is executing.

Type:

int

device_number

Device number

Type:

int

device_type

Device type

Type:

int

end

CUPTI end timestamp

Type:

int

end_line_no

For an OpenACC construct, this contains the line number of the end of the construct. A negative or zero value means the line number is not known.

Type:

int

event_kind

CUPTI OpenACC event kind (CUpti_OpenAccEventKind)

Type:

int

external_id

The OpenACC correlation ID. Valid only if deviceType is acc_device_nvidia. If not 0, it uniquely identifies this record. It is identical to the externalId in the preceding external correlation record of type CUPTI_EXTERNAL_CORRELATION_KIND_OPENACC.

Type:

int

static from_buffer(buffer)

Create an ActivityOpenAccLaunch instance with the memory from the given buffer.

static from_data(data)

Create an ActivityOpenAccLaunch instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_open_acc_launch_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityOpenAccLaunch instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func_end_line_no

The last line number of the function named in func_name. A negative or zero value means the line number is not known.

Type:

int

func_line_no

The line number of the first line of the function named in func_name. A negative or zero value means the line number is not known.

Type:

int

func_name

A pointer to a null-terminated string containing the name of the function in which the event occurred.

Type:

str

implicit

1 for any implicit event, such as an implicit wait at a synchronous data construct 0 otherwise

Type:

int

kernel_name

A pointer to null-terminated string containing the name of the kernel being launched, if known, or a null pointer if not.

Type:

str

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_OPENACC_LAUNCH.

Type:

int

line_no

The line number of the directive or program construct or the starting line number of the OpenACC construct corresponding to the event. A negative or zero value means the line number is not known.

Type:

int

num_gangs

The number of gangs created for this kernel launch

Type:

int

num_workers

The number of workers created for this kernel launch

Type:

int

parent_construct

CUPTI OpenACC parent construct kind (CUpti_OpenAccConstructKind) Note that for applications using PGI OpenACC runtime < 16.1, this will always be CUPTI_OPENACC_CONSTRUCT_KIND_UNKNOWN.

Type:

int

ptr

Get the pointer address to the data as Python int.

src_file

A pointer to null-terminated string containing the name of or path to the source file, if known, or a null pointer if not.

Type:

str

start

CUPTI start timestamp

Type:

int

thread_id

ThreadId

Type:

int

vector_length

The number of vector lanes created for this kernel launch

Type:

int

version

Version number

Type:

int

class cupti.cupti.ActivityOpenAccOther

Bases: object

Empty-initialize an instance of CUpti_ActivityOpenAccOther.

See also

CUpti_ActivityOpenAccOther

async_

Value of async() clause of the corresponding directive

Type:

int

async_map

Internal asynchronous queue number used

Type:

int

cu_context_id

CUDA context id Valid only if deviceType is acc_device_nvidia.

Type:

int

cu_device_id

CUDA device id Valid only if deviceType is acc_device_nvidia.

Type:

int

cu_process_id

The ID of the process where the OpenACC activity is executing.

Type:

int

cu_stream_id

CUDA stream id Valid only if deviceType is acc_device_nvidia.

Type:

int

cu_thread_id

The ID of the thread where the OpenACC activity is executing.

Type:

int

device_number

Device number

Type:

int

device_type

Device type

Type:

int

end

CUPTI end timestamp

Type:

int

end_line_no

For an OpenACC construct, this contains the line number of the end of the construct. A negative or zero value means the line number is not known.

Type:

int

event_kind

CUPTI OpenACC event kind (CUpti_OpenAccEventKind)

Type:

int

external_id

The OpenACC correlation ID. Valid only if deviceType is acc_device_nvidia. If not 0, it uniquely identifies this record. It is identical to the externalId in the preceding external correlation record of type CUPTI_EXTERNAL_CORRELATION_KIND_OPENACC.

Type:

int

static from_buffer(buffer)

Create an ActivityOpenAccOther instance with the memory from the given buffer.

static from_data(data)

Create an ActivityOpenAccOther instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_open_acc_other_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityOpenAccOther instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func_end_line_no

The last line number of the function named in func_name. A negative or zero value means the line number is not known.

Type:

int

func_line_no

The line number of the first line of the function named in func_name. A negative or zero value means the line number is not known.

Type:

int

func_name

A pointer to a null-terminated string containing the name of the function in which the event occurred.

Type:

str

implicit

1 for any implicit event, such as an implicit wait at a synchronous data construct 0 otherwise

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_OPENACC_OTHER.

Type:

int

line_no

The line number of the directive or program construct or the starting line number of the OpenACC construct corresponding to the event. A negative or zero value means the line number is not known.

Type:

int

parent_construct

CUPTI OpenACC parent construct kind (CUpti_OpenAccConstructKind) Note that for applications using PGI OpenACC runtime < 16.1, this will always be CUPTI_OPENACC_CONSTRUCT_KIND_UNKNOWN.

Type:

int

ptr

Get the pointer address to the data as Python int.

src_file

A pointer to null-terminated string containing the name of or path to the source file, if known, or a null pointer if not.

Type:

str

start

CUPTI start timestamp

Type:

int

thread_id

ThreadId

Type:

int

version

Version number

Type:

int

class cupti.cupti.ActivityOpenMp

Bases: object

Empty-initialize an instance of CUpti_ActivityOpenMp.

See also

CUpti_ActivityOpenMp

cu_process_id

The ID of the process where the OpenMP activity is executing.

Type:

int

cu_thread_id

The ID of the thread where the OpenMP activity is executing.

Type:

int

end

CUPTI end timestamp

Type:

int

event_kind

CUPTI OpenMP event kind (CUpti_OpenMpEventKind)

Type:

int

static from_buffer(buffer)

Create an ActivityOpenMp instance with the memory from the given buffer.

static from_data(data)

Create an ActivityOpenMp instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_open_mp_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityOpenMp instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The kind of this activity.

Type:

int

ptr

Get the pointer address to the data as Python int.

start

CUPTI start timestamp

Type:

int

thread_id

ThreadId

Type:

int

version

Version number

Type:

int

class cupti.cupti.ActivityOverhead3

Bases: object

Empty-initialize an instance of CUpti_ActivityOverhead3.

See also

CUpti_ActivityOverhead3

correlation_id

The correlation ID of the overhead operation to which records belong to. This ID is identical to the correlation ID in the driver or runtime API activity record that launched the overhead operation. In some cases, it can be zero, such as for CUPTI_ACTIVITY_OVERHEAD_CUPTI_BUFFER_FLUSH records.

Type:

int

end

The end timestamp for the overhead, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the overhead.

Type:

int

static from_buffer(buffer)

Create an ActivityOverhead3 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityOverhead3 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_overhead3_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityOverhead3 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_OVERHEAD.

Type:

int

object_id

The identifier for the activity object. ‘objectKind’ indicates which ID is valid for this record.

Type:

ActivityObjectKindId

object_kind

The kind of activity object that the overhead is associated with.

Type:

int

overhead_data

Pointer to the struct with additional details about the overhead. Refer CUpti_ActivityOverheadKind enum and the corresponding structure to typecast and access additional overhead data. Client is responsible for freeing this memory using the free function when done.

Type:

int

overhead_kind

The kind of overhead, CUPTI, DRIVER, COMPILER etc.

Type:

int

ptr

Get the pointer address to the data as Python int.

start

The start timestamp for the overhead, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the overhead.

Type:

int

class cupti.cupti.ActivityOverheadCommandBufferFullData

Bases: object

Empty-initialize an instance of CUpti_ActivityOverheadCommandBufferFullData.

See also

CUpti_ActivityOverheadCommandBufferFullData

channel_id

The channel ID of the command buffer.

Type:

int

channel_type

The channel type of the command buffer.

Type:

int

command_buffer_length

The remaining space in the command buffer. This field will always be zero when the command buffer is full, making it not useful in such cases.

Type:

int

static from_buffer(buffer)

Create an ActivityOverheadCommandBufferFullData instance with the memory from the given buffer.

static from_data(data)

Create an ActivityOverheadCommandBufferFullData instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_overhead_command_buffer_full_data_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityOverheadCommandBufferFullData instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.ActivityPreemption

Bases: object

Empty-initialize an instance of CUpti_ActivityPreemption.

See also

CUpti_ActivityPreemption

block_x

The X-dimension of the block that is preempted

Type:

int

block_y

The Y-dimension of the block that is preempted

Type:

int

block_z

The Z-dimension of the block that is preempted

Type:

int

static from_buffer(buffer)

Create an ActivityPreemption instance with the memory from the given buffer.

static from_data(data)

Create an ActivityPreemption instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_preemption_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityPreemption instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

grid_id

The grid-id of the block that is preempted

Type:

int

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_PREEMPTION

Type:

int

pad

Undefined. Reserved for internal use.

Type:

int

preemption_kind

kind of the preemption

Type:

int

ptr

Get the pointer address to the data as Python int.

timestamp

The timestamp of the preemption, in ns. A value of 0 indicates that timestamp information could not be collected for the preemption.

Type:

int

class cupti.cupti.ActivityStream

Bases: object

Empty-initialize an instance of CUpti_ActivityStream.

See also

CUpti_ActivityStream

context_id

The ID of the context where the stream was created.

Type:

int

correlation_id

The correlation ID of the API to which this result is associated.

Type:

int

flag

Flags associated with the stream.

Type:

int

static from_buffer(buffer)

Create an ActivityStream instance with the memory from the given buffer.

static from_data(data)

Create an ActivityStream instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_stream_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ActivityStream instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_STREAM.

Type:

int

priority

The clamped priority for the stream.

Type:

int

ptr

Get the pointer address to the data as Python int.

stream_id

A unique stream ID to identify the stream.

Type:

int

class cupti.cupti.ActivitySynchronization2

Bases: object

Empty-initialize an instance of CUpti_ActivitySynchronization2.

See also

CUpti_ActivitySynchronization2

context_id

The ID of the context for which the synchronization API is called. In case of context synchronization API it is the context id for which the API is called. In case of stream/event synchronization it is the ID of the context where the stream/event was created.

Type:

int

correlation_id

The correlation ID of the API to which this result is associated.

Type:

int

cuda_event_id

The event ID for which the synchronization API is called. A CUPTI_SYNCHRONIZATION_INVALID_VALUE value indicate the field is not applicable for this record. Not valid for cuCtxSynchronize, cuStreamSynchronize.

Type:

int

cuda_event_sync_id

A unique ID to associate event synchronization records with the latest CUDA Event record. Similar field is added in CUpti_ActivityCudaEvent3 to associate synchronization record to the CUDA Event record. The same CUDA event can be used multiple times, so the event id will not be unique to correlate the synchronization record with the latest CUDA Event record. This field will be unique and can be used to do the required correlation. A CUPTI_SYNCHRONIZATION_INVALID_VALUE value indicates that the field is not applicable for this record. Valid only for synchronization records related to CUDA Events.

Type:

int

end

The end timestamp for the function, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the function.

Type:

int

static from_buffer(buffer)

Create an ActivitySynchronization2 instance with the memory from the given buffer.

static from_data(data)

Create an ActivitySynchronization2 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_synchronization2_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivitySynchronization2 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_SYNCHRONIZATION.

Type:

int

pad

Undefined. Reserved for internal use.

Type:

int

ptr

Get the pointer address to the data as Python int.

return_value

The return value for the synchronization record. Use cuptiActivityEnableAllSyncRecords API to enable/disable collection of synchronization records with return value being non-zero. This will be a CUresult value.

Type:

int

start

The start timestamp for the function, in ns. A value of 0 for both the start and end timestamps indicates that timestamp information could not be collected for the function.

Type:

int

stream_id

The compute stream for which the synchronization API is called. A CUPTI_SYNCHRONIZATION_INVALID_VALUE value indicate the field is not applicable for this record. Not valid for cuCtxSynchronize, cuEventSynchronize.

Type:

int

type

The type of record.

Type:

int

Deprecated since version 12.8.0.

class cupti.cupti.ActivitySynchronization

Bases: object

Empty-initialize an instance of CUpti_ActivitySynchronization.

See also

CUpti_ActivitySynchronization

context_id

int:

correlation_id

int:

cuda_event_id

int:

end

int:

static from_buffer(buffer)

Create an ActivitySynchronization instance with the memory from the given buffer.

static from_data(data)

Create an ActivitySynchronization instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_synchronization_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivitySynchronization instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

int:

ptr

Get the pointer address to the data as Python int.

start

int:

stream_id

int:

type

int:

class cupti.cupti.ActivityUnifiedMemoryCounter3

Bases: object

Empty-initialize an instance of CUpti_ActivityUnifiedMemoryCounter3.

See also

CUpti_ActivityUnifiedMemoryCounter3

address

This is the virtual base address of the page/s being transferred. For cpu and gpu faults, the virtual address for the page that faulted.

Type:

int

counter_kind

The Unified Memory counter kind

Type:

int

dst_id

The ID of the destination CPU/device involved in the memory transfer or remote map operation. Ignore this field if counterKind is CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_GPU_PAGE_FAULT or CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_CPU_PAGE_FAULT_COUNT or CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_THRASHING or CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_THROTTLING

Type:

int

end

The end timestamp of the counter, in ns. Ignore this field if counterKind is CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_CPU_PAGE_FAULT_COUNT or CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_THRASHING or CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_REMOTE_MAP. For counterKind CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_BYTES_TRANSFER_HTOD and CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_BYTES_TRANSFER_DTOH, timestamp is captured when activity finishes on GPU. For counterKind CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_GPU_PAGE_FAULT, timestamp is captured when CUDA driver queues the replay of faulting memory accesses on the GPU For counterKind CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_THROTTLING, timestamp is captured when throttling operation was finished by CUDA driver

Type:

int

flags_

The flags associated with this record. See enums CUpti_ActivityUnifiedMemoryAccessType if counterKind is CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_GPU_PAGE_FAULT and CUpti_ActivityUnifiedMemoryMigrationCause if counterKind is CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_BYTES_TRANSFER_HTOD or CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_BYTES_TRANSFER_HTOD and CUpti_ActivityUnifiedMemoryRemoteMapCause if counterKind is CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_REMOTE_MAP and CUpti_ActivityFlag if counterKind is CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_THRASHING or CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_THROTTLING

Type:

int

static from_buffer(buffer)

Create an ActivityUnifiedMemoryCounter3 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityUnifiedMemoryCounter3 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_unified_memory_counter3_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityUnifiedMemoryCounter3 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

The activity record kind, must be CUPTI_ACTIVITY_KIND_UNIFIED_MEMORY_COUNTER

Type:

int

pad

Undefined. Reserved for internal use.

Type:

int

process_id

The ID of the process to which this record belongs to.

Type:

int

processors

(array of length 5).The bitmask of devices involved in the operation. For counterKind CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_THRASHING, it is a bitwise ORing of the device IDs fighting for the memory region. processors[0] represents the device ID of the device 0 to device 63, processors[1] represents device ID of device 64 to device 127 and so on. Ignore this field if counterKind is CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_CPU_PAGE_FAULT_COUNT or CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_GPU_PAGE_FAULT or CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_THROTTLING or CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_REMOTE_MAP or CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_BYTES_TRANSFER_HTOD or CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_BYTES_TRANSFER_DTOH or CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_DTOD or CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_GPU_FAULT_REPLAY

Type:

uint64

ptr

Get the pointer address to the data as Python int.

src_id

The ID of the source CPU/device involved in the memory transfer, page fault, thrashing, throttling or remote map operation. For counterKind CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_THRASHING, it is a bitwise ORing of the device IDs fighting for the memory region, ONLY if there are less than 32 devices. Ignore this field if counterKind is CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_CPU_PAGE_FAULT_COUNT

Type:

int

start

The start timestamp of the counter, in ns. For counterKind CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_BYTES_TRANSFER_HTOD and CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_BYTES_TRANSFER_DTOH, timestamp is captured when activity starts on GPU. For counterKind CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_GPU_PAGE_FAULT and CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_CPU_PAGE_FAULT_COUNT, timestamp is captured when CUDA driver started processing the fault. For counterKind CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_THRASHING, timestamp is captured when CUDA driver detected thrashing of memory region. For counterKind CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_THROTTLING, timestamp is captured when throttling operation was started by CUDA driver. For counterKind CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_REMOTE_MAP, timestamp is captured when CUDA driver has pushed all required operations to the processor specified by dstId.

Type:

int

stream_id

The ID of the stream causing the transfer. This value of this field is invalid.

Type:

int

value

Value of the counter For counterKind CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_BYTES_TRANSFER_HTOD, CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_BYTES_TRANSFER_DTOH, CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_THREASHING and CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_REMOTE_MAP, it is the size of the memory region in bytes. For counterKind CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_GPU_PAGE_FAULT, it is the number of page fault groups for the same page. For counterKind CUPTI_ACTIVITY_UNIFIED_MEMORY_COUNTER_KIND_CPU_PAGE_FAULT_COUNT, it is the program counter for the instruction that caused fault.

Type:

int

Deprecated since version 12.8.0.

class cupti.cupti.ActivityUnifiedMemoryCounter2

Bases: object

Empty-initialize an instance of CUpti_ActivityUnifiedMemoryCounter2.

See also

CUpti_ActivityUnifiedMemoryCounter2

address

int:

counter_kind

int:

dst_id

int:

end

int:

flags_

int:

static from_buffer(buffer)

Create an ActivityUnifiedMemoryCounter2 instance with the memory from the given buffer.

static from_data(data)

Create an ActivityUnifiedMemoryCounter2 instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype activity_unified_memory_counter2_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ActivityUnifiedMemoryCounter2 instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

int:

pad

int:

process_id

int:

ptr

Get the pointer address to the data as Python int.

src_id

int:

start

int:

stream_id

int:

value

int:

class cupti.cupti.ActivityUnifiedMemoryCounterConfig

Bases: object

Empty-initialize an array of CUpti_ActivityUnifiedMemoryCounterConfig. The resulting object is of length size and of dtype activity_unified_memory_counter_config_dtype. If default-constructed, the instance represents a single struct.

Parameters:

size (int) – number of structs, default=1.

See also

CUpti_ActivityUnifiedMemoryCounterConfig

device_id

Device id of the target device. This is relevant only for single device scopes. (deprecated in CUDA 7.0)

Type:

Union[uint32, int]

enable

Control to enable/disable the counter. To enable the counter set it to non-zero value while disable is indicated by zero.

Type:

Union[uint32, int]

static from_buffer(buffer)

Create an ActivityUnifiedMemoryCounterConfig instance with the memory from the given buffer.

static from_data(data)

Create an ActivityUnifiedMemoryCounterConfig instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a 1D array of dtype activity_unified_memory_counter_config_dtype holding the data.

static from_ptr(

ptr: int,

size: int = 1,

readonly: bool = False,

owner=None,

)

Create an ActivityUnifiedMemoryCounterConfig instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• size (int) – number of structs, default=1.

• readonly (bool) – whether the data is read-only (to the user). default is False.

• owner (object) – object that owns the memory at ptr. A strong reference is kept so the backing storage outlives this wrapper.

kind

Unified Memory counter Counter kind

Type:

Union[int32, int]

ptr

Get the pointer address to the data as Python int.

scope

Unified Memory counter Counter scope. (deprecated in CUDA 7.0)

Type:

Union[int32, int]

class cupti.cupti.MetricValue

Bases: object

Empty-initialize an instance of CUpti_MetricValue.

See also

CUpti_MetricValue

static from_buffer(buffer)

Create an MetricValue instance with the memory from the given buffer.

static from_data(data)

Create an MetricValue instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype metric_value_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an MetricValue instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

metric_value_double

float:

metric_value_int64

int:

metric_value_nvtx_extended_payload

Value for CUPTI_METRIC_VALUE_KIND_NVTX_EXTENDED_PAYLOAD.

Type:

int

metric_value_percent

float:

metric_value_throughput

int:

metric_value_uint64

int:

metric_value_utilization_level

int:

ptr

Get the pointer address to the data as Python int.

Callback API

Functions

cupti.cupti.enable_all_domains(enable: int, subscriber: int)

Enable or disable all callbacks in all domains.

Parameters:

• enable (uint32_t) – New enable state for all callbacks in all domain. Zero disables all callbacks, non-zero enables all callbacks.

• subscriber (intptr_t) – Handle to callback subscription.

See also

cuptiEnableAllDomains

cupti.cupti.enable_callback(enable: int, subscriber: int, domain: int, cbid: int)

Enable or disabled callbacks for a specific domain and callback ID.

Parameters:

• enable (uint32_t) – New enable state for the callback. Zero disables the callback, non-zero enables the callback.

• subscriber (intptr_t) – Handle to callback subscription.

• domain (CallbackDomain) – The domain of the callback.

• cbid (uint32_t) – The ID of the callback.

See also

cuptiEnableCallback

cupti.cupti.enable_domain(enable: int, subscriber: int, domain: int)

Enable or disabled all callbacks for a specific domain.

Parameters:

• enable (uint32_t) – New enable state for all callbacks in the domain. Zero disables all callbacks, non-zero enables all callbacks.

• subscriber (intptr_t) – Handle to callback subscription.

• domain (CallbackDomain) – The domain of the callback.

See also

cuptiEnableDomain

cupti.cupti.get_callback_name(domain: int, cbid: int)

Get the name of a callback for a specific domain and callback ID.

Parameters:

• domain (CallbackDomain) – The domain of the callback.

• cbid (uint32_t) – The ID of the callback.

Returns:

Returns name of the callback for the specified domain and callback ID

Return type:

name (str)

See also

cuptiGetCallbackName

cupti.cupti.get_callback_state(subscriber: int, domain: int, cbid: int) → int

Get the current enabled/disabled state of a callback for a specific domain and function ID.

Parameters:

• subscriber (intptr_t) – Handle to the initialize subscriber.

• domain (CallbackDomain) – The domain of the callback.

• cbid (uint32_t) – The ID of the callback.

Returns:

Returns non-zero if callback enabled, zero if not

enabled.

Return type:

uint32_t

See also

cuptiGetCallbackState

cupti.cupti.get_enabled_callbacks(subscriber: int, domain: int)

Get the callback ids currently enabled for a subscriber in a callback domain.

Parameters:

• subscriber (intptr_t) – The subscriber handle. Pass 0 (NULL) to get the union of callbacks enabled across all subscribers for the domain.

• domain (CallbackDomain) – The callback domain to query.

Returns:

the enabled callback ids for the domain (empty list if none are enabled).

Return type:

list[int]

See also

cuptiGetEnabledCallbacks

cupti.cupti.subscribe(callback, userdata) → int

Initialize a callback subscriber with a callback function and user data.

Parameters:

• callback (object) – The callback function.

• userdata (object) – A pointer to user data. This data will be passed to the callback function via the userdata parameter.

Returns:

Returns handle to initialize subscriber.

Return type:

intptr_t

See also

cuptiSubscribe

cupti.cupti.subscribe_v2(callback, userdata, p_params) → int

Initialize a callback subscriber with a callback function and user data.

Parameters:

• callback (object) – The callback function.

• userdata (object) – A pointer to user data. This data will be passed to the callback function via the userdata parameter.

• p_params (intptr_t) – A pointer to CUpti_SubscriberParams. Can be NULL.

Returns:

Returns handle to initialize subscriber.

Return type:

intptr_t

See also

cuptiSubscribe_v2

Note

For cupti.cupti.subscribe() and cupti.cupti.subscribe_v2(), the callback argument may be None for subscribers that only use the Activity APIs and do not need callbacks.

cupti.cupti.supported_domains()

Get the available callback domains.

Returns:

List of all available callback domains

Return type:

list[CallbackDomain]

See also

cuptiSupportedDomains

cupti.cupti.unsubscribe(subscriber: int)

Unregister a callback subscriber.

Parameters:

subscriber (intptr_t) – Handle to the initialize subscriber.

See also

cuptiUnsubscribe

Enums

class cupti.cupti.ApiCallbackSite(value)

Bases: IntEnum

Specifies the point in an API call that a callback is issued. Specifies the point in an API call that a callback is issued. This value is communicated to the callback function via CUpti_CallbackData.callbackSite.

See CUpti_ApiCallbackSite.

API_CBSITE_FORCE_INT = 2147483647

API_ENTER = 0

API_EXIT = 1

class cupti.cupti.CallbackDomain(value)

Bases: IntEnum

Callback domains. Callback domains. Each domain represents callback points for a group of related API functions or CUDA driver activity.

See CUpti_CallbackDomain.

DRIVER_API = 1

FORCE_INT = 2147483647

INVALID = 0

NVTX = 5

RESOURCE = 3

RUNTIME_API = 2

SIZE = 7

STATE = 6

SYNCHRONIZE = 4

class cupti.cupti.CallbackIdResource(value)

Bases: IntEnum

Callback IDs for resource domain. Callback IDs for resource domain, CUPTI_CB_DOMAIN_RESOURCE. This value is communicated to the callback function via the cbid parameter.

See CUpti_CallbackIdResource.

CONTEXT_CREATED = 1

CONTEXT_DESTROY_STARTING = 2

CU_INIT_FINISHED = 5

FORCE_INT = 2147483647

GRAPHEXEC_CREATED = 18

GRAPHEXEC_CREATE_STARTING = 17

GRAPHEXEC_DESTROY_STARTING = 19

GRAPHNODE_CLONED = 20

GRAPHNODE_CREATED = 13

GRAPHNODE_CREATE_STARTING = 12

GRAPHNODE_DEPENDENCY_CREATED = 15

GRAPHNODE_DEPENDENCY_DESTROY_STARTING = 16

GRAPHNODE_DESTROY_STARTING = 14

GRAPH_CLONED = 11

GRAPH_CREATED = 9

GRAPH_DESTROY_STARTING = 10

GRAPH_NODE_SET_PARAMS = 23

GRAPH_NODE_UPDATED = 22

INVALID = 0

LOGICAL_ENDPOINT_CREATED = 24

LOGICAL_ENDPOINT_DESTROYED = 25

MODULE_LOADED = 6

MODULE_PROFILED = 8

MODULE_UNLOAD_STARTING = 7

SIZE = 26

STREAM_ATTRIBUTE_CHANGED = 21

STREAM_CREATED = 3

STREAM_DESTROY_STARTING = 4

class cupti.cupti.CallbackIdState(value)

Bases: IntEnum

Callback IDs for state domain. Callback IDs for state domain, CUPTI_CB_DOMAIN_STATE. This value is communicated to the callback function via the cbid parameter.

See CUpti_CallbackIdState.

ERROR = 2

FATAL_ERROR = 1

FORCE_INT = 2147483647

INVALID = 0

SIZE = 4

WARNING = 3

class cupti.cupti.CallbackIdSync(value)

Bases: IntEnum

Callback IDs for synchronization domain. Callback IDs for synchronization domain, CUPTI_CB_DOMAIN_SYNCHRONIZE. This value is communicated to the callback function via the cbid parameter.

See CUpti_CallbackIdSync.

CONTEXT_SYNCHRONIZED = 2

FORCE_INT = 2147483647

INVALID = 0

SIZE = 3

STREAM_SYNCHRONIZED = 1

class cupti.cupti.DriverApiTraceCbid(value)

Bases: IntEnum

See CUpti_driver_api_trace_cbid.

FORCE_INT = 2147483647

INVALID = 0

SIZE = 877

cuArray3DCreate = 90

cuArray3DCreate_v2 = 274

cuArray3DGetDescriptor = 91

cuArray3DGetDescriptor_v2 = 275

cuArrayCreate = 87

cuArrayCreate_v2 = 272

cuArrayDestroy = 89

cuArrayGetDescriptor = 88

cuArrayGetDescriptor_v2 = 273

cuArrayGetMemoryRequirements = 654

cuArrayGetPlane = 597

cuArrayGetSparseProperties = 582

cuCheckpointOperationComplete = 861

cuCheckpointProcessCheckpoint = 771

cuCheckpointProcessGetRestoreThreadId = 768

cuCheckpointProcessGetState = 769

cuCheckpointProcessLock = 770

cuCheckpointProcessRestore = 772

cuCheckpointProcessUnlock = 773

cuCoredumpDeregisterCompleteCallback = 845

cuCoredumpDeregisterStartCallback = 844

cuCoredumpGetAttribute = 701

cuCoredumpGetAttributeGlobal = 702

cuCoredumpRegisterCompleteCallback = 843

cuCoredumpRegisterStartCallback = 842

cuCoredumpSetAttribute = 703

cuCoredumpSetAttributeGlobal = 704

cuCtxAttach = 12

cuCtxCreate = 10

cuCtxCreate_v2 = 235

cuCtxCreate_v3 = 645

cuCtxCreate_v4 = 757

cuCtxDestroy = 11

cuCtxDestroy_v2 = 322

cuCtxDetach = 13

cuCtxDisablePeerAccess = 314

cuCtxEnablePeerAccess = 313

cuCtxFromGreenCtx = 753

cuCtxGetApiVersion = 296

cuCtxGetCacheConfig = 299

cuCtxGetCurrent = 304

cuCtxGetDevResource = 746

cuCtxGetDevice = 16

cuCtxGetDevice_v2 = 795

cuCtxGetExecAffinity = 646

cuCtxGetFlags = 391

cuCtxGetId = 695

cuCtxGetLimit = 137

cuCtxGetSharedMemConfig = 337

cuCtxGetStreamPriorityRange = 370

cuCtxPopCurrent = 15

cuCtxPopCurrent_v2 = 324

cuCtxPushCurrent = 14

cuCtxPushCurrent_v2 = 323

cuCtxRecordEvent = 755

cuCtxResetPersistingL2Cache = 568

cuCtxSetCacheConfig = 300

cuCtxSetCurrent = 303

cuCtxSetFlags = 705

cuCtxSetLimit = 136

cuCtxSetSharedMemConfig = 336

cuCtxSynchronize = 17

cuCtxSynchronize_v2 = 800

cuCtxWaitEvent = 756

cuD3D10CtxCreate = 139

cuD3D10CtxCreateOnDevice = 212

cuD3D10CtxCreate_v2 = 236

cuD3D10GetDevice = 138

cuD3D10GetDevices = 211

cuD3D10GetDirect3DDevice = 297

cuD3D10MapResources = 143

cuD3D10RegisterResource = 141

cuD3D10ResourceGetMappedArray = 146

cuD3D10ResourceGetMappedPitch = 149

cuD3D10ResourceGetMappedPitch_v2 = 262

cuD3D10ResourceGetMappedPointer = 147

cuD3D10ResourceGetMappedPointer_v2 = 260

cuD3D10ResourceGetMappedSize = 148

cuD3D10ResourceGetMappedSize_v2 = 261

cuD3D10ResourceGetSurfaceDimensions = 150

cuD3D10ResourceGetSurfaceDimensions_v2 = 263

cuD3D10ResourceSetMapFlags = 145

cuD3D10UnmapResources = 144

cuD3D10UnregisterResource = 142

cuD3D11CtxCreate = 152

cuD3D11CtxCreateOnDevice = 210

cuD3D11CtxCreate_v2 = 237

cuD3D11GetDevice = 151

cuD3D11GetDevices = 209

cuD3D11GetDirect3DDevice = 298

cuD3D9Begin = 168

cuD3D9CtxCreate = 155

cuD3D9CtxCreateOnDevice = 214

cuD3D9CtxCreate_v2 = 238

cuD3D9End = 169

cuD3D9GetDevice = 154

cuD3D9GetDevices = 213

cuD3D9GetDirect3DDevice = 157

cuD3D9MapResources = 160

cuD3D9MapVertexBuffer = 171

cuD3D9MapVertexBuffer_v2 = 268

cuD3D9RegisterResource = 158

cuD3D9RegisterVertexBuffer = 170

cuD3D9ResourceGetMappedArray = 164

cuD3D9ResourceGetMappedPitch = 167

cuD3D9ResourceGetMappedPitch_v2 = 267

cuD3D9ResourceGetMappedPointer = 165

cuD3D9ResourceGetMappedPointer_v2 = 265

cuD3D9ResourceGetMappedSize = 166

cuD3D9ResourceGetMappedSize_v2 = 266

cuD3D9ResourceGetSurfaceDimensions = 163

cuD3D9ResourceGetSurfaceDimensions_v2 = 264

cuD3D9ResourceSetMapFlags = 162

cuD3D9UnmapResources = 161

cuD3D9UnmapVertexBuffer = 172

cuD3D9UnregisterResource = 159

cuD3D9UnregisterVertexBuffer = 173

cuDestroyExternalMemory = 488

cuDestroyExternalSemaphore = 494

cuDevResourceGenerateDesc = 748

cuDevSmResourceSplit = 822

cuDevSmResourceSplitByCount = 751

cuDeviceCanAccessPeer = 312

cuDeviceComputeCapability = 6

cuDeviceGet = 3

cuDeviceGetAttribute = 9

cuDeviceGetByPCIBusId = 331

cuDeviceGetCliqueCount = 866

cuDeviceGetCliqueInfo = 867

cuDeviceGetCount = 4

cuDeviceGetDefaultMemPool = 606

cuDeviceGetDevResource = 745

cuDeviceGetExecAffinitySupport = 644

cuDeviceGetFabricClusterUuid = 865

cuDeviceGetGraphMemAttribute = 641

cuDeviceGetHostAtomicCapabilities = 805

cuDeviceGetLuid = 532

cuDeviceGetMemPool = 610

cuDeviceGetName = 5

cuDeviceGetNvSciSyncAttributes = 542

cuDeviceGetP2PAtomicCapabilities = 804

cuDeviceGetP2PAttribute = 454

cuDeviceGetPCIBusId = 332

cuDeviceGetProperties = 8

cuDeviceGetTexture1DLinearMaxWidth = 579

cuDeviceGetUuid = 482

cuDeviceGetUuid_v2 = 647

cuDeviceGraphMemTrim = 640

cuDevicePrimaryCtxGetState = 392

cuDevicePrimaryCtxRelease = 387

cuDevicePrimaryCtxRelease_v2 = 544

cuDevicePrimaryCtxReset = 389

cuDevicePrimaryCtxReset_v2 = 545

cuDevicePrimaryCtxRetain = 386

cuDevicePrimaryCtxSetFlags = 388

cuDevicePrimaryCtxSetFlags_v2 = 546

cuDeviceRegisterAsyncNotification = 735

cuDeviceSetGraphMemAttribute = 642

cuDeviceSetMemPool = 609

cuDeviceTotalMem = 7

cuDeviceTotalMem_v2 = 259

cuDeviceUnregisterAsyncNotification = 736

cuDriverGetVersion = 2

cuEGLStreamConsumerAcquireFrame = 395

cuEGLStreamConsumerConnect = 393

cuEGLStreamConsumerConnectWithFlags = 470

cuEGLStreamConsumerDisconnect = 394

cuEGLStreamConsumerReleaseFrame = 396

cuEGLStreamProducerConnect = 446

cuEGLStreamProducerDisconnect = 447

cuEGLStreamProducerPresentFrame = 448

cuEGLStreamProducerReturnFrame = 453

cuEventCreate = 118

cuEventCreateFromEGLSync = 479

cuEventDestroy = 122

cuEventDestroy_v2 = 325

cuEventElapsedTime = 123

cuEventElapsedTime_v2 = 780

cuEventQuery = 120

cuEventRecord = 119

cuEventRecordWithFlags = 587

cuEventRecordWithFlags_ptsz = 588

cuEventRecord_ptsz = 441

cuEventSynchronize = 121

cuExternalMemoryGetMappedBuffer = 486

cuExternalMemoryGetMappedMipmappedArray = 487

cuFlushGPUDirectRDMAWrites = 627

cuFuncGetAttribute = 85

cuFuncGetModule = 566

cuFuncGetName = 718

cuFuncGetParamCount = 835

cuFuncGetParamInfo = 733

cuFuncIsLoaded = 741

cuFuncLoad = 742

cuFuncSetAttribute = 481

cuFuncSetBlockShape = 83

cuFuncSetCacheConfig = 86

cuFuncSetSharedMemConfig = 338

cuFuncSetSharedSize = 84

cuGLCtxCreate = 174

cuGLCtxCreate_v2 = 239

cuGLGetDevices = 333

cuGLGetDevices_v2 = 385

cuGLInit = 178

cuGLMapBufferObject = 180

cuGLMapBufferObjectAsync = 184

cuGLMapBufferObjectAsync_v2 = 270

cuGLMapBufferObjectAsync_v2_ptsz = 445

cuGLMapBufferObject_v2 = 269

cuGLMapBufferObject_v2_ptds = 417

cuGLRegisterBufferObject = 179

cuGLSetBufferObjectMapFlags = 183

cuGLUnmapBufferObject = 181

cuGLUnmapBufferObjectAsync = 185

cuGLUnregisterBufferObject = 182

cuGetErrorName = 373

cuGetErrorString = 372

cuGetExportTable = 135

cuGetProcAddress = 626

cuGetProcAddress_v2 = 677

cuGraphAddBatchMemOpNode = 669

cuGraphAddChildGraphNode = 525

cuGraphAddDependencies = 518

cuGraphAddDependencies_v2 = 727

cuGraphAddEmptyNode = 526

cuGraphAddEventRecordNode = 589

cuGraphAddEventWaitNode = 590

cuGraphAddExternalSemaphoresSignalNode = 618

cuGraphAddExternalSemaphoresWaitNode = 621

cuGraphAddHostNode = 530

cuGraphAddKernelNode = 502

cuGraphAddKernelNode_v2 = 689

cuGraphAddMemAllocNode = 638

cuGraphAddMemFreeNode = 639

cuGraphAddMemcpyNode = 504

cuGraphAddMemsetNode = 506

cuGraphAddNode = 712

cuGraphAddNode_v2 = 723

cuGraphAddNode_v3 = 863

cuGraphBatchMemOpNodeGetParams = 670

cuGraphBatchMemOpNodeSetParams = 671

cuGraphChildGraphNodeGetGraph = 529

cuGraphClone = 523

cuGraphConditionalHandleCreate = 722

cuGraphCreate = 501

cuGraphDebugDotPrint = 628

cuGraphDestroy = 517

cuGraphDestroyNode = 522

cuGraphEventRecordNodeGetEvent = 591

cuGraphEventRecordNodeSetEvent = 593

cuGraphEventWaitNodeGetEvent = 592

cuGraphEventWaitNodeSetEvent = 594

cuGraphExecBatchMemOpNodeSetParams = 672

cuGraphExecChildGraphNodeSetParams = 586

cuGraphExecDestroy = 516

cuGraphExecEventRecordNodeSetEvent = 595

cuGraphExecEventWaitNodeSetEvent = 596

cuGraphExecExternalSemaphoresSignalNodeSetParams = 624

cuGraphExecExternalSemaphoresWaitNodeSetParams = 625

cuGraphExecGetFlags = 658

cuGraphExecGetId = 813

cuGraphExecHostNodeSetParams = 564

cuGraphExecKernelNodeSetParams = 538

cuGraphExecKernelNodeSetParams_v2 = 692

cuGraphExecMemcpyNodeSetParams = 562

cuGraphExecMemsetNodeSetParams = 563

cuGraphExecNodeSetParams = 714

cuGraphExecUpdate = 561

cuGraphExecUpdate_v2 = 696

cuGraphExternalSemaphoresSignalNodeGetParams = 619

cuGraphExternalSemaphoresSignalNodeSetParams = 620

cuGraphExternalSemaphoresWaitNodeGetParams = 622

cuGraphExternalSemaphoresWaitNodeSetParams = 623

cuGraphGetEdges = 535

cuGraphGetEdges_v2 = 724

cuGraphGetId = 812

cuGraphGetNodes = 534

cuGraphGetRootNodes = 510

cuGraphHostNodeGetParams = 531

cuGraphHostNodeSetParams = 533

cuGraphInstantiate = 513

cuGraphInstantiateWithFlags = 643

cuGraphInstantiateWithParams = 656

cuGraphInstantiateWithParams_ptsz = 657

cuGraphInstantiate_v2 = 578

cuGraphKernelNodeCopyAttributes = 569

cuGraphKernelNodeGetAttribute = 570

cuGraphKernelNodeGetParams = 503

cuGraphKernelNodeGetParams_v2 = 690

cuGraphKernelNodeSetAttribute = 571

cuGraphKernelNodeSetParams = 521

cuGraphKernelNodeSetParams_v2 = 691

cuGraphLaunch = 514

cuGraphLaunch_ptsz = 515

cuGraphMemAllocNodeGetParams = 648

cuGraphMemFreeNodeGetParams = 649

cuGraphMemcpyNodeGetParams = 505

cuGraphMemcpyNodeSetParams = 520

cuGraphMemsetNodeGetParams = 507

cuGraphMemsetNodeSetParams = 508

cuGraphNodeFindInClone = 524

cuGraphNodeGetContainingGraph = 809

cuGraphNodeGetDependencies = 511

cuGraphNodeGetDependencies_v2 = 725

cuGraphNodeGetDependentNodes = 512

cuGraphNodeGetDependentNodes_v2 = 726

cuGraphNodeGetEnabled = 651

cuGraphNodeGetLocalId = 810

cuGraphNodeGetParams = 841

cuGraphNodeGetToolsId = 811

cuGraphNodeGetType = 509

cuGraphNodeSetEnabled = 650

cuGraphNodeSetParams = 713

cuGraphNodeSetParams_v2 = 864

cuGraphReleaseUserObject = 637

cuGraphRemoveDependencies = 519

cuGraphRemoveDependencies_v2 = 728

cuGraphRetainUserObject = 636

cuGraphUpload = 580

cuGraphUpload_ptsz = 581

cuGraphicsD3D10RegisterResource = 140

cuGraphicsD3D11RegisterResource = 153

cuGraphicsD3D9RegisterResource = 156

cuGraphicsEGLRegisterImage = 390

cuGraphicsGLRegisterBuffer = 175

cuGraphicsGLRegisterImage = 176

cuGraphicsMapResources = 133

cuGraphicsMapResources_ptsz = 443

cuGraphicsResourceGetMappedEglFrame = 449

cuGraphicsResourceGetMappedMipmappedArray = 360

cuGraphicsResourceGetMappedPointer = 130

cuGraphicsResourceGetMappedPointer_v2 = 258

cuGraphicsResourceSetMapFlags = 132

cuGraphicsResourceSetMapFlags_v2 = 380

cuGraphicsSubResourceGetMappedArray = 129

cuGraphicsUnmapResources = 134

cuGraphicsUnmapResources_ptsz = 444

cuGraphicsUnregisterResource = 128

cuGraphicsVDPAURegisterOutputSurface = 189

cuGraphicsVDPAURegisterVideoSurface = 188

cuGreenCtxCreate = 743

cuGreenCtxDestroy = 744

cuGreenCtxGetDevResource = 747

cuGreenCtxGetId = 782

cuGreenCtxRecordEvent = 749

cuGreenCtxStreamCreate = 758

cuGreenCtxWaitEvent = 750

cuImportExternalMemory = 485

cuImportExternalSemaphore = 489

cuInit = 1

cuIpcCloseMemHandle = 330

cuIpcGetEventHandle = 334

cuIpcGetMemHandle = 328

cuIpcOpenEventHandle = 335

cuIpcOpenMemHandle = 329

cuIpcOpenMemHandle_v2 = 567

cuKernelGetAttribute = 686

cuKernelGetFunction = 683

cuKernelGetLibrary = 754

cuKernelGetName = 719

cuKernelGetParamCount = 836

cuKernelGetParamInfo = 734

cuKernelSetAttribute = 687

cuKernelSetCacheConfig = 688

cuLaunch = 115

cuLaunchCooperativeKernel = 477

cuLaunchCooperativeKernelMultiDevice = 480

cuLaunchCooperativeKernel_ptsz = 478

cuLaunchGrid = 116

cuLaunchGridAsync = 117

cuLaunchHostFunc = 527

cuLaunchHostFunc_ptsz = 528

cuLaunchHostFunc_v2 = 831

cuLaunchHostFunc_v2_ptsz = 832

cuLaunchKernel = 307

cuLaunchKernelEx = 652

cuLaunchKernelEx_ptsz = 653

cuLaunchKernel_ptsz = 442

cuLibraryEnumerateKernels = 740

cuLibraryGetGlobal = 684

cuLibraryGetKernel = 681

cuLibraryGetKernelCount = 739

cuLibraryGetManaged = 685

cuLibraryGetModule = 682

cuLibraryGetUnifiedFunction = 700

cuLibraryLoadData = 678

cuLibraryLoadFromFile = 679

cuLibraryUnload = 680

cuLinkAddData = 363

cuLinkAddData_v2 = 382

cuLinkAddFile = 364

cuLinkAddFile_v2 = 383

cuLinkComplete = 365

cuLinkCreate = 362

cuLinkCreate_v2 = 381

cuLinkDestroy = 366

cuLogicalEndpointAddDevice = 851

cuLogicalEndpointBindAddr = 853

cuLogicalEndpointBindMem = 854

cuLogicalEndpointCreate = 850

cuLogicalEndpointDestroy = 852

cuLogicalEndpointExport = 856

cuLogicalEndpointGetLimits = 858

cuLogicalEndpointIdRelease = 849

cuLogicalEndpointIdReserve = 848

cuLogicalEndpointImport = 857

cuLogicalEndpointQuery = 859

cuLogicalEndpointUnbind = 855

cuLogsCurrent = 765

cuLogsDumpToFile = 766

cuLogsDumpToMemory = 767

cuLogsRegisterCallback = 763

cuLogsUnregisterCallback = 764

cuMemAddressFree = 548

cuMemAddressReserve = 547

cuMemAdvise = 457

cuMemAdvise_v2 = 715

cuMemAlloc = 29

cuMemAllocAsync = 598

cuMemAllocAsync_ptsz = 599

cuMemAllocFromPoolAsync = 611

cuMemAllocFromPoolAsync_ptsz = 612

cuMemAllocHost = 37

cuMemAllocHost_v2 = 294

cuMemAllocManaged = 371

cuMemAllocPitch = 31

cuMemAllocPitch_v2 = 244

cuMemAlloc_v2 = 243

cuMemBatchDecompressAsync = 761

cuMemBatchDecompressAsync_ptsz = 762

cuMemCreate = 549

cuMemDiscardAndPrefetchBatchAsync = 791

cuMemDiscardAndPrefetchBatchAsync_ptsz = 792

cuMemDiscardBatchAsync = 789

cuMemDiscardBatchAsync_ptsz = 790

cuMemExportToShareableHandle = 554

cuMemFree = 33

cuMemFreeAsync = 600

cuMemFreeAsync_ptsz = 601

cuMemFreeHost = 38

cuMemFree_v2 = 245

cuMemGetAccess = 558

cuMemGetAddressRange = 35

cuMemGetAddressRange_v2 = 246

cuMemGetAllocationGranularity = 556

cuMemGetAllocationPropertiesFromHandle = 557

cuMemGetDefaultMemPool = 801

cuMemGetHandleForAddressRange = 674

cuMemGetInfo = 27

cuMemGetInfo_v2 = 242

cuMemGetLocationInfo = 860

cuMemGetMemPool = 802

cuMemHostAlloc = 39

cuMemHostGetDevicePointer = 40

cuMemHostGetDevicePointer_v2 = 247

cuMemHostGetFlags = 42

cuMemHostRegister = 301

cuMemHostRegister_v2 = 379

cuMemHostUnregister = 302

cuMemImportFromShareableHandle = 555

cuMemMap = 551

cuMemMapArrayAsync = 584

cuMemMapArrayAsync_ptsz = 585

cuMemPoolCreate = 607

cuMemPoolDestroy = 608

cuMemPoolExportPointer = 615

cuMemPoolExportToShareableHandle = 613

cuMemPoolGetAccess = 617

cuMemPoolGetAttribute = 604

cuMemPoolImportFromShareableHandle = 614

cuMemPoolImportPointer = 616

cuMemPoolSetAccess = 605

cuMemPoolSetAttribute = 603

cuMemPoolTrimTo = 602

cuMemPrefetchAsync = 467

cuMemPrefetchAsync_ptsz = 468

cuMemPrefetchAsync_v2 = 716

cuMemPrefetchAsync_v2_ptsz = 717

cuMemPrefetchBatchAsync = 784

cuMemPrefetchBatchAsync_ptsz = 785

cuMemRangeGetAttribute = 471

cuMemRangeGetAttributes = 472

cuMemRelease = 550

cuMemRetainAllocationHandle = 565

cuMemSetAccess = 553

cuMemSetMemPool = 803

cuMemUnmap = 552

cuMemcpy = 305

cuMemcpy2D = 56

cuMemcpy2DAsync = 68

cuMemcpy2DAsync_v2 = 289

cuMemcpy2DAsync_v2_ptsz = 424

cuMemcpy2DUnaligned = 57

cuMemcpy2DUnaligned_v2 = 288

cuMemcpy2DUnaligned_v2_ptds = 406

cuMemcpy2D_v2 = 287

cuMemcpy2D_v2_ptds = 405

cuMemcpy3D = 58

cuMemcpy3DAsync = 69

cuMemcpy3DAsync_v2 = 291

cuMemcpy3DAsync_v2_ptsz = 425

cuMemcpy3DBatchAsync = 778

cuMemcpy3DBatchAsync_ptsz = 779

cuMemcpy3DBatchAsync_v2 = 798

cuMemcpy3DBatchAsync_v2_ptsz = 799

cuMemcpy3DPeer = 320

cuMemcpy3DPeerAsync = 321

cuMemcpy3DPeerAsync_ptsz = 427

cuMemcpy3DPeer_ptds = 410

cuMemcpy3DWithAttributesAsync = 839

cuMemcpy3DWithAttributesAsync_ptsz = 840

cuMemcpy3D_v2 = 290

cuMemcpy3D_v2_ptds = 407

cuMemcpyAsync = 306

cuMemcpyAsync_ptsz = 418

cuMemcpyAtoA = 55

cuMemcpyAtoA_v2 = 286

cuMemcpyAtoA_v2_ptds = 404

cuMemcpyAtoD = 51

cuMemcpyAtoD_v2 = 284

cuMemcpyAtoD_v2_ptds = 401

cuMemcpyAtoH = 54

cuMemcpyAtoHAsync = 67

cuMemcpyAtoHAsync_v2 = 283

cuMemcpyAtoHAsync_v2_ptsz = 420

cuMemcpyAtoH_v2 = 282

cuMemcpyAtoH_v2_ptds = 403

cuMemcpyBatchAsync = 776

cuMemcpyBatchAsync_ptsz = 777

cuMemcpyBatchAsync_v2 = 796

cuMemcpyBatchAsync_v2_ptsz = 797

cuMemcpyDtoA = 49

cuMemcpyDtoA_v2 = 285

cuMemcpyDtoA_v2_ptds = 400

cuMemcpyDtoD = 47

cuMemcpyDtoDAsync = 64

cuMemcpyDtoDAsync_v2 = 281

cuMemcpyDtoDAsync_v2_ptsz = 423

cuMemcpyDtoD_v2 = 280

cuMemcpyDtoD_v2_ptds = 399

cuMemcpyDtoH = 45

cuMemcpyDtoHAsync = 62

cuMemcpyDtoHAsync_v2 = 279

cuMemcpyDtoHAsync_v2_ptsz = 422

cuMemcpyDtoH_v2 = 278

cuMemcpyDtoH_v2_ptds = 398

cuMemcpyHtoA = 53

cuMemcpyHtoAAsync = 66

cuMemcpyHtoAAsync_v2 = 293

cuMemcpyHtoAAsync_v2_ptsz = 419

cuMemcpyHtoA_v2 = 292

cuMemcpyHtoA_v2_ptds = 402

cuMemcpyHtoD = 43

cuMemcpyHtoDAsync = 60

cuMemcpyHtoDAsync_v2 = 277

cuMemcpyHtoDAsync_v2_ptsz = 421

cuMemcpyHtoD_v2 = 276

cuMemcpyHtoD_v2_ptds = 397

cuMemcpyPeer = 318

cuMemcpyPeerAsync = 319

cuMemcpyPeerAsync_ptsz = 426

cuMemcpyPeer_ptds = 409

cuMemcpyWithAttributesAsync = 837

cuMemcpyWithAttributesAsync_ptsz = 838

cuMemcpy_ptds = 408

cuMemsetD16 = 73

cuMemsetD16Async = 218

cuMemsetD16Async_ptsz = 429

cuMemsetD16_v2 = 250

cuMemsetD16_v2_ptds = 412

cuMemsetD2D16 = 79

cuMemsetD2D16Async = 224

cuMemsetD2D16Async_ptsz = 432

cuMemsetD2D16_v2 = 253

cuMemsetD2D16_v2_ptds = 415

cuMemsetD2D32 = 81

cuMemsetD2D32Async = 226

cuMemsetD2D32Async_ptsz = 433

cuMemsetD2D32_v2 = 254

cuMemsetD2D32_v2_ptds = 416

cuMemsetD2D8 = 77

cuMemsetD2D8Async = 222

cuMemsetD2D8Async_ptsz = 431

cuMemsetD2D8_v2 = 252

cuMemsetD2D8_v2_ptds = 414

cuMemsetD32 = 75

cuMemsetD32Async = 220

cuMemsetD32Async_ptsz = 430

cuMemsetD32_v2 = 251

cuMemsetD32_v2_ptds = 413

cuMemsetD8 = 71

cuMemsetD8Async = 216

cuMemsetD8Async_ptsz = 428

cuMemsetD8_v2 = 249

cuMemsetD8_v2_ptds = 411

cuMipmappedArrayCreate = 347

cuMipmappedArrayDestroy = 349

cuMipmappedArrayGetLevel = 348

cuMipmappedArrayGetMemoryRequirements = 655

cuMipmappedArrayGetSparseProperties = 583

cuModuleEnumerateFunctions = 738

cuModuleGetFunction = 23

cuModuleGetFunctionCount = 737

cuModuleGetGlobal = 24

cuModuleGetGlobal_v2 = 241

cuModuleGetLoadingMode = 673

cuModuleGetSurfRef = 190

cuModuleGetTexRef = 26

cuModuleLoad = 18

cuModuleLoadData = 19

cuModuleLoadDataEx = 20

cuModuleLoadFatBinary = 21

cuModuleUnload = 22

cuMulticastAddDevice = 707

cuMulticastBindAddr = 709

cuMulticastBindAddr_v2 = 821

cuMulticastBindMem = 708

cuMulticastBindMem_v2 = 820

cuMulticastCreate = 706

cuMulticastGetGranularity = 711

cuMulticastUnbind = 710

cuNNSetAllocator = 466

cuOccupancyAvailableDynamicSMemPerBlock = 543

cuOccupancyMaxActiveBlocksPerMultiprocessor = 374

cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags = 451

cuOccupancyMaxActiveClusters = 676

cuOccupancyMaxPotentialBlockSize = 384

cuOccupancyMaxPotentialBlockSizeWithFlags = 452

cuOccupancyMaxPotentialClusterSize = 675

cuParamSetSize = 110

cuParamSetTexRef = 114

cuParamSetf = 112

cuParamSeti = 111

cuParamSetv = 113

cuPointerGetAttribute = 310

cuPointerGetAttributes = 450

cuPointerSetAttribute = 378

cuProfilerInitialize = 311

cuProfilerStart = 308

cuProfilerStop = 309

cuSignalExternalSemaphoresAsync = 490

cuSignalExternalSemaphoresAsync_ptsz = 491

cuStreamAddCallback = 346

cuStreamAddCallback_ptsz = 437

cuStreamAttachMemAsync = 377

cuStreamAttachMemAsync_ptsz = 438

cuStreamBatchMemOp = 462

cuStreamBatchMemOp_ptsz = 463

cuStreamBatchMemOp_v2 = 667

cuStreamBatchMemOp_v2_ptsz = 668

cuStreamBeginCapture = 495

cuStreamBeginCaptureToCig = 783

cuStreamBeginCaptureToCig_ptsz = 814

cuStreamBeginCaptureToGraph = 720

cuStreamBeginCaptureToGraph_ptsz = 721

cuStreamBeginCapture_ptsz = 496

cuStreamBeginCapture_v2 = 539

cuStreamBeginCapture_v2_ptsz = 540

cuStreamBeginRecaptureToGraph = 846

cuStreamBeginRecaptureToGraph_ptsz = 847

cuStreamCopyAttributes = 572

cuStreamCopyAttributes_ptsz = 573

cuStreamCreate = 124

cuStreamCreateWithPriority = 367

cuStreamDestroy = 127

cuStreamDestroy_v2 = 326

cuStreamEndCapture = 497

cuStreamEndCaptureToCig = 815

cuStreamEndCaptureToCig_ptsz = 816

cuStreamEndCapture_ptsz = 498

cuStreamGetAttribute = 574

cuStreamGetAttribute_ptsz = 575

cuStreamGetCaptureInfo = 536

cuStreamGetCaptureInfo_ptsz = 537

cuStreamGetCaptureInfo_v2 = 629

cuStreamGetCaptureInfo_v2_ptsz = 630

cuStreamGetCaptureInfo_v3 = 729

cuStreamGetCaptureInfo_v3_ptsz = 730

cuStreamGetCtx = 483

cuStreamGetCtx_ptsz = 484

cuStreamGetCtx_v2 = 759

cuStreamGetCtx_v2_ptsz = 760

cuStreamGetDevResource = 807

cuStreamGetDevResource_ptsz = 808

cuStreamGetDevice = 774

cuStreamGetDevice_ptsz = 775

cuStreamGetFlags = 369

cuStreamGetFlags_ptsz = 435

cuStreamGetGreenCtx = 752

cuStreamGetId = 693

cuStreamGetId_ptsz = 694

cuStreamGetPriority = 368

cuStreamGetPriority_ptsz = 434

cuStreamIsCapturing = 499

cuStreamIsCapturing_ptsz = 500

cuStreamQuery = 125

cuStreamQuery_ptsz = 439

cuStreamSetAttribute = 576

cuStreamSetAttribute_ptsz = 577

cuStreamSynchronize = 126

cuStreamSynchronize_ptsz = 440

cuStreamUpdateCaptureDependencies = 631

cuStreamUpdateCaptureDependencies_ptsz = 632

cuStreamUpdateCaptureDependencies_v2 = 731

cuStreamUpdateCaptureDependencies_v2_ptsz = 732

cuStreamWaitEvent = 295

cuStreamWaitEvent_ptsz = 436

cuStreamWaitValue32 = 458

cuStreamWaitValue32_ptsz = 459

cuStreamWaitValue32_v2 = 659

cuStreamWaitValue32_v2_ptsz = 660

cuStreamWaitValue64 = 473

cuStreamWaitValue64_ptsz = 474

cuStreamWaitValue64_v2 = 661

cuStreamWaitValue64_v2_ptsz = 662

cuStreamWriteValue32 = 460

cuStreamWriteValue32_ptsz = 461

cuStreamWriteValue32_v2 = 663

cuStreamWriteValue32_v2_ptsz = 664

cuStreamWriteValue64 = 475

cuStreamWriteValue64_ptsz = 476

cuStreamWriteValue64_v2 = 665

cuStreamWriteValue64_v2_ptsz = 666

cuSurfObjectCreate = 343

cuSurfObjectDestroy = 344

cuSurfObjectGetResourceDesc = 345

cuSurfRefGetArray = 196

cuSurfRefSetArray = 194

cuTensorMapEncodeIm2col = 698

cuTensorMapEncodeIm2colWide = 781

cuTensorMapEncodeTiled = 697

cuTensorMapReplaceAddress = 699

cuTexObjectCreate = 339

cuTexObjectDestroy = 340

cuTexObjectGetResourceDesc = 341

cuTexObjectGetResourceViewDesc = 361

cuTexObjectGetTextureDesc = 342

cuTexRefCreate = 92

cuTexRefDestroy = 93

cuTexRefGetAddress = 103

cuTexRefGetAddressMode = 106

cuTexRefGetAddress_v2 = 257

cuTexRefGetArray = 105

cuTexRefGetBorderColor = 456

cuTexRefGetFilterMode = 107

cuTexRefGetFlags = 109

cuTexRefGetFormat = 108

cuTexRefGetMaxAnisotropy = 359

cuTexRefGetMipmapFilterMode = 356

cuTexRefGetMipmapLevelBias = 357

cuTexRefGetMipmapLevelClamp = 358

cuTexRefGetMipmappedArray = 355

cuTexRefSetAddress = 95

cuTexRefSetAddress2D = 97

cuTexRefSetAddress2D_v2 = 256

cuTexRefSetAddress2D_v3 = 327

cuTexRefSetAddressMode = 100

cuTexRefSetAddress_v2 = 255

cuTexRefSetArray = 94

cuTexRefSetBorderColor = 455

cuTexRefSetFilterMode = 101

cuTexRefSetFlags = 102

cuTexRefSetFormat = 99

cuTexRefSetMaxAnisotropy = 354

cuTexRefSetMipmapFilterMode = 351

cuTexRefSetMipmapLevelBias = 352

cuTexRefSetMipmapLevelClamp = 353

cuTexRefSetMipmappedArray = 350

cuThreadExchangeStreamCaptureMode = 541

cuUserObjectCreate = 633

cuUserObjectRelease = 635

cuUserObjectRetain = 634

cuVDPAUCtxCreate = 187

cuVDPAUCtxCreate_v2 = 240

cuVDPAUGetDevice = 186

cuWGLGetDevice = 177

cuWaitExternalSemaphoresAsync = 492

cuWaitExternalSemaphoresAsync_ptsz = 493

cuda104 = 104

cuda131 = 131

cuda191 = 191

cuda192 = 192

cuda193 = 193

cuda195 = 195

cuda197 = 197

cuda198 = 198

cuda199 = 199

cuda200 = 200

cuda201 = 201

cuda202 = 202

cuda203 = 203

cuda204 = 204

cuda205 = 205

cuda206 = 206

cuda207 = 207

cuda208 = 208

cuda215 = 215

cuda217 = 217

cuda219 = 219

cuda221 = 221

cuda223 = 223

cuda225 = 225

cuda227 = 227

cuda228 = 228

cuda229 = 229

cuda230 = 230

cuda231 = 231

cuda232 = 232

cuda233 = 233

cuda234 = 234

cuda248 = 248

cuda25 = 25

cuda271 = 271

cuda28 = 28

cuda30 = 30

cuda315 = 315

cuda316 = 316

cuda317 = 317

cuda32 = 32

cuda34 = 34

cuda36 = 36

cuda375 = 375

cuda376 = 376

cuda41 = 41

cuda44 = 44

cuda46 = 46

cuda464 = 464

cuda465 = 465

cuda469 = 469

cuda48 = 48

cuda50 = 50

cuda52 = 52

cuda559 = 559

cuda560 = 560

cuda59 = 59

cuda61 = 61

cuda63 = 63

cuda65 = 65

cuda70 = 70

cuda72 = 72

cuda74 = 74

cuda76 = 76

cuda78 = 78

cuda786 = 786

cuda787 = 787

cuda788 = 788

cuda793 = 793

cuda794 = 794

cuda80 = 80

cuda806 = 806

cuda817 = 817

cuda818 = 818

cuda819 = 819

cuda82 = 82

cuda823 = 823

cuda824 = 824

cuda825 = 825

cuda826 = 826

cuda827 = 827

cuda828 = 828

cuda829 = 829

cuda830 = 830

cuda833 = 833

cuda834 = 834

cuda862 = 862

cuda868 = 868

cuda869 = 869

cuda870 = 870

cuda871 = 871

cuda872 = 872

cuda873 = 873

cuda874 = 874

cuda875 = 875

cuda876 = 876

cuda96 = 96

cuda98 = 98

class cupti.cupti.LogicalEndpointType(value)

Bases: IntEnum

Logical endpoint type reported in CUpti_LogicalEndpointResourceData. These values mirror the CUDA driver’s CUlogicalEndpointType.

See CUpti_LogicalEndpointType.

FORCE_INT = 2147483647

INVALID = 0

MULTICAST = 2

UNICAST = 1

class cupti.cupti.NvtxApiTraceCbid(value)

Bases: IntEnum

See CUpti_nvtx_api_trace_cbid.

FORCE_INT = 2147483647

INVALID = 0

SIZE = 50

nvtxDomainCreateA = 41

nvtxDomainCreateW = 42

nvtxDomainDestroy = 43

nvtxDomainMarkEx = 30

nvtxDomainNameCategoryA = 37

nvtxDomainNameCategoryW = 38

nvtxDomainRangeEnd = 32

nvtxDomainRangePop = 34

nvtxDomainRangePushEx = 33

nvtxDomainRangeStartEx = 31

nvtxDomainRegisterStringA = 39

nvtxDomainRegisterStringW = 40

nvtxDomainResourceCreate = 35

nvtxDomainResourceDestroy = 36

nvtxDomainSyncUserAcquireFailed = 47

nvtxDomainSyncUserAcquireStart = 46

nvtxDomainSyncUserAcquireSuccess = 48

nvtxDomainSyncUserCreate = 44

nvtxDomainSyncUserDestroy = 45

nvtxDomainSyncUserReleasing = 49

nvtxMarkA = 1

nvtxMarkEx = 3

nvtxMarkW = 2

nvtxNameCategoryA = 12

nvtxNameCategoryW = 13

nvtxNameCuContextA = 18

nvtxNameCuContextW = 19

nvtxNameCuDeviceA = 16

nvtxNameCuDeviceW = 17

nvtxNameCuEventA = 22

nvtxNameCuEventW = 23

nvtxNameCuStreamA = 20

nvtxNameCuStreamW = 21

nvtxNameCudaDeviceA = 24

nvtxNameCudaDeviceW = 25

nvtxNameCudaEventA = 28

nvtxNameCudaEventW = 29

nvtxNameCudaStreamA = 26

nvtxNameCudaStreamW = 27

nvtxNameOsThreadA = 14

nvtxNameOsThreadW = 15

nvtxRangeEnd = 7

nvtxRangePop = 11

nvtxRangePushA = 8

nvtxRangePushEx = 10

nvtxRangePushW = 9

nvtxRangeStartA = 4

nvtxRangeStartEx = 6

nvtxRangeStartW = 5

class cupti.cupti.RuntimeApiTraceCbid(value)

Bases: IntEnum

See CUpti_runtime_api_trace_cbid.

FORCE_INT = 2147483647

INVALID = 0

SIZE = 567

cuda470_v12060 = 470

cuda471_v12060 = 471

cuda472_v12060 = 472

cuda473_v12060 = 473

cuda474_v12060 = 474

cuda475_v12060 = 475

cuda476_v12060 = 476

cuda477_v12060 = 477

cuda478_v12060 = 478

cuda479_v12060 = 479

cuda531 = 531

cuda531_v13010 = 531

cuda532 = 532

cuda532_v13010 = 532

cuda545 = 545

cuda545_v13010 = 545

cuda546 = 546

cuda546_v13010 = 546

cuda547 = 547

cuda547_v13010 = 547

cuda553 = 553

cuda553_v13020 = 553

cuda559 = 559

cuda560 = 560

cuda561 = 561

cuda564 = 564

cuda566 = 566

cudaArrayGetInfo_v4010 = 181

cudaArrayGetMemoryRequirements_v11060 = 428

cudaArrayGetPlane_v11020 = 381

cudaArrayGetSparseProperties_v11010 = 359

cudaBindSurfaceToArray_v3020 = 61

cudaBindTexture2D_v3020 = 56

cudaBindTextureToArray_v3020 = 57

cudaBindTextureToMipmappedArray_v5000 = 195

cudaBindTexture_v3020 = 55

cudaChooseDevice_v3020 = 5

cudaConfigureCall_v3020 = 8

cudaCreateChannelDesc_v3020 = 7

cudaCreateSurfaceObject_v5000 = 189

cudaCreateTextureObject_v2_v11080 = 434

cudaCreateTextureObject_v5000 = 185

cudaCtxResetPersistingL2Cache_v11000 = 337

cudaD3D10GetDevice_v3020 = 88

cudaD3D10GetDevices_v3020 = 89

cudaD3D10GetDirect3DDevice_v3020 = 149

cudaD3D10MapResources_v3020 = 94

cudaD3D10RegisterResource_v3020 = 92

cudaD3D10ResourceGetMappedArray_v3020 = 98

cudaD3D10ResourceGetMappedPitch_v3020 = 101

cudaD3D10ResourceGetMappedPointer_v3020 = 99

cudaD3D10ResourceGetMappedSize_v3020 = 100

cudaD3D10ResourceGetSurfaceDimensions_v3020 = 97

cudaD3D10ResourceSetMapFlags_v3020 = 96

cudaD3D10SetDirect3DDevice_v3020 = 90

cudaD3D10UnmapResources_v3020 = 95

cudaD3D10UnregisterResource_v3020 = 93

cudaD3D11GetDevice_v3020 = 84

cudaD3D11GetDevices_v3020 = 85

cudaD3D11GetDirect3DDevice_v3020 = 148

cudaD3D11SetDirect3DDevice_v3020 = 86

cudaD3D9Begin_v3020 = 117

cudaD3D9End_v3020 = 118

cudaD3D9GetDevice_v3020 = 102

cudaD3D9GetDevices_v3020 = 103

cudaD3D9GetDirect3DDevice_v3020 = 105

cudaD3D9MapResources_v3020 = 109

cudaD3D9MapVertexBuffer_v3020 = 121

cudaD3D9RegisterResource_v3020 = 107

cudaD3D9RegisterVertexBuffer_v3020 = 119

cudaD3D9ResourceGetMappedArray_v3020 = 113

cudaD3D9ResourceGetMappedPitch_v3020 = 116

cudaD3D9ResourceGetMappedPointer_v3020 = 114

cudaD3D9ResourceGetMappedSize_v3020 = 115

cudaD3D9ResourceGetSurfaceDimensions_v3020 = 112

cudaD3D9ResourceSetMapFlags_v3020 = 111

cudaD3D9SetDirect3DDevice_v3020 = 104

cudaD3D9UnmapResources_v3020 = 110

cudaD3D9UnmapVertexBuffer_v3020 = 122

cudaD3D9UnregisterResource_v3020 = 108

cudaD3D9UnregisterVertexBuffer_v3020 = 120

cudaDestroyExternalMemory_v10000 = 277

cudaDestroyExternalSemaphore_v10000 = 283

cudaDestroySurfaceObject_v5000 = 190

cudaDestroyTextureObject_v5000 = 186

cudaDevResourceGenerateDesc_v13010 = 525

cudaDevSmResourceSplitByCount_v13010 = 524

cudaDevSmResourceSplit_v13010 = 549

cudaDeviceCanAccessPeer_v4000 = 154

cudaDeviceDisablePeerAccess_v4000 = 156

cudaDeviceEnablePeerAccess_v4000 = 155

cudaDeviceFlushGPUDirectRDMAWrites_v11030 = 405

cudaDeviceGetAttribute_v5000 = 200

cudaDeviceGetByPCIBusId_v4010 = 173

cudaDeviceGetCacheConfig_v3020 = 168

cudaDeviceGetDefaultMemPool_v11020 = 372

cudaDeviceGetDevResource_v13010 = 523

cudaDeviceGetExecutionCtx_v13010 = 548

cudaDeviceGetGraphMemAttribute_v11040 = 424

cudaDeviceGetHostAtomicCapabilities_v13000 = 521

cudaDeviceGetLimit_v3020 = 166

cudaDeviceGetMemPool_v11020 = 386

cudaDeviceGetNvSciSyncAttributes_v10020 = 328

cudaDeviceGetP2PAtomicCapabilities_v13000 = 522

cudaDeviceGetP2PAttribute_v8000 = 255

cudaDeviceGetPCIBusId_v4010 = 174

cudaDeviceGetSharedMemConfig_v4020 = 183

cudaDeviceGetStreamPriorityRange_v5050 = 205

cudaDeviceGetTexture1DLinearMaxWidth_v11010 = 347

cudaDeviceGraphMemTrim_v11040 = 423

cudaDeviceRegisterAsyncNotification_v12040 = 465

cudaDeviceReset_v3020 = 164

cudaDeviceSetCacheConfig_v3020 = 169

cudaDeviceSetGraphMemAttribute_v11040 = 425

cudaDeviceSetLimit_v3020 = 167

cudaDeviceSetMemPool_v11020 = 385

cudaDeviceSetSharedMemConfig_v4020 = 184

cudaDeviceSynchronize_v3020 = 165

cudaDeviceUnregisterAsyncNotification_v12040 = 466

cudaDriverGetVersion_v3020 = 1

cudaEGLStreamConsumerAcquireFrame_v7000 = 259

cudaEGLStreamConsumerConnectWithFlags_v7000 = 268

cudaEGLStreamConsumerConnect_v7000 = 257

cudaEGLStreamConsumerDisconnect_v7000 = 258

cudaEGLStreamConsumerReleaseFrame_v7000 = 260

cudaEGLStreamProducerConnect_v7000 = 261

cudaEGLStreamProducerDisconnect_v7000 = 262

cudaEGLStreamProducerPresentFrame_v7000 = 263

cudaEGLStreamProducerReturnFrame_v7000 = 264

cudaEventCreateFromEGLSync_v9000 = 271

cudaEventCreateWithFlags_v3020 = 134

cudaEventCreate_v3020 = 133

cudaEventDestroy_v3020 = 136

cudaEventElapsedTime_v12080 = 486

cudaEventElapsedTime_v2_v12080 = 486

cudaEventElapsedTime_v3020 = 139

cudaEventQuery_v3020 = 138

cudaEventRecordWithFlags_ptsz_v11010 = 371

cudaEventRecordWithFlags_v11010 = 370

cudaEventRecord_ptsz_v7000 = 242

cudaEventRecord_v3020 = 135

cudaEventSynchronize_v3020 = 137

cudaExecutionCtxDestroy_v13010 = 527

cudaExecutionCtxGetDevResource_v13010 = 528

cudaExecutionCtxGetDevice_v13010 = 529

cudaExecutionCtxGetId_v13010 = 530

cudaExecutionCtxRecordEvent_v13010 = 543

cudaExecutionCtxStreamCreate_v13010 = 533

cudaExecutionCtxSynchronize_v13010 = 534

cudaExecutionCtxWaitEvent_v13010 = 544

cudaExternalMemoryGetMappedBuffer_v10000 = 275

cudaExternalMemoryGetMappedMipmappedArray_v10000 = 276

cudaFreeArray_v3020 = 24

cudaFreeAsync_ptsz_v11020 = 376

cudaFreeAsync_v11020 = 375

cudaFreeHost_v3020 = 26

cudaFreeMipmappedArray_v5000 = 194

cudaFree_v3020 = 22

cudaFuncGetAttributes_v3020 = 15

cudaFuncGetName_v12030 = 451

cudaFuncGetParamCount_v13020 = 552

cudaFuncGetParamInfo_v12040 = 467

cudaFuncSetAttribute_v9000 = 273

cudaFuncSetCacheConfig_v3020 = 14

cudaFuncSetSharedMemConfig_v4020 = 182

cudaGLGetDevices_v4010 = 175

cudaGLMapBufferObjectAsync_v3020 = 69

cudaGLMapBufferObject_v3020 = 65

cudaGLRegisterBufferObject_v3020 = 64

cudaGLSetBufferObjectMapFlags_v3020 = 68

cudaGLSetGLDevice_v3020 = 63

cudaGLUnmapBufferObjectAsync_v3020 = 70

cudaGLUnmapBufferObject_v3020 = 66

cudaGLUnregisterBufferObject_v3020 = 67

cudaGetChannelDesc_v3020 = 6

cudaGetDeviceCount_v3020 = 3

cudaGetDeviceFlags_v7000 = 212

cudaGetDeviceProperties_v12000 = 440

cudaGetDeviceProperties_v2_v12000 = 440

cudaGetDeviceProperties_v3020 = 4

cudaGetDevice_v3020 = 17

cudaGetDriverEntryPointByVersion_ptsz_v12050 = 469

cudaGetDriverEntryPointByVersion_v12050 = 468

cudaGetDriverEntryPoint_ptsz_v11030 = 407

cudaGetDriverEntryPoint_v11030 = 406

cudaGetErrorName_v6050 = 209

cudaGetErrorString_v3020 = 12

cudaGetExportTable_v13000 = 493

cudaGetFuncBySymbol_v11000 = 336

cudaGetKernel_v12000 = 439

cudaGetLastError_v3020 = 10

cudaGetMipmappedArrayLevel_v5000 = 193

cudaGetSurfaceObjectResourceDesc_v5000 = 191

cudaGetSurfaceReference_v3020 = 62

cudaGetSymbolAddress_v3020 = 53

cudaGetSymbolSize_v3020 = 54

cudaGetTextureAlignmentOffset_v3020 = 59

cudaGetTextureObjectResourceDesc_v5000 = 187

cudaGetTextureObjectResourceViewDesc_v5000 = 199

cudaGetTextureObjectTextureDesc_v2_v11080 = 435

cudaGetTextureObjectTextureDesc_v5000 = 188

cudaGetTextureReference_v3020 = 60

cudaGraphAddChildGraphNode_v10000 = 298

cudaGraphAddDependencies_v10000 = 307

cudaGraphAddDependencies_v12030 = 458

cudaGraphAddDependencies_v2_v12030 = 458

cudaGraphAddEmptyNode_v10000 = 300

cudaGraphAddEventRecordNode_v11010 = 362

cudaGraphAddEventWaitNode_v11010 = 365

cudaGraphAddExternalSemaphoresSignalNode_v11020 = 397

cudaGraphAddExternalSemaphoresWaitNode_v11020 = 400

cudaGraphAddHostNode_v10000 = 296

cudaGraphAddKernelNode_v10000 = 289

cudaGraphAddMemAllocNode_v11040 = 419

cudaGraphAddMemFreeNode_v11040 = 421

cudaGraphAddMemcpyNode1D_v11010 = 352

cudaGraphAddMemcpyNodeFromSymbol_v11010 = 351

cudaGraphAddMemcpyNodeToSymbol_v11010 = 350

cudaGraphAddMemcpyNode_v10000 = 290

cudaGraphAddMemsetNode_v10000 = 293

cudaGraphAddNode_v12020 = 445

cudaGraphAddNode_v12030 = 460

cudaGraphAddNode_v2_v12030 = 460

cudaGraphChildGraphNodeGetGraph_v10000 = 299

cudaGraphClone_v10000 = 301

cudaGraphConditionalHandleCreate_v12030 = 454

cudaGraphConditionalHandleCreate_v2_v13010 = 535

cudaGraphCreate_v10000 = 286

cudaGraphDebugDotPrint_v11030 = 408

cudaGraphDestroyNode_v10000 = 309

cudaGraphDestroy_v10000 = 314

cudaGraphEventRecordNodeGetEvent_v11010 = 363

cudaGraphEventRecordNodeSetEvent_v11010 = 364

cudaGraphEventWaitNodeGetEvent_v11010 = 366

cudaGraphEventWaitNodeSetEvent_v11010 = 367

cudaGraphExecChildGraphNodeSetParams_v11010 = 361

cudaGraphExecDestroy_v10000 = 313

cudaGraphExecEventRecordNodeSetEvent_v11010 = 368

cudaGraphExecEventWaitNodeSetEvent_v11010 = 369

cudaGraphExecExternalSemaphoresSignalNodeSetParams_v11020 = 403

cudaGraphExecExternalSemaphoresWaitNodeSetParams_v11020 = 404

cudaGraphExecGetFlags_v12000 = 438

cudaGraphExecGetId_v13010 = 542

cudaGraphExecHostNodeSetParams_v10020 = 334

cudaGraphExecKernelNodeSetParams_v10010 = 326

cudaGraphExecMemcpyNodeSetParams1D_v11010 = 358

cudaGraphExecMemcpyNodeSetParamsFromSymbol_v11010 = 357

cudaGraphExecMemcpyNodeSetParamsToSymbol_v11010 = 356

cudaGraphExecMemcpyNodeSetParams_v10020 = 332

cudaGraphExecMemsetNodeSetParams_v10020 = 333

cudaGraphExecNodeSetParams_v12020 = 447

cudaGraphExecUpdate_v10020 = 335

cudaGraphExternalSemaphoresSignalNodeGetParams_v11020 = 398

cudaGraphExternalSemaphoresSignalNodeSetParams_v11020 = 399

cudaGraphExternalSemaphoresWaitNodeGetParams_v11020 = 401

cudaGraphExternalSemaphoresWaitNodeSetParams_v11020 = 402

cudaGraphGetEdges_v10000 = 323

cudaGraphGetEdges_v12030 = 455

cudaGraphGetEdges_v2_v12030 = 455

cudaGraphGetId_v13010 = 541

cudaGraphGetNodes_v10000 = 322

cudaGraphGetRootNodes_v10000 = 304

cudaGraphHostNodeGetParams_v10000 = 297

cudaGraphHostNodeSetParams_v10000 = 321

cudaGraphInstantiateWithFlags_v11040 = 418

cudaGraphInstantiateWithParams_ptsz_v12000 = 437

cudaGraphInstantiateWithParams_v12000 = 436

cudaGraphInstantiate_v10000 = 310

cudaGraphInstantiate_v12000 = 443

cudaGraphKernelNodeCopyAttributes_v11000 = 338

cudaGraphKernelNodeGetAttribute_v11000 = 339

cudaGraphKernelNodeGetParams_v10000 = 287

cudaGraphKernelNodeSetAttribute_v11000 = 340

cudaGraphKernelNodeSetParams_v10000 = 288

cudaGraphLaunch_ptsz_v10000 = 312

cudaGraphLaunch_v10000 = 311

cudaGraphMemAllocNodeGetParams_v11040 = 420

cudaGraphMemFreeNodeGetParams_v11040 = 422

cudaGraphMemcpyNodeGetParams_v10000 = 291

cudaGraphMemcpyNodeSetParams1D_v11010 = 355

cudaGraphMemcpyNodeSetParamsFromSymbol_v11010 = 354

cudaGraphMemcpyNodeSetParamsToSymbol_v11010 = 353

cudaGraphMemcpyNodeSetParams_v10000 = 292

cudaGraphMemsetNodeGetParams_v10000 = 294

cudaGraphMemsetNodeSetParams_v10000 = 295

cudaGraphNodeFindInClone_v10000 = 302

cudaGraphNodeGetContainingGraph_v13010 = 538

cudaGraphNodeGetDependencies_v10000 = 305

cudaGraphNodeGetDependencies_v12030 = 456

cudaGraphNodeGetDependencies_v2_v12030 = 456

cudaGraphNodeGetDependentNodes_v10000 = 306

cudaGraphNodeGetDependentNodes_v12030 = 457

cudaGraphNodeGetDependentNodes_v2_v12030 = 457

cudaGraphNodeGetEnabled_v11060 = 427

cudaGraphNodeGetLocalId_v13010 = 539

cudaGraphNodeGetParams_v13020 = 558

cudaGraphNodeGetToolsId_v13010 = 540

cudaGraphNodeGetType_v10000 = 303

cudaGraphNodeSetEnabled_v11060 = 426

cudaGraphNodeSetParams_v12020 = 446

cudaGraphReleaseUserObject_v11030 = 417

cudaGraphRemoveDependencies_v10000 = 308

cudaGraphRemoveDependencies_v12030 = 459

cudaGraphRemoveDependencies_v2_v12030 = 459

cudaGraphRetainUserObject_v11030 = 416

cudaGraphUpload_ptsz_v10000 = 349

cudaGraphUpload_v10000 = 348

cudaGraphicsD3D10RegisterResource_v3020 = 91

cudaGraphicsD3D11RegisterResource_v3020 = 87

cudaGraphicsD3D9RegisterResource_v3020 = 106

cudaGraphicsEGLRegisterImage_v7000 = 256

cudaGraphicsGLRegisterBuffer_v3020 = 73

cudaGraphicsGLRegisterImage_v3020 = 72

cudaGraphicsMapResources_v3020 = 76

cudaGraphicsResourceGetMappedEglFrame_v7000 = 265

cudaGraphicsResourceGetMappedMipmappedArray_v5000 = 196

cudaGraphicsResourceGetMappedPointer_v3020 = 78

cudaGraphicsResourceSetMapFlags_v3020 = 75

cudaGraphicsSubResourceGetMappedArray_v3020 = 79

cudaGraphicsUnmapResources_v3020 = 77

cudaGraphicsUnregisterResource_v3020 = 74

cudaGraphicsVDPAURegisterOutputSurface_v3020 = 83

cudaGraphicsVDPAURegisterVideoSurface_v3020 = 82

cudaGreenCtxCreate_v13010 = 526

cudaHostAlloc_v3020 = 27

cudaHostGetDevicePointer_v3020 = 28

cudaHostGetFlags_v3020 = 29

cudaHostRegister_v4000 = 152

cudaHostUnregister_v4000 = 153

cudaImportExternalMemory_v10000 = 274

cudaImportExternalSemaphore_v10000 = 278

cudaInitDevice_v12000 = 444

cudaIpcCloseMemHandle_v4010 = 180

cudaIpcGetEventHandle_v4010 = 176

cudaIpcGetMemHandle_v4010 = 178

cudaIpcOpenEventHandle_v4010 = 177

cudaIpcOpenMemHandle_v4010 = 179

cudaKernelSetAttributeForDevice_v12060 = 479

cudaLaunchCooperativeKernelMultiDevice_v9000 = 272

cudaLaunchCooperativeKernel_ptsz_v9000 = 270

cudaLaunchCooperativeKernel_v9000 = 269

cudaLaunchHostFunc_ptsz_v10000 = 285

cudaLaunchHostFunc_v10000 = 284

cudaLaunchHostFunc_v2_ptsz_v13020 = 551

cudaLaunchHostFunc_v2_v13020 = 550

cudaLaunchKernelExC_ptsz_v11060 = 431

cudaLaunchKernelExC_v11060 = 430

cudaLaunchKernel_ptsz_v7000 = 214

cudaLaunchKernel_v7000 = 211

cudaLaunch_ptsz_v7000 = 213

cudaLaunch_v3020 = 13

cudaLibraryEnumerateKernels_v12060 = 478

cudaLibraryGetGlobal_v12060 = 474

cudaLibraryGetKernelCount_v12060 = 477

cudaLibraryGetKernel_v12060 = 473

cudaLibraryGetManaged_v12060 = 475

cudaLibraryGetUnifiedFunction_v12060 = 476

cudaLibraryLoadData_v12060 = 470

cudaLibraryLoadFromFile_v12060 = 471

cudaLibraryUnload_v12060 = 472

cudaLogsCurrent_v13000 = 515

cudaLogsDumpToFile_v13000 = 516

cudaLogsDumpToMemory_v13000 = 517

cudaLogsRegisterCallback_v13000 = 513

cudaLogsUnregisterCallback_v13000 = 514

cudaMalloc3DArray_v3020 = 141

cudaMalloc3D_v3020 = 140

cudaMallocArray_v3020 = 23

cudaMallocAsync_ptsz_v11020 = 374

cudaMallocAsync_v11020 = 373

cudaMallocFromPoolAsync_ptsz_v11020 = 392

cudaMallocFromPoolAsync_v11020 = 391

cudaMallocHost_v3020 = 25

cudaMallocManaged_v6000 = 206

cudaMallocMipmappedArray_v5000 = 192

cudaMallocPitch_v3020 = 21

cudaMalloc_v3020 = 20

cudaMemAdvise_v12020 = 448

cudaMemAdvise_v2_v12020 = 448

cudaMemAdvise_v8000 = 254

cudaMemDiscardAndPrefetchBatchAsync_ptsz_v13000 = 492

cudaMemDiscardAndPrefetchBatchAsync_v13000 = 491

cudaMemDiscardBatchAsync_ptsz_v13000 = 490

cudaMemDiscardBatchAsync_v13000 = 489

cudaMemGetDefaultMemPool_v13000 = 518

cudaMemGetInfo_v3020 = 30

cudaMemGetLocationInfo_v13040 = 565

cudaMemGetMemPool_v13000 = 519

cudaMemPoolCreate_v11020 = 383

cudaMemPoolDestroy_v11020 = 384

cudaMemPoolExportPointer_v11020 = 389

cudaMemPoolExportToShareableHandle_v11020 = 387

cudaMemPoolGetAccess_v11020 = 382

cudaMemPoolGetAttribute_v11020 = 379

cudaMemPoolImportFromShareableHandle_v11020 = 388

cudaMemPoolImportPointer_v11020 = 390

cudaMemPoolSetAccess_v11020 = 380

cudaMemPoolSetAttribute_v11020 = 378

cudaMemPoolTrimTo_v11020 = 377

cudaMemPrefetchAsync_ptsz_v12020 = 450

cudaMemPrefetchAsync_ptsz_v8000 = 253

cudaMemPrefetchAsync_v12020 = 449

cudaMemPrefetchAsync_v2_ptsz_v12020 = 450

cudaMemPrefetchAsync_v2_v12020 = 449

cudaMemPrefetchAsync_v8000 = 252

cudaMemPrefetchBatchAsync_ptsz_v13000 = 488

cudaMemPrefetchBatchAsync_v13000 = 487

cudaMemRangeGetAttribute_v8000 = 266

cudaMemRangeGetAttributes_v8000 = 267

cudaMemSetMemPool_v13000 = 520

cudaMemcpy2DArrayToArray_ptds_v7000 = 222

cudaMemcpy2DArrayToArray_v3020 = 38

cudaMemcpy2DAsync_ptsz_v7000 = 228

cudaMemcpy2DAsync_v3020 = 44

cudaMemcpy2DFromArrayAsync_ptsz_v7000 = 230

cudaMemcpy2DFromArrayAsync_v3020 = 46

cudaMemcpy2DFromArray_ptds_v7000 = 220

cudaMemcpy2DFromArray_v3020 = 36

cudaMemcpy2DToArrayAsync_ptsz_v7000 = 229

cudaMemcpy2DToArrayAsync_v3020 = 45

cudaMemcpy2DToArray_ptds_v7000 = 218

cudaMemcpy2DToArray_v3020 = 34

cudaMemcpy2D_ptds_v7000 = 216

cudaMemcpy2D_v3020 = 32

cudaMemcpy3DAsync_ptsz_v7000 = 246

cudaMemcpy3DAsync_v3020 = 145

cudaMemcpy3DBatchAsync_ptsz_v12080 = 485

cudaMemcpy3DBatchAsync_ptsz_v13000 = 512

cudaMemcpy3DBatchAsync_v12080 = 484

cudaMemcpy3DBatchAsync_v13000 = 511

cudaMemcpy3DPeerAsync_ptsz_v7000 = 250

cudaMemcpy3DPeerAsync_v4000 = 163

cudaMemcpy3DPeer_ptds_v7000 = 249

cudaMemcpy3DPeer_v4000 = 162

cudaMemcpy3DWithAttributesAsync_ptsz_v13020 = 557

cudaMemcpy3DWithAttributesAsync_v13020 = 556

cudaMemcpy3D_ptds_v7000 = 245

cudaMemcpy3D_v3020 = 144

cudaMemcpyArrayToArray_ptds_v7000 = 221

cudaMemcpyArrayToArray_v3020 = 37

cudaMemcpyAsync_ptsz_v7000 = 225

cudaMemcpyAsync_v3020 = 41

cudaMemcpyBatchAsync_ptsz_v12080 = 483

cudaMemcpyBatchAsync_ptsz_v13000 = 510

cudaMemcpyBatchAsync_v12080 = 482

cudaMemcpyBatchAsync_v13000 = 509

cudaMemcpyFromArrayAsync_ptsz_v7000 = 227

cudaMemcpyFromArrayAsync_v3020 = 43

cudaMemcpyFromArray_ptds_v7000 = 219

cudaMemcpyFromArray_v3020 = 35

cudaMemcpyFromSymbolAsync_ptsz_v7000 = 232

cudaMemcpyFromSymbolAsync_v3020 = 48

cudaMemcpyFromSymbol_ptds_v7000 = 224

cudaMemcpyFromSymbol_v3020 = 40

cudaMemcpyPeerAsync_v4000 = 161

cudaMemcpyPeer_v4000 = 160

cudaMemcpyToArrayAsync_ptsz_v7000 = 226

cudaMemcpyToArrayAsync_v3020 = 42

cudaMemcpyToArray_ptds_v7000 = 217

cudaMemcpyToArray_v3020 = 33

cudaMemcpyToSymbolAsync_ptsz_v7000 = 231

cudaMemcpyToSymbolAsync_v3020 = 47

cudaMemcpyToSymbol_ptds_v7000 = 223

cudaMemcpyToSymbol_v3020 = 39

cudaMemcpyWithAttributesAsync_ptsz_v13020 = 555

cudaMemcpyWithAttributesAsync_v13020 = 554

cudaMemcpy_ptds_v7000 = 215

cudaMemcpy_v3020 = 31

cudaMemset2DAsync_ptsz_v7000 = 236

cudaMemset2DAsync_v3020 = 52

cudaMemset2D_ptds_v7000 = 234

cudaMemset2D_v3020 = 50

cudaMemset3DAsync_ptsz_v7000 = 244

cudaMemset3DAsync_v3020 = 143

cudaMemset3D_ptds_v7000 = 243

cudaMemset3D_v3020 = 142

cudaMemsetAsync_ptsz_v7000 = 235

cudaMemsetAsync_v3020 = 51

cudaMemset_ptds_v7000 = 233

cudaMemset_v3020 = 49

cudaMipmappedArrayGetMemoryRequirements_v11060 = 429

cudaMipmappedArrayGetSparseProperties_v11010 = 360

cudaOccupancyAvailableDynamicSMemPerBlock_v10200 = 329

cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags_v7000 = 251

cudaOccupancyMaxActiveBlocksPerMultiprocessor_v6000 = 207

cudaOccupancyMaxActiveBlocksPerMultiprocessor_v6050 = 210

cudaOccupancyMaxActiveClusters_v11070 = 433

cudaOccupancyMaxPotentialClusterSize_v11070 = 432

cudaPeekAtLastError_v3020 = 11

cudaPeerGetDevicePointer_v4000 = 159

cudaPeerRegister_v4000 = 157

cudaPeerUnregister_v4000 = 158

cudaPointerGetAttributes_v4000 = 151

cudaProfilerInitialize_v4000 = 170

cudaProfilerStart_v4000 = 171

cudaProfilerStop_v4000 = 172

cudaRuntimeGetVersion_v3020 = 2

cudaSetDeviceFlags_v3020 = 19

cudaSetDevice_v3020 = 16

cudaSetDoubleForDevice_v3020 = 124

cudaSetDoubleForHost_v3020 = 125

cudaSetValidDevices_v3020 = 18

cudaSetupArgument_v3020 = 9

cudaSignalExternalSemaphoresAsync_ptsz_v10000 = 280

cudaSignalExternalSemaphoresAsync_ptsz_v11020 = 394

cudaSignalExternalSemaphoresAsync_v10000 = 279

cudaSignalExternalSemaphoresAsync_v11020 = 393

cudaSignalExternalSemaphoresAsync_v2_ptsz_v11020 = 394

cudaSignalExternalSemaphoresAsync_v2_v11020 = 393

cudaStreamAddCallback_ptsz_v7000 = 248

cudaStreamAddCallback_v5000 = 197

cudaStreamAttachMemAsync_ptsz_v7000 = 241

cudaStreamAttachMemAsync_v6000 = 208

cudaStreamBeginCaptureToGraph_ptsz_v12030 = 453

cudaStreamBeginCaptureToGraph_v12030 = 452

cudaStreamBeginCapture_ptsz_v10000 = 316

cudaStreamBeginCapture_v10000 = 315

cudaStreamBeginRecaptureToGraph_ptsz_v13030 = 563

cudaStreamBeginRecaptureToGraph_v13030 = 562

cudaStreamCopyAttributes_ptsz_v11000 = 342

cudaStreamCopyAttributes_v11000 = 341

cudaStreamCreateWithFlags_v5000 = 198

cudaStreamCreateWithPriority_v5050 = 202

cudaStreamCreate_v3020 = 129

cudaStreamDestroy_v3020 = 130

cudaStreamDestroy_v5050 = 201

cudaStreamEndCapture_ptsz_v10000 = 320

cudaStreamEndCapture_v10000 = 319

cudaStreamGetAttribute_ptsz_v11000 = 344

cudaStreamGetAttribute_v11000 = 343

cudaStreamGetCaptureInfo_ptsz_v10010 = 325

cudaStreamGetCaptureInfo_ptsz_v12030 = 462

cudaStreamGetCaptureInfo_v10010 = 324

cudaStreamGetCaptureInfo_v12030 = 461

cudaStreamGetCaptureInfo_v2_ptsz_v11030 = 410

cudaStreamGetCaptureInfo_v2_v11030 = 409

cudaStreamGetCaptureInfo_v3_ptsz_v12030 = 462

cudaStreamGetCaptureInfo_v3_v12030 = 461

cudaStreamGetDevResource_ptsz_v13010 = 537

cudaStreamGetDevResource_v13010 = 536

cudaStreamGetDevice_ptsz_v12080 = 481

cudaStreamGetDevice_v12080 = 480

cudaStreamGetFlags_ptsz_v7000 = 238

cudaStreamGetFlags_v5050 = 204

cudaStreamGetId_ptsz_v12000 = 442

cudaStreamGetId_v12000 = 441

cudaStreamGetPriority_ptsz_v7000 = 237

cudaStreamGetPriority_v5050 = 203

cudaStreamIsCapturing_ptsz_v10000 = 318

cudaStreamIsCapturing_v10000 = 317

cudaStreamQuery_ptsz_v7000 = 240

cudaStreamQuery_v3020 = 132

cudaStreamSetAttribute_ptsz_v11000 = 346

cudaStreamSetAttribute_v11000 = 345

cudaStreamSetFlags_ptsz_v10200 = 331

cudaStreamSetFlags_v10200 = 330

cudaStreamSynchronize_ptsz_v7000 = 239

cudaStreamSynchronize_v3020 = 131

cudaStreamUpdateCaptureDependencies_ptsz_v11030 = 412

cudaStreamUpdateCaptureDependencies_ptsz_v12030 = 464

cudaStreamUpdateCaptureDependencies_v11030 = 411

cudaStreamUpdateCaptureDependencies_v12030 = 463

cudaStreamUpdateCaptureDependencies_v2_ptsz_v12030 = 464

cudaStreamUpdateCaptureDependencies_v2_v12030 = 463

cudaStreamWaitEvent_ptsz_v7000 = 247

cudaStreamWaitEvent_v3020 = 147

cudaThreadExchangeStreamCaptureMode_v10010 = 327

cudaThreadExit_v3020 = 123

cudaThreadGetCacheConfig_v3020 = 150

cudaThreadGetLimit_v3020 = 127

cudaThreadSetCacheConfig_v3020 = 146

cudaThreadSetLimit_v3020 = 128

cudaThreadSynchronize_v3020 = 126

cudaUnbindTexture_v3020 = 58

cudaUserObjectCreate_v11030 = 413

cudaUserObjectRelease_v11030 = 415

cudaUserObjectRetain_v11030 = 414

cudaVDPAUGetDevice_v3020 = 80

cudaVDPAUSetVDPAUDevice_v3020 = 81

cudaWGLGetDevice_v3020 = 71

cudaWaitExternalSemaphoresAsync_ptsz_v10000 = 282

cudaWaitExternalSemaphoresAsync_ptsz_v11020 = 396

cudaWaitExternalSemaphoresAsync_v10000 = 281

cudaWaitExternalSemaphoresAsync_v11020 = 395

cudaWaitExternalSemaphoresAsync_v2_ptsz_v11020 = 396

cudaWaitExternalSemaphoresAsync_v2_v11020 = 395

Classes

class cupti.cupti.CallbackData

Bases: object

Empty-initialize an instance of CUpti_CallbackData.

See also

CUpti_CallbackData

callback_site

Point in the runtime or driver function from where the callback was issued.

Type:

int

context

Driver context current to the thread, or null if no context is current. This value can change from the entry to exit callback of a runtime API function if the runtime initializes a context.

Type:

int

context_uid

Unique ID for the CUDA context associated with the thread. The UIDs are assigned sequentially as contexts are created and are unique within a process.

Type:

int

correlation_data

Pointer to data shared between the entry and exit callbacks of a given runtime or drive API function invocation. This field can be used to pass 64-bit values from the entry callback to the corresponding exit callback.

Type:

int

correlation_id

The activity record correlation ID for this callback. For a driver domain callback (i.e. domain CUPTI_CB_DOMAIN_DRIVER_API) this ID will equal the correlation ID in the CUpti_ActivityAPI record corresponding to the CUDA driver function call. For a runtime domain callback (i.e. domain CUPTI_CB_DOMAIN_RUNTIME_API) this ID will equal the correlation ID in the CUpti_ActivityAPI record corresponding to the CUDA runtime function call. Within the callback, this ID can be recorded to correlate user data with the activity record. This field is new in 4.1.

Type:

int

static from_buffer(buffer)

Create an CallbackData instance with the memory from the given buffer.

static from_data(data)

Create an CallbackData instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype callback_data_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an CallbackData instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

function_name

Name of the runtime or driver API function which issued the callback. This string is a global constant and so may be accessed outside of the callback.

Type:

str

function_params

int: Pointer to the arguments passed to the runtime or driver API call.

To access the contents, cast this pointer using the appropriate Function Parameter class and its from_ptr method. See Function Parameter Classes for CUDA and NVTX APIs for available classes and usage examples.

function_return_value

Pointer to the return value of the runtime or driver API call. This field is only valid within the exit::CUPTI_API_EXIT callback. For a runtime API functionReturnValue points to a cudaError_t. For a driver API functionReturnValue points to a CUresult.

Type:

int

ptr

Get the pointer address to the data as Python int.

symbol_name

Name of the symbol operated on by the runtime or driver API function which issued the callback. This entry is valid only for driver and runtime launch callbacks, where it returns the name of the kernel.

Type:

str

class cupti.cupti.GraphData

Bases: object

Empty-initialize an instance of CUpti_GraphData.

See also

CUpti_GraphData

dependency

The dependent graph node

Type:

int

static from_buffer(buffer)

Create an GraphData instance with the memory from the given buffer.

static from_data(data)

Create an GraphData instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype graph_data_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an GraphData instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

CUDA graph

Type:

int

graph_exec

CUDA executable graph

Type:

int

node

CUDA graph node

Type:

int

node_type

Type of the node

Type:

int

original_graph

The original CUDA graph from which graph is cloned

Type:

int

original_node

The original CUDA graph node from which node is cloned

Type:

int

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.LogicalEndpointResourceData

Bases: object

Empty-initialize an instance of CUpti_LogicalEndpointResourceData.

See also

CUpti_LogicalEndpointResourceData

flags_

Flags associated with the logical endpoint construction. Valid only for the created callback.

Type:

int

static from_buffer(buffer)

Create an LogicalEndpointResourceData instance with the memory from the given buffer.

static from_data(data)

Create an LogicalEndpointResourceData instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype logical_endpoint_resource_data_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an LogicalEndpointResourceData instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

logical_endpoint_id

Identifier of the logical endpoint. Valid for both created and destroyed callbacks.

Type:

int

ptr

Get the pointer address to the data as Python int.

size_

Size in bytes of the region bound to the logical endpoint. Valid for both created and destroyed callbacks.

Type:

int

type

Type of the logical endpoint. Valid only for the created callback; set to CUPTI_LOGICAL_ENDPOINT_TYPE_INVALID for the destroyed callback.

Type:

int

type_info

_py_anon_pod60:

class cupti.cupti.ModuleResourceData

Bases: object

Empty-initialize an instance of CUpti_ModuleResourceData.

See also

CUpti_ModuleResourceData

cubin_size

The size of the cubin.

Type:

int

static from_buffer(buffer)

Create an ModuleResourceData instance with the memory from the given buffer.

static from_data(data)

Create an ModuleResourceData instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype module_resource_data_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ModuleResourceData instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

module_id

Identifier to associate with the CUDA module.

Type:

int

p_cubin

Pointer to the associated cubin.

Type:

str

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxData

Bases: object

Empty-initialize an instance of CUpti_NvtxData.

See also

CUpti_NvtxData

static from_buffer(buffer)

Create an NvtxData instance with the memory from the given buffer.

static from_data(data)

Create an NvtxData instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_data_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an NvtxData instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

function_name

Name of the NVTX API function which issued the callback. This string is a global constant and so may be accessed outside of the callback.

Type:

str

function_params

int: Pointer to the arguments passed to the NVTX API call.

To access the contents, cast this pointer using the appropriate Function Parameter class and its from_ptr method. See Function Parameter Classes for CUDA and NVTX APIs for available classes and usage examples.

function_return_value

Pointer to the return value of the NVTX API call. See nvToolsExt.h for each NVTX API function’s return value.

Type:

int

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.ResourceData

Bases: object

Empty-initialize an instance of CUpti_ResourceData.

See also

CUpti_ResourceData

context

For CUPTI_CBID_RESOURCE_CONTEXT_CREATED and CUPTI_CBID_RESOURCE_CONTEXT_DESTROY_STARTING, the context being created or destroyed. For CUPTI_CBID_RESOURCE_STREAM_CREATED and CUPTI_CBID_RESOURCE_STREAM_DESTROY_STARTING, the context containing the stream being created or destroyed.

Type:

int

static from_buffer(buffer)

Create an ResourceData instance with the memory from the given buffer.

static from_data(data)

Create an ResourceData instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype resource_data_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an ResourceData instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

resource_descriptor

Reserved for future use.

Type:

int

resource_handle

_py_anon_pod0:

class cupti.cupti.StateData

Bases: object

Empty-initialize an instance of CUpti_StateData.

See also

CUpti_StateData

static from_buffer(buffer)

Create an StateData instance with the memory from the given buffer.

static from_data(data)

Create an StateData instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype state_data_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an StateData instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

notification

Data passed along with the callback Ids Enum CUpti_CallbackIdState used to denote callback ids

Type:

_py_anon_pod2

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.StreamAttrData

Bases: object

Empty-initialize an instance of CUpti_StreamAttrData.

See also

CUpti_StreamAttrData

attr

The type of the CUDA stream attribute

Type:

int

static from_buffer(buffer)

Create an StreamAttrData instance with the memory from the given buffer.

static from_data(data)

Create an StreamAttrData instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype stream_attr_data_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an StreamAttrData instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

The CUDA stream handle for the attribute

Type:

int

value

The value of the CUDA stream attribute

Type:

int

class cupti.cupti.SubscriberParams

Empty-initialize an instance of CUpti_SubscriberParams.

See also

CUpti_SubscriberParams

allow_multiple_subscribers

Request to allow multiple subscribers. If 0, requesting to disallow multiple subscribers. If 1, requesting to allow multiple subscribers. Note that cuptiSubscribe_v2 will return CUPTI_ERROR_MULTIPLE_SUBSCRIBERS_NOT_SUPPORTED if the request cannot be fulfilled. Note that support is currently only for CUPTI shared library and not for CUPTI static library, but static library will be supported in future releases.

Type:

int

static from_buffer(buffer)

Create an _SubscriberParams instance with the memory from the given buffer.

static from_data(data)

Create an _SubscriberParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype _subscriber_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an SubscriberParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

old_subscriber_name

In case of multiple subscribers not allowed, and a CUPTI_ERROR_MULTIPLE_SUBSCRIBERS_NOT_SUPPORTED return code, the name of the incompatible tool or the existing CUPTI subscriber will be written to this location. Size should be greater than or equal to CUPTI_OLD_SUBSCRIBER_NAME_MIN_LEN to avoid truncation. Can be NULL. If multiple subscribers are allowed, this will be the name of the first subscriber, but CUPTI_ERROR_MULTIPLE_SUBSCRIBERS_NOT_SUPPORTED will not be returned.

Type:

str

ptr

Get the pointer address to the data as Python int.

struct_size

Size of the data structure. CUPTI client should set the size of the structure. It will be used in CUPTI to check what fields are available in the structure. Used to preserve backward compatibility.

Type:

int

subscriber_name

Name given to the subscriber. The subscriber name need not include the “CUPTI” prefix, as the CUPTI library automatically adds it as “CUPTI for <subscriberName>”. Can be NULL. An internal copy is created. Size must not exceed CUPTI_SUBSCRIBER_NAME_MAX_LEN to avoid truncation.

Type:

str

cupti.cupti.SUBSCRIBER_PARAMS_SIZE

The size, in bytes, of the cupti.cupti.SubscriberParams structure for this version of the bindings. Assign it to the struct_size member of cupti.cupti.SubscriberParams before calling cupti.cupti.subscribe_v2().

class cupti.cupti.SynchronizeData

Bases: object

Empty-initialize an instance of CUpti_SynchronizeData.

See also

CUpti_SynchronizeData

context

The context of the stream being synchronized.

Type:

int

static from_buffer(buffer)

Create an SynchronizeData instance with the memory from the given buffer.

static from_data(data)

Create an SynchronizeData instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype synchronize_data_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an SynchronizeData instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

The stream being synchronized.

Type:

int

Function Parameter Classes for CUDA and NVTX APIs

Each CUPTI callback delivers a function-parameters pointer to the arguments of the intercepted API call — CallbackData.function_params for CUDA runtime and driver API calls, and NvtxData.function_params for NVTX API calls. Use the corresponding Function Parameter class and its from_ptr() class method to access those arguments in a type-safe way.

Example — inspect cudaLaunchKernel() grid and block dimensions from a callback:

from cupti import cupti

def my_callback(user_data, domain, callback_id, cbdata):
    if domain == cupti.CallbackDomain.RUNTIME_API:
        # Inspect cudaLaunchKernel parameters at API entry
        if (callback_id == cupti.RuntimeApiTraceCbid.cudaLaunchKernel_v7000
                and cbdata.callback_site == cupti.ApiCallbackSite.API_ENTER):
            params = cupti.CudaLaunchKernelV7000Params.from_ptr(
                cbdata.function_params, readonly=True
            )
            cbid_name = cupti.RuntimeApiTraceCbid(callback_id).name
            print(f"{cbid_name}: grid={params.grid_dim}, block={params.block_dim}")

Function Parameter Classes for CUDA Runtime APIs

class cupti.cupti.CudaDeviceSetLimitV3020Params

Bases: object

Empty-initialize an instance of cudaDeviceSetLimit_v3020_params.

See also

cudaDeviceSetLimit_v3020_params

static from_buffer(buffer)

Create an CudaDeviceSetLimitV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceSetLimitV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_set_limit_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceSetLimitV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

limit

:

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CudaDeviceGetLimitV3020Params

Bases: object

Empty-initialize an instance of cudaDeviceGetLimit_v3020_params.

See also

cudaDeviceGetLimit_v3020_params

static from_buffer(buffer)

Create an CudaDeviceGetLimitV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetLimitV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_limit_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetLimitV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

limit

:

p_value

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGetTexture1DLinearMaxWidthV11010Params

Bases: object

Empty-initialize an instance of cudaDeviceGetTexture1DLinearMaxWidth_v11010_params.

See also

cudaDeviceGetTexture1DLinearMaxWidth_v11010_params

device_

int:

fmt_desc

int:

static from_buffer(buffer)

Create an CudaDeviceGetTexture1DLinearMaxWidthV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetTexture1DLinearMaxWidthV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_texture1d_linear_max_width_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetTexture1DLinearMaxWidthV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

max_width_in_elements

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGetCacheConfigV3020Params

Bases: object

Empty-initialize an instance of cudaDeviceGetCacheConfig_v3020_params.

See also

cudaDeviceGetCacheConfig_v3020_params

static from_buffer(buffer)

Create an CudaDeviceGetCacheConfigV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetCacheConfigV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_cache_config_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetCacheConfigV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_cache_config

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGetStreamPriorityRangeV5050Params

Bases: object

Empty-initialize an instance of cudaDeviceGetStreamPriorityRange_v5050_params.

See also

cudaDeviceGetStreamPriorityRange_v5050_params

static from_buffer(buffer)

Create an CudaDeviceGetStreamPriorityRangeV5050Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetStreamPriorityRangeV5050Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_stream_priority_range_v5050params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetStreamPriorityRangeV5050Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

greatest_priority

int:

least_priority

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceSetCacheConfigV3020Params

Bases: object

Empty-initialize an instance of cudaDeviceSetCacheConfig_v3020_params.

See also

cudaDeviceSetCacheConfig_v3020_params

cache_config

:

static from_buffer(buffer)

Create an CudaDeviceSetCacheConfigV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceSetCacheConfigV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_set_cache_config_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceSetCacheConfigV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGetByPCIBusIdV4010Params

Bases: object

Empty-initialize an instance of cudaDeviceGetByPCIBusId_v4010_params.

See also

cudaDeviceGetByPCIBusId_v4010_params

device_

int:

static from_buffer(buffer)

Create an CudaDeviceGetByPCIBusIdV4010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetByPCIBusIdV4010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_by_pci_bus_id_v4010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetByPCIBusIdV4010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pci_bus_id

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGetPCIBusIdV4010Params

Bases: object

Empty-initialize an instance of cudaDeviceGetPCIBusId_v4010_params.

See also

cudaDeviceGetPCIBusId_v4010_params

device_

int:

static from_buffer(buffer)

Create an CudaDeviceGetPCIBusIdV4010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetPCIBusIdV4010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_pci_bus_id_v4010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetPCIBusIdV4010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

len

int:

pci_bus_id

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaIpcGetEventHandleV4010Params

Bases: object

Empty-initialize an instance of cudaIpcGetEventHandle_v4010_params.

See also

cudaIpcGetEventHandle_v4010_params

event

int:

static from_buffer(buffer)

Create an CudaIpcGetEventHandleV4010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaIpcGetEventHandleV4010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_ipc_get_event_handle_v4010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaIpcGetEventHandleV4010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaIpcOpenEventHandleV4010Params

Bases: object

Empty-initialize an instance of cudaIpcOpenEventHandle_v4010_params.

See also

cudaIpcOpenEventHandle_v4010_params

event

int:

static from_buffer(buffer)

Create an CudaIpcOpenEventHandleV4010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaIpcOpenEventHandleV4010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_ipc_open_event_handle_v4010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaIpcOpenEventHandleV4010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaIpcGetMemHandleV4010Params

Bases: object

Empty-initialize an instance of cudaIpcGetMemHandle_v4010_params.

See also

cudaIpcGetMemHandle_v4010_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaIpcGetMemHandleV4010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaIpcGetMemHandleV4010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_ipc_get_mem_handle_v4010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaIpcGetMemHandleV4010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaIpcOpenMemHandleV4010Params

Bases: object

Empty-initialize an instance of cudaIpcOpenMemHandle_v4010_params.

See also

cudaIpcOpenMemHandle_v4010_params

dev_ptr

int:

flags_

int:

static from_buffer(buffer)

Create an CudaIpcOpenMemHandleV4010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaIpcOpenMemHandleV4010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_ipc_open_mem_handle_v4010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaIpcOpenMemHandleV4010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaIpcCloseMemHandleV4010Params

Bases: object

Empty-initialize an instance of cudaIpcCloseMemHandle_v4010_params.

See also

cudaIpcCloseMemHandle_v4010_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaIpcCloseMemHandleV4010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaIpcCloseMemHandleV4010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_ipc_close_mem_handle_v4010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaIpcCloseMemHandleV4010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceFlushGPUDirectRDMAWritesV11030Params

Bases: object

Empty-initialize an instance of cudaDeviceFlushGPUDirectRDMAWrites_v11030_params.

See also

cudaDeviceFlushGPUDirectRDMAWrites_v11030_params

static from_buffer(buffer)

Create an CudaDeviceFlushGPUDirectRDMAWritesV11030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceFlushGPUDirectRDMAWritesV11030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_flush_gpu_direct_rdma_writes_v11030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceFlushGPUDirectRDMAWritesV11030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

scope

:

target

:

class cupti.cupti.CudaDeviceRegisterAsyncNotificationV12040Params

Bases: object

Empty-initialize an instance of cudaDeviceRegisterAsyncNotification_v12040_params.

See also

cudaDeviceRegisterAsyncNotification_v12040_params

callback

int:

callback_func

int:

device_

int:

static from_buffer(buffer)

Create an CudaDeviceRegisterAsyncNotificationV12040Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceRegisterAsyncNotificationV12040Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_register_async_notification_v12040params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceRegisterAsyncNotificationV12040Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

user_data

int:

class cupti.cupti.CudaDeviceUnregisterAsyncNotificationV12040Params

Bases: object

Empty-initialize an instance of cudaDeviceUnregisterAsyncNotification_v12040_params.

See also

cudaDeviceUnregisterAsyncNotification_v12040_params

callback

int:

device_

int:

static from_buffer(buffer)

Create an CudaDeviceUnregisterAsyncNotificationV12040Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceUnregisterAsyncNotificationV12040Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_unregister_async_notification_v12040params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceUnregisterAsyncNotificationV12040Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGetSharedMemConfigV4020Params

Bases: object

Empty-initialize an instance of cudaDeviceGetSharedMemConfig_v4020_params.

See also

cudaDeviceGetSharedMemConfig_v4020_params

static from_buffer(buffer)

Create an CudaDeviceGetSharedMemConfigV4020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetSharedMemConfigV4020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_shared_mem_config_v4020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetSharedMemConfigV4020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_config

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceSetSharedMemConfigV4020Params

Bases: object

Empty-initialize an instance of cudaDeviceSetSharedMemConfig_v4020_params.

See also

cudaDeviceSetSharedMemConfig_v4020_params

config

:

static from_buffer(buffer)

Create an CudaDeviceSetSharedMemConfigV4020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceSetSharedMemConfigV4020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_set_shared_mem_config_v4020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceSetSharedMemConfigV4020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGetErrorNameV6050Params

Bases: object

Empty-initialize an instance of cudaGetErrorName_v6050_params.

See also

cudaGetErrorName_v6050_params

error

:

static from_buffer(buffer)

Create an CudaGetErrorNameV6050Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetErrorNameV6050Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_error_name_v6050params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetErrorNameV6050Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGetErrorStringV3020Params

Bases: object

Empty-initialize an instance of cudaGetErrorString_v3020_params.

See also

cudaGetErrorString_v3020_params

error

:

static from_buffer(buffer)

Create an CudaGetErrorStringV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetErrorStringV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_error_string_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetErrorStringV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGetDeviceCountV3020Params

Bases: object

Empty-initialize an instance of cudaGetDeviceCount_v3020_params.

See also

cudaGetDeviceCount_v3020_params

count

int:

static from_buffer(buffer)

Create an CudaGetDeviceCountV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetDeviceCountV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_device_count_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetDeviceCountV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGetAttributeV5000Params

Bases: object

Empty-initialize an instance of cudaDeviceGetAttribute_v5000_params.

See also

cudaDeviceGetAttribute_v5000_params

attr

:

device_

int:

static from_buffer(buffer)

Create an CudaDeviceGetAttributeV5000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetAttributeV5000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_attribute_v5000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetAttributeV5000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CudaDeviceGetDefaultMemPoolV11020Params

Bases: object

Empty-initialize an instance of cudaDeviceGetDefaultMemPool_v11020_params.

See also

cudaDeviceGetDefaultMemPool_v11020_params

device_

int:

static from_buffer(buffer)

Create an CudaDeviceGetDefaultMemPoolV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetDefaultMemPoolV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_default_mem_pool_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetDefaultMemPoolV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceSetMemPoolV11020Params

Bases: object

Empty-initialize an instance of cudaDeviceSetMemPool_v11020_params.

See also

cudaDeviceSetMemPool_v11020_params

device_

int:

static from_buffer(buffer)

Create an CudaDeviceSetMemPoolV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceSetMemPoolV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_set_mem_pool_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceSetMemPoolV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGetMemPoolV11020Params

Bases: object

Empty-initialize an instance of cudaDeviceGetMemPool_v11020_params.

See also

cudaDeviceGetMemPool_v11020_params

device_

int:

static from_buffer(buffer)

Create an CudaDeviceGetMemPoolV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetMemPoolV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_mem_pool_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetMemPoolV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGetNvSciSyncAttributesV10020Params

Bases: object

Empty-initialize an instance of cudaDeviceGetNvSciSyncAttributes_v10020_params.

See also

cudaDeviceGetNvSciSyncAttributes_v10020_params

device_

int:

flags_

int:

static from_buffer(buffer)

Create an CudaDeviceGetNvSciSyncAttributesV10020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetNvSciSyncAttributesV10020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_nv_sci_sync_attributes_v10020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetNvSciSyncAttributesV10020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

nv_sci_sync_attr_list

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGetP2PAttributeV8000Params

Bases: object

Empty-initialize an instance of cudaDeviceGetP2PAttribute_v8000_params.

See also

cudaDeviceGetP2PAttribute_v8000_params

attr

:

dst_device

int:

static from_buffer(buffer)

Create an CudaDeviceGetP2PAttributeV8000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetP2PAttributeV8000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_p2p_attribute_v8000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetP2PAttributeV8000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_device

int:

value

int:

class cupti.cupti.CudaChooseDeviceV3020Params

Bases: object

Empty-initialize an instance of cudaChooseDevice_v3020_params.

See also

cudaChooseDevice_v3020_params

device_

int:

static from_buffer(buffer)

Create an CudaChooseDeviceV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaChooseDeviceV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_choose_device_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaChooseDeviceV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

prop

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaInitDeviceV12000Params

Bases: object

Empty-initialize an instance of cudaInitDevice_v12000_params.

See also

cudaInitDevice_v12000_params

device_

int:

device_flags

int:

flags_

int:

static from_buffer(buffer)

Create an CudaInitDeviceV12000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaInitDeviceV12000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_init_device_v12000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaInitDeviceV12000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaSetDeviceV3020Params

Bases: object

Empty-initialize an instance of cudaSetDevice_v3020_params.

See also

cudaSetDevice_v3020_params

device_

int:

static from_buffer(buffer)

Create an CudaSetDeviceV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaSetDeviceV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_set_device_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaSetDeviceV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGetDeviceV3020Params

Bases: object

Empty-initialize an instance of cudaGetDevice_v3020_params.

See also

cudaGetDevice_v3020_params

device_

int:

static from_buffer(buffer)

Create an CudaGetDeviceV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetDeviceV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_device_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetDeviceV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaSetValidDevicesV3020Params

Bases: object

Empty-initialize an instance of cudaSetValidDevices_v3020_params.

See also

cudaSetValidDevices_v3020_params

device_arr

int:

static from_buffer(buffer)

Create an CudaSetValidDevicesV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaSetValidDevicesV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_set_valid_devices_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaSetValidDevicesV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

len

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaSetDeviceFlagsV3020Params

Bases: object

Empty-initialize an instance of cudaSetDeviceFlags_v3020_params.

See also

cudaSetDeviceFlags_v3020_params

flags_

int:

static from_buffer(buffer)

Create an CudaSetDeviceFlagsV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaSetDeviceFlagsV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_set_device_flags_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaSetDeviceFlagsV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGetDeviceFlagsV7000Params

Bases: object

Empty-initialize an instance of cudaGetDeviceFlags_v7000_params.

See also

cudaGetDeviceFlags_v7000_params

flags_

int:

static from_buffer(buffer)

Create an CudaGetDeviceFlagsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetDeviceFlagsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_device_flags_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetDeviceFlagsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaStreamCreateV3020Params

Bases: object

Empty-initialize an instance of cudaStreamCreate_v3020_params.

See also

cudaStreamCreate_v3020_params

static from_buffer(buffer)

Create an CudaStreamCreateV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamCreateV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_create_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamCreateV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaStreamCreateWithFlagsV5000Params

Bases: object

Empty-initialize an instance of cudaStreamCreateWithFlags_v5000_params.

See also

cudaStreamCreateWithFlags_v5000_params

flags_

int:

static from_buffer(buffer)

Create an CudaStreamCreateWithFlagsV5000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamCreateWithFlagsV5000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_create_with_flags_v5000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamCreateWithFlagsV5000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaStreamCreateWithPriorityV5050Params

Bases: object

Empty-initialize an instance of cudaStreamCreateWithPriority_v5050_params.

See also

cudaStreamCreateWithPriority_v5050_params

flags_

int:

static from_buffer(buffer)

Create an CudaStreamCreateWithPriorityV5050Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamCreateWithPriorityV5050Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_create_with_priority_v5050params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamCreateWithPriorityV5050Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_stream

int:

priority

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaStreamGetPriorityPtszV7000Params

Bases: object

Empty-initialize an instance of cudaStreamGetPriority_ptsz_v7000_params.

See also

cudaStreamGetPriority_ptsz_v7000_params

static from_buffer(buffer)

Create an CudaStreamGetPriorityPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamGetPriorityPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_get_priority_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamGetPriorityPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

priority

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaStreamGetFlagsPtszV7000Params

Bases: object

Empty-initialize an instance of cudaStreamGetFlags_ptsz_v7000_params.

See also

cudaStreamGetFlags_ptsz_v7000_params

flags_

int:

static from_buffer(buffer)

Create an CudaStreamGetFlagsPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamGetFlagsPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_get_flags_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamGetFlagsPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaStreamGetIdPtszV12000Params

Bases: object

Empty-initialize an instance of cudaStreamGetId_ptsz_v12000_params.

See also

cudaStreamGetId_ptsz_v12000_params

static from_buffer(buffer)

Create an CudaStreamGetIdPtszV12000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamGetIdPtszV12000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_get_id_ptsz_v12000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamGetIdPtszV12000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

stream_id

int:

class cupti.cupti.CudaStreamGetDevicePtszV12080Params

Bases: object

Empty-initialize an instance of cudaStreamGetDevice_ptsz_v12080_params.

See also

cudaStreamGetDevice_ptsz_v12080_params

device_

int:

static from_buffer(buffer)

Create an CudaStreamGetDevicePtszV12080Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamGetDevicePtszV12080Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_get_device_ptsz_v12080params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamGetDevicePtszV12080Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaStreamCopyAttributesPtszV11000Params

Bases: object

Empty-initialize an instance of cudaStreamCopyAttributes_ptsz_v11000_params.

See also

cudaStreamCopyAttributes_ptsz_v11000_params

dst

int:

static from_buffer(buffer)

Create an CudaStreamCopyAttributesPtszV11000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamCopyAttributesPtszV11000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_copy_attributes_ptsz_v11000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamCopyAttributesPtszV11000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src

int:

class cupti.cupti.CudaStreamGetAttributePtszV11000Params

Bases: object

Empty-initialize an instance of cudaStreamGetAttribute_ptsz_v11000_params.

See also

cudaStreamGetAttribute_ptsz_v11000_params

attr

:

static from_buffer(buffer)

Create an CudaStreamGetAttributePtszV11000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamGetAttributePtszV11000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_get_attribute_ptsz_v11000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamGetAttributePtszV11000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

value_out

int:

class cupti.cupti.CudaStreamSetAttributePtszV11000Params

Bases: object

Empty-initialize an instance of cudaStreamSetAttribute_ptsz_v11000_params.

See also

cudaStreamSetAttribute_ptsz_v11000_params

attr

:

static from_buffer(buffer)

Create an CudaStreamSetAttributePtszV11000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamSetAttributePtszV11000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_set_attribute_ptsz_v11000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamSetAttributePtszV11000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CudaStreamDestroyV5050Params

Bases: object

Empty-initialize an instance of cudaStreamDestroy_v5050_params.

See also

cudaStreamDestroy_v5050_params

static from_buffer(buffer)

Create an CudaStreamDestroyV5050Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamDestroyV5050Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_destroy_v5050params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamDestroyV5050Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamWaitEventPtszV7000Params

Bases: object

Empty-initialize an instance of cudaStreamWaitEvent_ptsz_v7000_params.

See also

cudaStreamWaitEvent_ptsz_v7000_params

event

int:

flags_

int:

static from_buffer(buffer)

Create an CudaStreamWaitEventPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamWaitEventPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_wait_event_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamWaitEventPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamAddCallbackPtszV7000Params

Bases: object

Empty-initialize an instance of cudaStreamAddCallback_ptsz_v7000_params.

See also

cudaStreamAddCallback_ptsz_v7000_params

callback

int:

flags_

int:

static from_buffer(buffer)

Create an CudaStreamAddCallbackPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamAddCallbackPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_add_callback_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamAddCallbackPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

user_data

int:

class cupti.cupti.CudaStreamSynchronizePtszV7000Params

Bases: object

Empty-initialize an instance of cudaStreamSynchronize_ptsz_v7000_params.

See also

cudaStreamSynchronize_ptsz_v7000_params

static from_buffer(buffer)

Create an CudaStreamSynchronizePtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamSynchronizePtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_synchronize_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamSynchronizePtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamQueryPtszV7000Params

Bases: object

Empty-initialize an instance of cudaStreamQuery_ptsz_v7000_params.

See also

cudaStreamQuery_ptsz_v7000_params

static from_buffer(buffer)

Create an CudaStreamQueryPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamQueryPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_query_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamQueryPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamAttachMemAsyncPtszV7000Params

Bases: object

Empty-initialize an instance of cudaStreamAttachMemAsync_ptsz_v7000_params.

See also

cudaStreamAttachMemAsync_ptsz_v7000_params

dev_ptr

int:

flags_

int:

static from_buffer(buffer)

Create an CudaStreamAttachMemAsyncPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamAttachMemAsyncPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_attach_mem_async_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamAttachMemAsyncPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

length

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamBeginCapturePtszV10000Params

Bases: object

Empty-initialize an instance of cudaStreamBeginCapture_ptsz_v10000_params.

See also

cudaStreamBeginCapture_ptsz_v10000_params

static from_buffer(buffer)

Create an CudaStreamBeginCapturePtszV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamBeginCapturePtszV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_begin_capture_ptsz_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamBeginCapturePtszV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mode

:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamBeginCaptureToGraphPtszV12030Params

Bases: object

Empty-initialize an instance of cudaStreamBeginCaptureToGraph_ptsz_v12030_params.

See also

cudaStreamBeginCaptureToGraph_ptsz_v12030_params

dependencies

int:

dependency_data

int:

static from_buffer(buffer)

Create an CudaStreamBeginCaptureToGraphPtszV12030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamBeginCaptureToGraphPtszV12030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_begin_capture_to_graph_ptsz_v12030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamBeginCaptureToGraphPtszV12030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

mode

:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaThreadExchangeStreamCaptureModeV10010Params

Bases: object

Empty-initialize an instance of cudaThreadExchangeStreamCaptureMode_v10010_params.

See also

cudaThreadExchangeStreamCaptureMode_v10010_params

static from_buffer(buffer)

Create an CudaThreadExchangeStreamCaptureModeV10010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaThreadExchangeStreamCaptureModeV10010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_thread_exchange_stream_capture_mode_v10010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaThreadExchangeStreamCaptureModeV10010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mode

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaStreamEndCapturePtszV10000Params

Bases: object

Empty-initialize an instance of cudaStreamEndCapture_ptsz_v10000_params.

See also

cudaStreamEndCapture_ptsz_v10000_params

static from_buffer(buffer)

Create an CudaStreamEndCapturePtszV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamEndCapturePtszV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_end_capture_ptsz_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamEndCapturePtszV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_graph

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamIsCapturingPtszV10000Params

Bases: object

Empty-initialize an instance of cudaStreamIsCapturing_ptsz_v10000_params.

See also

cudaStreamIsCapturing_ptsz_v10000_params

static from_buffer(buffer)

Create an CudaStreamIsCapturingPtszV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamIsCapturingPtszV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_is_capturing_ptsz_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamIsCapturingPtszV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_capture_status

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaEventCreateV3020Params

Bases: object

Empty-initialize an instance of cudaEventCreate_v3020_params.

See also

cudaEventCreate_v3020_params

event

int:

static from_buffer(buffer)

Create an CudaEventCreateV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaEventCreateV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_event_create_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaEventCreateV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaEventCreateWithFlagsV3020Params

Bases: object

Empty-initialize an instance of cudaEventCreateWithFlags_v3020_params.

See also

cudaEventCreateWithFlags_v3020_params

event

int:

flags_

int:

static from_buffer(buffer)

Create an CudaEventCreateWithFlagsV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaEventCreateWithFlagsV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_event_create_with_flags_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaEventCreateWithFlagsV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaEventRecordPtszV7000Params

Bases: object

Empty-initialize an instance of cudaEventRecord_ptsz_v7000_params.

See also

cudaEventRecord_ptsz_v7000_params

event

int:

static from_buffer(buffer)

Create an CudaEventRecordPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaEventRecordPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_event_record_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaEventRecordPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaEventRecordWithFlagsPtszV11010Params

Bases: object

Empty-initialize an instance of cudaEventRecordWithFlags_ptsz_v11010_params.

See also

cudaEventRecordWithFlags_ptsz_v11010_params

event

int:

flags_

int:

static from_buffer(buffer)

Create an CudaEventRecordWithFlagsPtszV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaEventRecordWithFlagsPtszV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_event_record_with_flags_ptsz_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaEventRecordWithFlagsPtszV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaEventQueryV3020Params

Bases: object

Empty-initialize an instance of cudaEventQuery_v3020_params.

See also

cudaEventQuery_v3020_params

event

int:

static from_buffer(buffer)

Create an CudaEventQueryV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaEventQueryV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_event_query_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaEventQueryV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaEventSynchronizeV3020Params

Bases: object

Empty-initialize an instance of cudaEventSynchronize_v3020_params.

See also

cudaEventSynchronize_v3020_params

event

int:

static from_buffer(buffer)

Create an CudaEventSynchronizeV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaEventSynchronizeV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_event_synchronize_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaEventSynchronizeV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaEventDestroyV3020Params

Bases: object

Empty-initialize an instance of cudaEventDestroy_v3020_params.

See also

cudaEventDestroy_v3020_params

event

int:

static from_buffer(buffer)

Create an CudaEventDestroyV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaEventDestroyV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_event_destroy_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaEventDestroyV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaImportExternalMemoryV10000Params

Bases: object

Empty-initialize an instance of cudaImportExternalMemory_v10000_params.

See also

cudaImportExternalMemory_v10000_params

ext_mem_out

int:

static from_buffer(buffer)

Create an CudaImportExternalMemoryV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaImportExternalMemoryV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_import_external_memory_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaImportExternalMemoryV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mem_handle_desc

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaExternalMemoryGetMappedBufferV10000Params

Bases: object

Empty-initialize an instance of cudaExternalMemoryGetMappedBuffer_v10000_params.

See also

cudaExternalMemoryGetMappedBuffer_v10000_params

buffer_desc

int:

dev_ptr

int:

ext_mem

int:

static from_buffer(buffer)

Create an CudaExternalMemoryGetMappedBufferV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaExternalMemoryGetMappedBufferV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_external_memory_get_mapped_buffer_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaExternalMemoryGetMappedBufferV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaExternalMemoryGetMappedMipmappedArrayV10000Params

Bases: object

Empty-initialize an instance of cudaExternalMemoryGetMappedMipmappedArray_v10000_params.

See also

cudaExternalMemoryGetMappedMipmappedArray_v10000_params

ext_mem

int:

static from_buffer(

buffer,

)

Create an CudaExternalMemoryGetMappedMipmappedArrayV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaExternalMemoryGetMappedMipmappedArrayV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_external_memory_get_mapped_mipmapped_array_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaExternalMemoryGetMappedMipmappedArrayV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mipmap

int:

mipmap_desc

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDestroyExternalMemoryV10000Params

Bases: object

Empty-initialize an instance of cudaDestroyExternalMemory_v10000_params.

See also

cudaDestroyExternalMemory_v10000_params

ext_mem

int:

static from_buffer(buffer)

Create an CudaDestroyExternalMemoryV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDestroyExternalMemoryV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_destroy_external_memory_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDestroyExternalMemoryV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaImportExternalSemaphoreV10000Params

Bases: object

Empty-initialize an instance of cudaImportExternalSemaphore_v10000_params.

See also

cudaImportExternalSemaphore_v10000_params

ext_sem_out

int:

static from_buffer(buffer)

Create an CudaImportExternalSemaphoreV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaImportExternalSemaphoreV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_import_external_semaphore_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaImportExternalSemaphoreV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

sem_handle_desc

int:

class cupti.cupti.CudaDestroyExternalSemaphoreV10000Params

Bases: object

Empty-initialize an instance of cudaDestroyExternalSemaphore_v10000_params.

See also

cudaDestroyExternalSemaphore_v10000_params

ext_sem

int:

static from_buffer(buffer)

Create an CudaDestroyExternalSemaphoreV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDestroyExternalSemaphoreV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_destroy_external_semaphore_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDestroyExternalSemaphoreV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLaunchKernelPtszV7000Params

Bases: object

Empty-initialize an instance of cudaLaunchKernel_ptsz_v7000_params.

See also

cudaLaunchKernel_ptsz_v7000_params

args

int:

block_dim

:

static from_buffer(buffer)

Create an CudaLaunchKernelPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLaunchKernelPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_launch_kernel_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLaunchKernelPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

grid_dim

:

ptr

Get the pointer address to the data as Python int.

shared_mem

int:

stream

int:

class cupti.cupti.CudaLaunchKernelExCPtszV11060Params

Bases: object

Empty-initialize an instance of cudaLaunchKernelExC_ptsz_v11060_params.

See also

cudaLaunchKernelExC_ptsz_v11060_params

args

int:

config

int:

static from_buffer(buffer)

Create an CudaLaunchKernelExCPtszV11060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLaunchKernelExCPtszV11060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_launch_kernel_ex_c_ptsz_v11060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLaunchKernelExCPtszV11060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLaunchCooperativeKernelPtszV9000Params

Bases: object

Empty-initialize an instance of cudaLaunchCooperativeKernel_ptsz_v9000_params.

See also

cudaLaunchCooperativeKernel_ptsz_v9000_params

args

int:

block_dim

:

static from_buffer(buffer)

Create an CudaLaunchCooperativeKernelPtszV9000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLaunchCooperativeKernelPtszV9000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_launch_cooperative_kernel_ptsz_v9000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLaunchCooperativeKernelPtszV9000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

grid_dim

:

ptr

Get the pointer address to the data as Python int.

shared_mem

int:

stream

int:

class cupti.cupti.CudaFuncSetCacheConfigV3020Params

Bases: object

Empty-initialize an instance of cudaFuncSetCacheConfig_v3020_params.

See also

cudaFuncSetCacheConfig_v3020_params

cache_config

:

static from_buffer(buffer)

Create an CudaFuncSetCacheConfigV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaFuncSetCacheConfigV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_func_set_cache_config_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaFuncSetCacheConfigV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaFuncGetAttributesV3020Params

Bases: object

Empty-initialize an instance of cudaFuncGetAttributes_v3020_params.

See also

cudaFuncGetAttributes_v3020_params

attr

int:

static from_buffer(buffer)

Create an CudaFuncGetAttributesV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaFuncGetAttributesV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_func_get_attributes_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaFuncGetAttributesV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaFuncSetAttributeV9000Params

Bases: object

Empty-initialize an instance of cudaFuncSetAttribute_v9000_params.

See also

cudaFuncSetAttribute_v9000_params

attr

:

static from_buffer(buffer)

Create an CudaFuncSetAttributeV9000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaFuncSetAttributeV9000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_func_set_attribute_v9000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaFuncSetAttributeV9000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CudaFuncGetNameV12030Params

Bases: object

Empty-initialize an instance of cudaFuncGetName_v12030_params.

See also

cudaFuncGetName_v12030_params

static from_buffer(buffer)

Create an CudaFuncGetNameV12030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaFuncGetNameV12030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_func_get_name_v12030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaFuncGetNameV12030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaFuncGetParamInfoV12040Params

Bases: object

Empty-initialize an instance of cudaFuncGetParamInfo_v12040_params.

See also

cudaFuncGetParamInfo_v12040_params

static from_buffer(buffer)

Create an CudaFuncGetParamInfoV12040Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaFuncGetParamInfoV12040Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_func_get_param_info_v12040params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaFuncGetParamInfoV12040Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

param_index

int:

param_offset

int:

param_size

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLaunchHostFuncPtszV10000Params

Bases: object

Empty-initialize an instance of cudaLaunchHostFunc_ptsz_v10000_params.

See also

cudaLaunchHostFunc_ptsz_v10000_params

fn

int:

static from_buffer(buffer)

Create an CudaLaunchHostFuncPtszV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLaunchHostFuncPtszV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_launch_host_func_ptsz_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLaunchHostFuncPtszV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

user_data

int:

class cupti.cupti.CudaFuncSetSharedMemConfigV4020Params

Bases: object

Empty-initialize an instance of cudaFuncSetSharedMemConfig_v4020_params.

See also

cudaFuncSetSharedMemConfig_v4020_params

config

:

static from_buffer(buffer)

Create an CudaFuncSetSharedMemConfigV4020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaFuncSetSharedMemConfigV4020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_func_set_shared_mem_config_v4020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaFuncSetSharedMemConfigV4020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaOccupancyMaxActiveBlocksPerMultiprocessorV6050Params

Bases: object

Empty-initialize an instance of cudaOccupancyMaxActiveBlocksPerMultiprocessor_v6050_params.

See also

cudaOccupancyMaxActiveBlocksPerMultiprocessor_v6050_params

block_size

int:

dynamic_s_mem_size

int:

static from_buffer(

buffer,

)

Create an CudaOccupancyMaxActiveBlocksPerMultiprocessorV6050Params instance with the memory from the given buffer.

static from_data(

data,

)

Create an CudaOccupancyMaxActiveBlocksPerMultiprocessorV6050Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_occupancy_max_active_blocks_per_multiprocessor_v6050params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaOccupancyMaxActiveBlocksPerMultiprocessorV6050Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

num_blocks

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaOccupancyAvailableDynamicSMemPerBlockV10200Params

Bases: object

Empty-initialize an instance of cudaOccupancyAvailableDynamicSMemPerBlock_v10200_params.

See also

cudaOccupancyAvailableDynamicSMemPerBlock_v10200_params

block_size

int:

dynamic_smem_size

int:

static from_buffer(

buffer,

)

Create an CudaOccupancyAvailableDynamicSMemPerBlockV10200Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaOccupancyAvailableDynamicSMemPerBlockV10200Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_occupancy_available_dynamic_s_mem_per_block_v10200params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaOccupancyAvailableDynamicSMemPerBlockV10200Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

num_blocks

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlagsV7000Params

Bases: object

Empty-initialize an instance of cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags_v7000_params.

See also

cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags_v7000_params

block_size

int:

dynamic_s_mem_size

int:

flags_

int:

static from_buffer(

buffer,

)

Create an CudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlagsV7000Params instance with the memory from the given buffer.

static from_data(

data,

)

Create an CudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlagsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_occupancy_max_active_blocks_per_multiprocessor_with_flags_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlagsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

num_blocks

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaOccupancyMaxPotentialClusterSizeV11070Params

Bases: object

Empty-initialize an instance of cudaOccupancyMaxPotentialClusterSize_v11070_params.

See also

cudaOccupancyMaxPotentialClusterSize_v11070_params

cluster_size

int:

static from_buffer(buffer)

Create an CudaOccupancyMaxPotentialClusterSizeV11070Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaOccupancyMaxPotentialClusterSizeV11070Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_occupancy_max_potential_cluster_size_v11070params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaOccupancyMaxPotentialClusterSizeV11070Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

launch_config

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaOccupancyMaxActiveClustersV11070Params

Bases: object

Empty-initialize an instance of cudaOccupancyMaxActiveClusters_v11070_params.

See also

cudaOccupancyMaxActiveClusters_v11070_params

static from_buffer(buffer)

Create an CudaOccupancyMaxActiveClustersV11070Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaOccupancyMaxActiveClustersV11070Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_occupancy_max_active_clusters_v11070params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaOccupancyMaxActiveClustersV11070Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

launch_config

int:

num_clusters

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMallocManagedV6000Params

Bases: object

Empty-initialize an instance of cudaMallocManaged_v6000_params.

See also

cudaMallocManaged_v6000_params

dev_ptr

int:

flags_

int:

static from_buffer(buffer)

Create an CudaMallocManagedV6000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMallocManagedV6000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_malloc_managed_v6000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMallocManagedV6000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CudaMallocV3020Params

Bases: object

Empty-initialize an instance of cudaMalloc_v3020_params.

See also

cudaMalloc_v3020_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMallocV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMallocV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_malloc_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMallocV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CudaMallocHostV3020Params

Bases: object

Empty-initialize an instance of cudaMallocHost_v3020_params.

See also

cudaMallocHost_v3020_params

static from_buffer(buffer)

Create an CudaMallocHostV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMallocHostV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_malloc_host_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMallocHostV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

ptr_

int:

size_

int:

class cupti.cupti.CudaMallocPitchV3020Params

Bases: object

Empty-initialize an instance of cudaMallocPitch_v3020_params.

See also

cudaMallocPitch_v3020_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMallocPitchV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMallocPitchV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_malloc_pitch_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMallocPitchV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

pitch

int:

ptr

Get the pointer address to the data as Python int.

width

int:

class cupti.cupti.CudaMallocArrayV3020Params

Bases: object

Empty-initialize an instance of cudaMallocArray_v3020_params.

See also

cudaMallocArray_v3020_params

array

int:

desc

int:

flags_

int:

static from_buffer(buffer)

Create an CudaMallocArrayV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMallocArrayV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_malloc_array_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMallocArrayV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

ptr

Get the pointer address to the data as Python int.

width

int:

class cupti.cupti.CudaFreeV3020Params

Bases: object

Empty-initialize an instance of cudaFree_v3020_params.

See also

cudaFree_v3020_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaFreeV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaFreeV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_free_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaFreeV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaFreeHostV3020Params

Bases: object

Empty-initialize an instance of cudaFreeHost_v3020_params.

See also

cudaFreeHost_v3020_params

static from_buffer(buffer)

Create an CudaFreeHostV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaFreeHostV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_free_host_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaFreeHostV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

ptr_

int:

class cupti.cupti.CudaFreeArrayV3020Params

Bases: object

Empty-initialize an instance of cudaFreeArray_v3020_params.

See also

cudaFreeArray_v3020_params

array

int:

static from_buffer(buffer)

Create an CudaFreeArrayV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaFreeArrayV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_free_array_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaFreeArrayV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaFreeMipmappedArrayV5000Params

Bases: object

Empty-initialize an instance of cudaFreeMipmappedArray_v5000_params.

See also

cudaFreeMipmappedArray_v5000_params

static from_buffer(buffer)

Create an CudaFreeMipmappedArrayV5000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaFreeMipmappedArrayV5000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_free_mipmapped_array_v5000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaFreeMipmappedArrayV5000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mipmapped_array

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaHostAllocV3020Params

Bases: object

Empty-initialize an instance of cudaHostAlloc_v3020_params.

See also

cudaHostAlloc_v3020_params

flags_

int:

static from_buffer(buffer)

Create an CudaHostAllocV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaHostAllocV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_host_alloc_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaHostAllocV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_host

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CudaHostRegisterV4000Params

Bases: object

Empty-initialize an instance of cudaHostRegister_v4000_params.

See also

cudaHostRegister_v4000_params

flags_

int:

static from_buffer(buffer)

Create an CudaHostRegisterV4000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaHostRegisterV4000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_host_register_v4000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaHostRegisterV4000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

ptr_

int:

size_

int:

class cupti.cupti.CudaHostUnregisterV4000Params

Bases: object

Empty-initialize an instance of cudaHostUnregister_v4000_params.

See also

cudaHostUnregister_v4000_params

static from_buffer(buffer)

Create an CudaHostUnregisterV4000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaHostUnregisterV4000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_host_unregister_v4000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaHostUnregisterV4000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

ptr_

int:

class cupti.cupti.CudaHostGetDevicePointerV3020Params

Bases: object

Empty-initialize an instance of cudaHostGetDevicePointer_v3020_params.

See also

cudaHostGetDevicePointer_v3020_params

flags_

int:

static from_buffer(buffer)

Create an CudaHostGetDevicePointerV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaHostGetDevicePointerV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_host_get_device_pointer_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaHostGetDevicePointerV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_device

int:

p_host

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaHostGetFlagsV3020Params

Bases: object

Empty-initialize an instance of cudaHostGetFlags_v3020_params.

See also

cudaHostGetFlags_v3020_params

static from_buffer(buffer)

Create an CudaHostGetFlagsV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaHostGetFlagsV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_host_get_flags_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaHostGetFlagsV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_flags

int:

p_host

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMalloc3DV3020Params

Bases: object

Empty-initialize an instance of cudaMalloc3D_v3020_params.

See also

cudaMalloc3D_v3020_params

static from_buffer(buffer)

Create an CudaMalloc3DV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMalloc3DV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_malloc3d_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMalloc3DV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pitched_dev_ptr

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMalloc3DArrayV3020Params

Bases: object

Empty-initialize an instance of cudaMalloc3DArray_v3020_params.

See also

cudaMalloc3DArray_v3020_params

array

int:

desc

int:

flags_

int:

static from_buffer(buffer)

Create an CudaMalloc3DArrayV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMalloc3DArrayV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_malloc3d_array_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMalloc3DArrayV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMallocMipmappedArrayV5000Params

Bases: object

Empty-initialize an instance of cudaMallocMipmappedArray_v5000_params.

See also

cudaMallocMipmappedArray_v5000_params

desc

int:

flags_

int:

static from_buffer(buffer)

Create an CudaMallocMipmappedArrayV5000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMallocMipmappedArrayV5000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_malloc_mipmapped_array_v5000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMallocMipmappedArrayV5000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mipmapped_array

int:

num_levels

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGetMipmappedArrayLevelV5000Params

Bases: object

Empty-initialize an instance of cudaGetMipmappedArrayLevel_v5000_params.

See also

cudaGetMipmappedArrayLevel_v5000_params

static from_buffer(buffer)

Create an CudaGetMipmappedArrayLevelV5000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetMipmappedArrayLevelV5000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_mipmapped_array_level_v5000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetMipmappedArrayLevelV5000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

level

int:

level_array

int:

mipmapped_array

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemcpy3DPtdsV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpy3D_ptds_v7000_params.

See also

cudaMemcpy3D_ptds_v7000_params

static from_buffer(buffer)

Create an CudaMemcpy3DPtdsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy3DPtdsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy3d_ptds_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy3DPtdsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemcpy3DPeerPtdsV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpy3DPeer_ptds_v7000_params.

See also

cudaMemcpy3DPeer_ptds_v7000_params

static from_buffer(buffer)

Create an CudaMemcpy3DPeerPtdsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy3DPeerPtdsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy3d_peer_ptds_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy3DPeerPtdsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemcpy3DAsyncPtszV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpy3DAsync_ptsz_v7000_params.

See also

cudaMemcpy3DAsync_ptsz_v7000_params

static from_buffer(buffer)

Create an CudaMemcpy3DAsyncPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy3DAsyncPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy3_dasync_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy3DAsyncPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaMemcpy3DPeerAsyncPtszV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpy3DPeerAsync_ptsz_v7000_params.

See also

cudaMemcpy3DPeerAsync_ptsz_v7000_params

static from_buffer(buffer)

Create an CudaMemcpy3DPeerAsyncPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy3DPeerAsyncPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy3d_peer_async_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy3DPeerAsyncPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaMemGetInfoV3020Params

Bases: object

Empty-initialize an instance of cudaMemGetInfo_v3020_params.

See also

cudaMemGetInfo_v3020_params

free

int:

static from_buffer(buffer)

Create an CudaMemGetInfoV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemGetInfoV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_get_info_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemGetInfoV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

total

int:

class cupti.cupti.CudaArrayGetInfoV4010Params

Bases: object

Empty-initialize an instance of cudaArrayGetInfo_v4010_params.

See also

cudaArrayGetInfo_v4010_params

array

int:

desc

int:

extent

int:

flags_

int:

static from_buffer(buffer)

Create an CudaArrayGetInfoV4010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaArrayGetInfoV4010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_array_get_info_v4010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaArrayGetInfoV4010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaArrayGetPlaneV11020Params

Bases: object

Empty-initialize an instance of cudaArrayGetPlane_v11020_params.

See also

cudaArrayGetPlane_v11020_params

static from_buffer(buffer)

Create an CudaArrayGetPlaneV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaArrayGetPlaneV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_array_get_plane_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaArrayGetPlaneV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_array

int:

p_plane_array

int:

plane_idx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaArrayGetMemoryRequirementsV11060Params

Bases: object

Empty-initialize an instance of cudaArrayGetMemoryRequirements_v11060_params.

See also

cudaArrayGetMemoryRequirements_v11060_params

array

int:

device_

int:

static from_buffer(buffer)

Create an CudaArrayGetMemoryRequirementsV11060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaArrayGetMemoryRequirementsV11060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_array_get_memory_requirements_v11060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaArrayGetMemoryRequirementsV11060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

memory_requirements

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMipmappedArrayGetMemoryRequirementsV11060Params

Bases: object

Empty-initialize an instance of cudaMipmappedArrayGetMemoryRequirements_v11060_params.

See also

cudaMipmappedArrayGetMemoryRequirements_v11060_params

device_

int:

static from_buffer(

buffer,

)

Create an CudaMipmappedArrayGetMemoryRequirementsV11060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMipmappedArrayGetMemoryRequirementsV11060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mipmapped_array_get_memory_requirements_v11060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMipmappedArrayGetMemoryRequirementsV11060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

memory_requirements

int:

mipmap

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaArrayGetSparsePropertiesV11010Params

Bases: object

Empty-initialize an instance of cudaArrayGetSparseProperties_v11010_params.

See also

cudaArrayGetSparseProperties_v11010_params

array

int:

static from_buffer(buffer)

Create an CudaArrayGetSparsePropertiesV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaArrayGetSparsePropertiesV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_array_get_sparse_properties_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaArrayGetSparsePropertiesV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

sparse_properties

int:

class cupti.cupti.CudaMipmappedArrayGetSparsePropertiesV11010Params

Bases: object

Empty-initialize an instance of cudaMipmappedArrayGetSparseProperties_v11010_params.

See also

cudaMipmappedArrayGetSparseProperties_v11010_params

static from_buffer(buffer)

Create an CudaMipmappedArrayGetSparsePropertiesV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMipmappedArrayGetSparsePropertiesV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mipmapped_array_get_sparse_properties_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMipmappedArrayGetSparsePropertiesV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mipmap

int:

ptr

Get the pointer address to the data as Python int.

sparse_properties

int:

class cupti.cupti.CudaMemcpyPtdsV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpy_ptds_v7000_params.

See also

cudaMemcpy_ptds_v7000_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyPtdsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyPtdsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_ptds_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyPtdsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

class cupti.cupti.CudaMemcpyPeerV4000Params

Bases: object

Empty-initialize an instance of cudaMemcpyPeer_v4000_params.

See also

cudaMemcpyPeer_v4000_params

count

int:

dst

int:

dst_device

int:

static from_buffer(buffer)

Create an CudaMemcpyPeerV4000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyPeerV4000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_peer_v4000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyPeerV4000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src

int:

src_device

int:

class cupti.cupti.CudaMemcpy2DPtdsV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpy2D_ptds_v7000_params.

See also

cudaMemcpy2D_ptds_v7000_params

dpitch

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpy2DPtdsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy2DPtdsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy2d_ptds_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy2DPtdsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

kind

:

ptr

Get the pointer address to the data as Python int.

spitch

int:

src

int:

width

int:

class cupti.cupti.CudaMemcpy2DToArrayPtdsV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpy2DToArray_ptds_v7000_params.

See also

cudaMemcpy2DToArray_ptds_v7000_params

dst

int:

static from_buffer(buffer)

Create an CudaMemcpy2DToArrayPtdsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy2DToArrayPtdsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy2d_to_array_ptds_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy2DToArrayPtdsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

height

int:

kind

:

ptr

Get the pointer address to the data as Python int.

spitch

int:

src

int:

w_offset

int:

width

int:

class cupti.cupti.CudaMemcpy2DFromArrayPtdsV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpy2DFromArray_ptds_v7000_params.

See also

cudaMemcpy2DFromArray_ptds_v7000_params

dpitch

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpy2DFromArrayPtdsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy2DFromArrayPtdsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy2d_from_array_ptds_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy2DFromArrayPtdsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

height

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

w_offset

int:

width

int:

class cupti.cupti.CudaMemcpy2DArrayToArrayPtdsV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpy2DArrayToArray_ptds_v7000_params.

See also

cudaMemcpy2DArrayToArray_ptds_v7000_params

dst

int:

static from_buffer(buffer)

Create an CudaMemcpy2DArrayToArrayPtdsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy2DArrayToArrayPtdsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy2d_array_to_array_ptds_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy2DArrayToArrayPtdsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset_dst

int:

h_offset_src

int:

height

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

w_offset_dst

int:

w_offset_src

int:

width

int:

class cupti.cupti.CudaMemcpyToSymbolPtdsV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpyToSymbol_ptds_v7000_params.

See also

cudaMemcpyToSymbol_ptds_v7000_params

count

int:

static from_buffer(buffer)

Create an CudaMemcpyToSymbolPtdsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyToSymbolPtdsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_to_symbol_ptds_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyToSymbolPtdsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

:

offset

int:

ptr

Get the pointer address to the data as Python int.

src

int:

symbol

int:

class cupti.cupti.CudaMemcpyFromSymbolPtdsV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpyFromSymbol_ptds_v7000_params.

See also

cudaMemcpyFromSymbol_ptds_v7000_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyFromSymbolPtdsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyFromSymbolPtdsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_from_symbol_ptds_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyFromSymbolPtdsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

:

offset

int:

ptr

Get the pointer address to the data as Python int.

symbol

int:

class cupti.cupti.CudaMemcpyAsyncPtszV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpyAsync_ptsz_v7000_params.

See also

cudaMemcpyAsync_ptsz_v7000_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyAsyncPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyAsyncPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_async_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyAsyncPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

stream

int:

class cupti.cupti.CudaMemcpyPeerAsyncV4000Params

Bases: object

Empty-initialize an instance of cudaMemcpyPeerAsync_v4000_params.

See also

cudaMemcpyPeerAsync_v4000_params

count

int:

dst

int:

dst_device

int:

static from_buffer(buffer)

Create an CudaMemcpyPeerAsyncV4000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyPeerAsyncV4000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_peer_async_v4000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyPeerAsyncV4000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src

int:

src_device

int:

stream

int:

class cupti.cupti.CudaMemcpy2DAsyncPtszV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpy2DAsync_ptsz_v7000_params.

See also

cudaMemcpy2DAsync_ptsz_v7000_params

dpitch

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpy2DAsyncPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy2DAsyncPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy2_dasync_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy2DAsyncPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

kind

:

ptr

Get the pointer address to the data as Python int.

spitch

int:

src

int:

stream

int:

width

int:

class cupti.cupti.CudaMemcpy2DToArrayAsyncPtszV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpy2DToArrayAsync_ptsz_v7000_params.

See also

cudaMemcpy2DToArrayAsync_ptsz_v7000_params

dst

int:

static from_buffer(buffer)

Create an CudaMemcpy2DToArrayAsyncPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy2DToArrayAsyncPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy2d_to_array_async_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy2DToArrayAsyncPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

height

int:

kind

:

ptr

Get the pointer address to the data as Python int.

spitch

int:

src

int:

stream

int:

w_offset

int:

width

int:

class cupti.cupti.CudaMemcpy2DFromArrayAsyncPtszV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpy2DFromArrayAsync_ptsz_v7000_params.

See also

cudaMemcpy2DFromArrayAsync_ptsz_v7000_params

dpitch

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpy2DFromArrayAsyncPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy2DFromArrayAsyncPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy2d_from_array_async_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy2DFromArrayAsyncPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

height

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

stream

int:

w_offset

int:

width

int:

class cupti.cupti.CudaMemcpyToSymbolAsyncPtszV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpyToSymbolAsync_ptsz_v7000_params.

See also

cudaMemcpyToSymbolAsync_ptsz_v7000_params

count

int:

static from_buffer(buffer)

Create an CudaMemcpyToSymbolAsyncPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyToSymbolAsyncPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_to_symbol_async_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyToSymbolAsyncPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

:

offset

int:

ptr

Get the pointer address to the data as Python int.

src

int:

stream

int:

symbol

int:

class cupti.cupti.CudaMemcpyFromSymbolAsyncPtszV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpyFromSymbolAsync_ptsz_v7000_params.

See also

cudaMemcpyFromSymbolAsync_ptsz_v7000_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyFromSymbolAsyncPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyFromSymbolAsyncPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_from_symbol_async_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyFromSymbolAsyncPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

:

offset

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

symbol

int:

class cupti.cupti.CudaMemsetPtdsV7000Params

Bases: object

Empty-initialize an instance of cudaMemset_ptds_v7000_params.

See also

cudaMemset_ptds_v7000_params

count

int:

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMemsetPtdsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemsetPtdsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memset_ptds_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemsetPtdsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CudaMemset2DPtdsV7000Params

Bases: object

Empty-initialize an instance of cudaMemset2D_ptds_v7000_params.

See also

cudaMemset2D_ptds_v7000_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMemset2DPtdsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemset2DPtdsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memset2d_ptds_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemset2DPtdsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

pitch

int:

ptr

Get the pointer address to the data as Python int.

value

int:

width

int:

class cupti.cupti.CudaMemset3DPtdsV7000Params

Bases: object

Empty-initialize an instance of cudaMemset3D_ptds_v7000_params.

See also

cudaMemset3D_ptds_v7000_params

static from_buffer(buffer)

Create an CudaMemset3DPtdsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemset3DPtdsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memset3d_ptds_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemset3DPtdsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CudaMemsetAsyncPtszV7000Params

Bases: object

Empty-initialize an instance of cudaMemsetAsync_ptsz_v7000_params.

See also

cudaMemsetAsync_ptsz_v7000_params

count

int:

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMemsetAsyncPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemsetAsyncPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memset_async_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemsetAsyncPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CudaMemset2DAsyncPtszV7000Params

Bases: object

Empty-initialize an instance of cudaMemset2DAsync_ptsz_v7000_params.

See also

cudaMemset2DAsync_ptsz_v7000_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMemset2DAsyncPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemset2DAsyncPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memset2_dasync_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemset2DAsyncPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

pitch

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

width

int:

class cupti.cupti.CudaMemset3DAsyncPtszV7000Params

Bases: object

Empty-initialize an instance of cudaMemset3DAsync_ptsz_v7000_params.

See also

cudaMemset3DAsync_ptsz_v7000_params

static from_buffer(buffer)

Create an CudaMemset3DAsyncPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemset3DAsyncPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memset3_dasync_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemset3DAsyncPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CudaGetSymbolAddressV3020Params

Bases: object

Empty-initialize an instance of cudaGetSymbolAddress_v3020_params.

See also

cudaGetSymbolAddress_v3020_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaGetSymbolAddressV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetSymbolAddressV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_symbol_address_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetSymbolAddressV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

symbol

int:

class cupti.cupti.CudaGetSymbolSizeV3020Params

Bases: object

Empty-initialize an instance of cudaGetSymbolSize_v3020_params.

See also

cudaGetSymbolSize_v3020_params

static from_buffer(buffer)

Create an CudaGetSymbolSizeV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetSymbolSizeV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_symbol_size_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetSymbolSizeV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

size_

int:

symbol

int:

class cupti.cupti.CudaMemRangeGetAttributeV8000Params

Bases: object

Empty-initialize an instance of cudaMemRangeGetAttribute_v8000_params.

See also

cudaMemRangeGetAttribute_v8000_params

attribute

:

count

int:

data_

int:

data_size

int:

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMemRangeGetAttributeV8000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemRangeGetAttributeV8000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_range_get_attribute_v8000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemRangeGetAttributeV8000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemRangeGetAttributesV8000Params

Bases: object

Empty-initialize an instance of cudaMemRangeGetAttributes_v8000_params.

See also

cudaMemRangeGetAttributes_v8000_params

attributes

int:

count

int:

data_

int:

data_sizes

int:

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMemRangeGetAttributesV8000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemRangeGetAttributesV8000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_range_get_attributes_v8000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemRangeGetAttributesV8000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_attributes

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemcpyToArrayPtdsV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpyToArray_ptds_v7000_params.

See also

cudaMemcpyToArray_ptds_v7000_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyToArrayPtdsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyToArrayPtdsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_to_array_ptds_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyToArrayPtdsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

w_offset

int:

class cupti.cupti.CudaMemcpyFromArrayPtdsV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpyFromArray_ptds_v7000_params.

See also

cudaMemcpyFromArray_ptds_v7000_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyFromArrayPtdsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyFromArrayPtdsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_from_array_ptds_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyFromArrayPtdsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

w_offset

int:

class cupti.cupti.CudaMemcpyArrayToArrayPtdsV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpyArrayToArray_ptds_v7000_params.

See also

cudaMemcpyArrayToArray_ptds_v7000_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyArrayToArrayPtdsV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyArrayToArrayPtdsV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_array_to_array_ptds_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyArrayToArrayPtdsV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset_dst

int:

h_offset_src

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

w_offset_dst

int:

w_offset_src

int:

class cupti.cupti.CudaMemcpyToArrayAsyncPtszV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpyToArrayAsync_ptsz_v7000_params.

See also

cudaMemcpyToArrayAsync_ptsz_v7000_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyToArrayAsyncPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyToArrayAsyncPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_to_array_async_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyToArrayAsyncPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

stream

int:

w_offset

int:

class cupti.cupti.CudaMemcpyFromArrayAsyncPtszV7000Params

Bases: object

Empty-initialize an instance of cudaMemcpyFromArrayAsync_ptsz_v7000_params.

See also

cudaMemcpyFromArrayAsync_ptsz_v7000_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyFromArrayAsyncPtszV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyFromArrayAsyncPtszV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_from_array_async_ptsz_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyFromArrayAsyncPtszV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

stream

int:

w_offset

int:

class cupti.cupti.CudaMallocAsyncPtszV11020Params

Bases: object

Empty-initialize an instance of cudaMallocAsync_ptsz_v11020_params.

See also

cudaMallocAsync_ptsz_v11020_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMallocAsyncPtszV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMallocAsyncPtszV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_malloc_async_ptsz_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMallocAsyncPtszV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CudaFreeAsyncPtszV11020Params

Bases: object

Empty-initialize an instance of cudaFreeAsync_ptsz_v11020_params.

See also

cudaFreeAsync_ptsz_v11020_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaFreeAsyncPtszV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaFreeAsyncPtszV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_free_async_ptsz_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaFreeAsyncPtszV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemPoolTrimToV11020Params

Bases: object

Empty-initialize an instance of cudaMemPoolTrimTo_v11020_params.

See also

cudaMemPoolTrimTo_v11020_params

static from_buffer(buffer)

Create an CudaMemPoolTrimToV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemPoolTrimToV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_pool_trim_to_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemPoolTrimToV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mem_pool

int:

min_bytes_to_keep

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemPoolSetAttributeV11020Params

Bases: object

Empty-initialize an instance of cudaMemPoolSetAttribute_v11020_params.

See also

cudaMemPoolSetAttribute_v11020_params

attr

:

static from_buffer(buffer)

Create an CudaMemPoolSetAttributeV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemPoolSetAttributeV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_pool_set_attribute_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemPoolSetAttributeV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CudaMemPoolGetAttributeV11020Params

Bases: object

Empty-initialize an instance of cudaMemPoolGetAttribute_v11020_params.

See also

cudaMemPoolGetAttribute_v11020_params

attr

:

static from_buffer(buffer)

Create an CudaMemPoolGetAttributeV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemPoolGetAttributeV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_pool_get_attribute_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemPoolGetAttributeV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CudaMemPoolSetAccessV11020Params

Bases: object

Empty-initialize an instance of cudaMemPoolSetAccess_v11020_params.

See also

cudaMemPoolSetAccess_v11020_params

count

int:

desc_list

int:

static from_buffer(buffer)

Create an CudaMemPoolSetAccessV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemPoolSetAccessV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_pool_set_access_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemPoolSetAccessV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemPoolGetAccessV11020Params

Bases: object

Empty-initialize an instance of cudaMemPoolGetAccess_v11020_params.

See also

cudaMemPoolGetAccess_v11020_params

flags_

int:

static from_buffer(buffer)

Create an CudaMemPoolGetAccessV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemPoolGetAccessV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_pool_get_access_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemPoolGetAccessV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

location

int:

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemPoolCreateV11020Params

Bases: object

Empty-initialize an instance of cudaMemPoolCreate_v11020_params.

See also

cudaMemPoolCreate_v11020_params

static from_buffer(buffer)

Create an CudaMemPoolCreateV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemPoolCreateV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_pool_create_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemPoolCreateV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mem_pool

int:

pool_props

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemPoolDestroyV11020Params

Bases: object

Empty-initialize an instance of cudaMemPoolDestroy_v11020_params.

See also

cudaMemPoolDestroy_v11020_params

static from_buffer(buffer)

Create an CudaMemPoolDestroyV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemPoolDestroyV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_pool_destroy_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemPoolDestroyV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMallocFromPoolAsyncPtszV11020Params

Bases: object

Empty-initialize an instance of cudaMallocFromPoolAsync_ptsz_v11020_params.

See also

cudaMallocFromPoolAsync_ptsz_v11020_params

static from_buffer(buffer)

Create an CudaMallocFromPoolAsyncPtszV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMallocFromPoolAsyncPtszV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_malloc_from_pool_async_ptsz_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMallocFromPoolAsyncPtszV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

ptr_

int:

size_

int:

stream

int:

class cupti.cupti.CudaMemPoolExportToShareableHandleV11020Params

Bases: object

Empty-initialize an instance of cudaMemPoolExportToShareableHandle_v11020_params.

See also

cudaMemPoolExportToShareableHandle_v11020_params

flags_

int:

static from_buffer(buffer)

Create an CudaMemPoolExportToShareableHandleV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemPoolExportToShareableHandleV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_pool_export_to_shareable_handle_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemPoolExportToShareableHandleV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle_type

:

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

shareable_handle

int:

class cupti.cupti.CudaMemPoolImportFromShareableHandleV11020Params

Bases: object

Empty-initialize an instance of cudaMemPoolImportFromShareableHandle_v11020_params.

See also

cudaMemPoolImportFromShareableHandle_v11020_params

flags_

int:

static from_buffer(buffer)

Create an CudaMemPoolImportFromShareableHandleV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemPoolImportFromShareableHandleV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_pool_import_from_shareable_handle_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemPoolImportFromShareableHandleV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle_type

:

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

shareable_handle

int:

class cupti.cupti.CudaMemPoolExportPointerV11020Params

Bases: object

Empty-initialize an instance of cudaMemPoolExportPointer_v11020_params.

See also

cudaMemPoolExportPointer_v11020_params

export_data

int:

static from_buffer(buffer)

Create an CudaMemPoolExportPointerV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemPoolExportPointerV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_pool_export_pointer_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemPoolExportPointerV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

ptr_

int:

class cupti.cupti.CudaMemPoolImportPointerV11020Params

Bases: object

Empty-initialize an instance of cudaMemPoolImportPointer_v11020_params.

See also

cudaMemPoolImportPointer_v11020_params

export_data

int:

static from_buffer(buffer)

Create an CudaMemPoolImportPointerV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemPoolImportPointerV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_pool_import_pointer_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemPoolImportPointerV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

ptr_

int:

class cupti.cupti.CudaPointerGetAttributesV4000Params

Bases: object

Empty-initialize an instance of cudaPointerGetAttributes_v4000_params.

See also

cudaPointerGetAttributes_v4000_params

attributes

int:

static from_buffer(buffer)

Create an CudaPointerGetAttributesV4000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaPointerGetAttributesV4000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_pointer_get_attributes_v4000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaPointerGetAttributesV4000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

ptr_

int:

class cupti.cupti.CudaDeviceCanAccessPeerV4000Params

Bases: object

Empty-initialize an instance of cudaDeviceCanAccessPeer_v4000_params.

See also

cudaDeviceCanAccessPeer_v4000_params

can_access_peer

int:

device_

int:

static from_buffer(buffer)

Create an CudaDeviceCanAccessPeerV4000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceCanAccessPeerV4000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_can_access_peer_v4000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceCanAccessPeerV4000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

peer_device

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceEnablePeerAccessV4000Params

Bases: object

Empty-initialize an instance of cudaDeviceEnablePeerAccess_v4000_params.

See also

cudaDeviceEnablePeerAccess_v4000_params

flags_

int:

static from_buffer(buffer)

Create an CudaDeviceEnablePeerAccessV4000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceEnablePeerAccessV4000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_enable_peer_access_v4000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceEnablePeerAccessV4000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

peer_device

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceDisablePeerAccessV4000Params

Bases: object

Empty-initialize an instance of cudaDeviceDisablePeerAccess_v4000_params.

See also

cudaDeviceDisablePeerAccess_v4000_params

static from_buffer(buffer)

Create an CudaDeviceDisablePeerAccessV4000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceDisablePeerAccessV4000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_disable_peer_access_v4000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceDisablePeerAccessV4000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

peer_device

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphicsUnregisterResourceV3020Params

Bases: object

Empty-initialize an instance of cudaGraphicsUnregisterResource_v3020_params.

See also

cudaGraphicsUnregisterResource_v3020_params

static from_buffer(buffer)

Create an CudaGraphicsUnregisterResourceV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphicsUnregisterResourceV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graphics_unregister_resource_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphicsUnregisterResourceV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

resource

int:

class cupti.cupti.CudaGraphicsResourceSetMapFlagsV3020Params

Bases: object

Empty-initialize an instance of cudaGraphicsResourceSetMapFlags_v3020_params.

See also

cudaGraphicsResourceSetMapFlags_v3020_params

flags_

int:

static from_buffer(buffer)

Create an CudaGraphicsResourceSetMapFlagsV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphicsResourceSetMapFlagsV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graphics_resource_set_map_flags_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphicsResourceSetMapFlagsV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

resource

int:

class cupti.cupti.CudaGraphicsMapResourcesV3020Params

Bases: object

Empty-initialize an instance of cudaGraphicsMapResources_v3020_params.

See also

cudaGraphicsMapResources_v3020_params

count

int:

static from_buffer(buffer)

Create an CudaGraphicsMapResourcesV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphicsMapResourcesV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graphics_map_resources_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphicsMapResourcesV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

resources

int:

stream

int:

class cupti.cupti.CudaGraphicsUnmapResourcesV3020Params

Bases: object

Empty-initialize an instance of cudaGraphicsUnmapResources_v3020_params.

See also

cudaGraphicsUnmapResources_v3020_params

count

int:

static from_buffer(buffer)

Create an CudaGraphicsUnmapResourcesV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphicsUnmapResourcesV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graphics_unmap_resources_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphicsUnmapResourcesV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

resources

int:

stream

int:

class cupti.cupti.CudaGraphicsResourceGetMappedPointerV3020Params

Bases: object

Empty-initialize an instance of cudaGraphicsResourceGetMappedPointer_v3020_params.

See also

cudaGraphicsResourceGetMappedPointer_v3020_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaGraphicsResourceGetMappedPointerV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphicsResourceGetMappedPointerV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graphics_resource_get_mapped_pointer_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphicsResourceGetMappedPointerV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

resource

int:

size_

int:

class cupti.cupti.CudaGraphicsSubResourceGetMappedArrayV3020Params

Bases: object

Empty-initialize an instance of cudaGraphicsSubResourceGetMappedArray_v3020_params.

See also

cudaGraphicsSubResourceGetMappedArray_v3020_params

array

int:

array_index

int:

static from_buffer(buffer)

Create an CudaGraphicsSubResourceGetMappedArrayV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphicsSubResourceGetMappedArrayV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graphics_sub_resource_get_mapped_array_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphicsSubResourceGetMappedArrayV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mip_level

int:

ptr

Get the pointer address to the data as Python int.

resource

int:

class cupti.cupti.CudaGraphicsResourceGetMappedMipmappedArrayV5000Params

Bases: object

Empty-initialize an instance of cudaGraphicsResourceGetMappedMipmappedArray_v5000_params.

See also

cudaGraphicsResourceGetMappedMipmappedArray_v5000_params

static from_buffer(

buffer,

)

Create an CudaGraphicsResourceGetMappedMipmappedArrayV5000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphicsResourceGetMappedMipmappedArrayV5000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graphics_resource_get_mapped_mipmapped_array_v5000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphicsResourceGetMappedMipmappedArrayV5000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mipmapped_array

int:

ptr

Get the pointer address to the data as Python int.

resource

int:

class cupti.cupti.CudaGetChannelDescV3020Params

Bases: object

Empty-initialize an instance of cudaGetChannelDesc_v3020_params.

See also

cudaGetChannelDesc_v3020_params

array

int:

desc

int:

static from_buffer(buffer)

Create an CudaGetChannelDescV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetChannelDescV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_channel_desc_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetChannelDescV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaCreateChannelDescV3020Params

Bases: object

Empty-initialize an instance of cudaCreateChannelDesc_v3020_params.

See also

cudaCreateChannelDesc_v3020_params

f

:

static from_buffer(buffer)

Create an CudaCreateChannelDescV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaCreateChannelDescV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_create_channel_desc_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaCreateChannelDescV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

w

int:

x

int:

y

int:

z

int:

class cupti.cupti.CudaCreateTextureObjectV5000Params

Bases: object

Empty-initialize an instance of cudaCreateTextureObject_v5000_params.

See also

cudaCreateTextureObject_v5000_params

static from_buffer(buffer)

Create an CudaCreateTextureObjectV5000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaCreateTextureObjectV5000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_create_texture_object_v5000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaCreateTextureObjectV5000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_res_desc

int:

p_res_view_desc

int:

p_tex_desc

int:

p_tex_object

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDestroyTextureObjectV5000Params

Bases: object

Empty-initialize an instance of cudaDestroyTextureObject_v5000_params.

See also

cudaDestroyTextureObject_v5000_params

static from_buffer(buffer)

Create an CudaDestroyTextureObjectV5000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDestroyTextureObjectV5000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_destroy_texture_object_v5000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDestroyTextureObjectV5000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

tex_object

:

class cupti.cupti.CudaGetTextureObjectResourceDescV5000Params

Bases: object

Empty-initialize an instance of cudaGetTextureObjectResourceDesc_v5000_params.

See also

cudaGetTextureObjectResourceDesc_v5000_params

static from_buffer(buffer)

Create an CudaGetTextureObjectResourceDescV5000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetTextureObjectResourceDescV5000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_texture_object_resource_desc_v5000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetTextureObjectResourceDescV5000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_res_desc

int:

ptr

Get the pointer address to the data as Python int.

tex_object

:

class cupti.cupti.CudaGetTextureObjectTextureDescV5000Params

Bases: object

Empty-initialize an instance of cudaGetTextureObjectTextureDesc_v5000_params.

See also

cudaGetTextureObjectTextureDesc_v5000_params

static from_buffer(buffer)

Create an CudaGetTextureObjectTextureDescV5000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetTextureObjectTextureDescV5000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_texture_object_texture_desc_v5000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetTextureObjectTextureDescV5000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_tex_desc

int:

ptr

Get the pointer address to the data as Python int.

tex_object

:

class cupti.cupti.CudaGetTextureObjectResourceViewDescV5000Params

Bases: object

Empty-initialize an instance of cudaGetTextureObjectResourceViewDesc_v5000_params.

See also

cudaGetTextureObjectResourceViewDesc_v5000_params

static from_buffer(buffer)

Create an CudaGetTextureObjectResourceViewDescV5000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetTextureObjectResourceViewDescV5000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_texture_object_resource_view_desc_v5000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetTextureObjectResourceViewDescV5000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_res_view_desc

int:

ptr

Get the pointer address to the data as Python int.

tex_object

:

class cupti.cupti.CudaCreateSurfaceObjectV5000Params

Bases: object

Empty-initialize an instance of cudaCreateSurfaceObject_v5000_params.

See also

cudaCreateSurfaceObject_v5000_params

static from_buffer(buffer)

Create an CudaCreateSurfaceObjectV5000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaCreateSurfaceObjectV5000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_create_surface_object_v5000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaCreateSurfaceObjectV5000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_res_desc

int:

p_surf_object

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDestroySurfaceObjectV5000Params

Bases: object

Empty-initialize an instance of cudaDestroySurfaceObject_v5000_params.

See also

cudaDestroySurfaceObject_v5000_params

static from_buffer(buffer)

Create an CudaDestroySurfaceObjectV5000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDestroySurfaceObjectV5000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_destroy_surface_object_v5000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDestroySurfaceObjectV5000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

surf_object

:

class cupti.cupti.CudaGetSurfaceObjectResourceDescV5000Params

Bases: object

Empty-initialize an instance of cudaGetSurfaceObjectResourceDesc_v5000_params.

See also

cudaGetSurfaceObjectResourceDesc_v5000_params

static from_buffer(buffer)

Create an CudaGetSurfaceObjectResourceDescV5000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetSurfaceObjectResourceDescV5000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_surface_object_resource_desc_v5000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetSurfaceObjectResourceDescV5000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_res_desc

int:

ptr

Get the pointer address to the data as Python int.

surf_object

:

class cupti.cupti.CudaDriverGetVersionV3020Params

Bases: object

Empty-initialize an instance of cudaDriverGetVersion_v3020_params.

See also

cudaDriverGetVersion_v3020_params

driver_version

int:

static from_buffer(buffer)

Create an CudaDriverGetVersionV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDriverGetVersionV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_driver_get_version_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDriverGetVersionV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaRuntimeGetVersionV3020Params

Bases: object

Empty-initialize an instance of cudaRuntimeGetVersion_v3020_params.

See also

cudaRuntimeGetVersion_v3020_params

static from_buffer(buffer)

Create an CudaRuntimeGetVersionV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaRuntimeGetVersionV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_runtime_get_version_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaRuntimeGetVersionV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

runtime_version

int:

class cupti.cupti.CudaGraphCreateV10000Params

Bases: object

Empty-initialize an instance of cudaGraphCreate_v10000_params.

See also

cudaGraphCreate_v10000_params

flags_

int:

static from_buffer(buffer)

Create an CudaGraphCreateV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphCreateV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_create_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphCreateV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_graph

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphAddKernelNodeV10000Params

Bases: object

Empty-initialize an instance of cudaGraphAddKernelNode_v10000_params.

See also

cudaGraphAddKernelNode_v10000_params

static from_buffer(buffer)

Create an CudaGraphAddKernelNodeV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddKernelNodeV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_kernel_node_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddKernelNodeV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

num_dependencies

int:

p_dependencies

int:

p_graph_node

int:

p_node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphKernelNodeGetParamsV10000Params

Bases: object

Empty-initialize an instance of cudaGraphKernelNodeGetParams_v10000_params.

See also

cudaGraphKernelNodeGetParams_v10000_params

static from_buffer(buffer)

Create an CudaGraphKernelNodeGetParamsV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphKernelNodeGetParamsV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_kernel_node_get_params_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphKernelNodeGetParamsV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

p_node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphKernelNodeSetParamsV10000Params

Bases: object

Empty-initialize an instance of cudaGraphKernelNodeSetParams_v10000_params.

See also

cudaGraphKernelNodeSetParams_v10000_params

static from_buffer(buffer)

Create an CudaGraphKernelNodeSetParamsV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphKernelNodeSetParamsV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_kernel_node_set_params_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphKernelNodeSetParamsV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

p_node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphKernelNodeCopyAttributesV11000Params

Bases: object

Empty-initialize an instance of cudaGraphKernelNodeCopyAttributes_v11000_params.

See also

cudaGraphKernelNodeCopyAttributes_v11000_params

static from_buffer(buffer)

Create an CudaGraphKernelNodeCopyAttributesV11000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphKernelNodeCopyAttributesV11000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_kernel_node_copy_attributes_v11000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphKernelNodeCopyAttributesV11000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_dst

int:

h_src

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphKernelNodeGetAttributeV11000Params

Bases: object

Empty-initialize an instance of cudaGraphKernelNodeGetAttribute_v11000_params.

See also

cudaGraphKernelNodeGetAttribute_v11000_params

attr

:

static from_buffer(buffer)

Create an CudaGraphKernelNodeGetAttributeV11000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphKernelNodeGetAttributeV11000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_kernel_node_get_attribute_v11000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphKernelNodeGetAttributeV11000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ptr

Get the pointer address to the data as Python int.

value_out

int:

class cupti.cupti.CudaGraphKernelNodeSetAttributeV11000Params

Bases: object

Empty-initialize an instance of cudaGraphKernelNodeSetAttribute_v11000_params.

See also

cudaGraphKernelNodeSetAttribute_v11000_params

attr

:

static from_buffer(buffer)

Create an CudaGraphKernelNodeSetAttributeV11000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphKernelNodeSetAttributeV11000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_kernel_node_set_attribute_v11000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphKernelNodeSetAttributeV11000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CudaGraphAddMemcpyNodeV10000Params

Bases: object

Empty-initialize an instance of cudaGraphAddMemcpyNode_v10000_params.

See also

cudaGraphAddMemcpyNode_v10000_params

static from_buffer(buffer)

Create an CudaGraphAddMemcpyNodeV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddMemcpyNodeV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_memcpy_node_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddMemcpyNodeV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

num_dependencies

int:

p_copy_params

int:

p_dependencies

int:

p_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphAddMemcpyNodeToSymbolV11010Params

Bases: object

Empty-initialize an instance of cudaGraphAddMemcpyNodeToSymbol_v11010_params.

See also

cudaGraphAddMemcpyNodeToSymbol_v11010_params

count

int:

static from_buffer(buffer)

Create an CudaGraphAddMemcpyNodeToSymbolV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddMemcpyNodeToSymbolV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_memcpy_node_to_symbol_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddMemcpyNodeToSymbolV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

kind

:

num_dependencies

int:

offset

int:

p_dependencies

int:

p_graph_node

int:

ptr

Get the pointer address to the data as Python int.

src

int:

symbol

int:

class cupti.cupti.CudaGraphAddMemcpyNodeFromSymbolV11010Params

Bases: object

Empty-initialize an instance of cudaGraphAddMemcpyNodeFromSymbol_v11010_params.

See also

cudaGraphAddMemcpyNodeFromSymbol_v11010_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaGraphAddMemcpyNodeFromSymbolV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddMemcpyNodeFromSymbolV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_memcpy_node_from_symbol_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddMemcpyNodeFromSymbolV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

kind

:

num_dependencies

int:

offset

int:

p_dependencies

int:

p_graph_node

int:

ptr

Get the pointer address to the data as Python int.

symbol

int:

class cupti.cupti.CudaGraphAddMemcpyNode1DV11010Params

Bases: object

Empty-initialize an instance of cudaGraphAddMemcpyNode1D_v11010_params.

See also

cudaGraphAddMemcpyNode1D_v11010_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaGraphAddMemcpyNode1DV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddMemcpyNode1DV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_memcpy_node1d_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddMemcpyNode1DV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

kind

:

num_dependencies

int:

p_dependencies

int:

p_graph_node

int:

ptr

Get the pointer address to the data as Python int.

src

int:

class cupti.cupti.CudaGraphMemcpyNodeGetParamsV10000Params

Bases: object

Empty-initialize an instance of cudaGraphMemcpyNodeGetParams_v10000_params.

See also

cudaGraphMemcpyNodeGetParams_v10000_params

static from_buffer(buffer)

Create an CudaGraphMemcpyNodeGetParamsV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphMemcpyNodeGetParamsV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_memcpy_node_get_params_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphMemcpyNodeGetParamsV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

p_node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphMemcpyNodeSetParamsV10000Params

Bases: object

Empty-initialize an instance of cudaGraphMemcpyNodeSetParams_v10000_params.

See also

cudaGraphMemcpyNodeSetParams_v10000_params

static from_buffer(buffer)

Create an CudaGraphMemcpyNodeSetParamsV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphMemcpyNodeSetParamsV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_memcpy_node_set_params_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphMemcpyNodeSetParamsV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

p_node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphMemcpyNodeSetParamsToSymbolV11010Params

Bases: object

Empty-initialize an instance of cudaGraphMemcpyNodeSetParamsToSymbol_v11010_params.

See also

cudaGraphMemcpyNodeSetParamsToSymbol_v11010_params

count

int:

static from_buffer(buffer)

Create an CudaGraphMemcpyNodeSetParamsToSymbolV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphMemcpyNodeSetParamsToSymbolV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_memcpy_node_set_params_to_symbol_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphMemcpyNodeSetParamsToSymbolV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

:

node

int:

offset

int:

ptr

Get the pointer address to the data as Python int.

src

int:

symbol

int:

class cupti.cupti.CudaGraphMemcpyNodeSetParamsFromSymbolV11010Params

Bases: object

Empty-initialize an instance of cudaGraphMemcpyNodeSetParamsFromSymbol_v11010_params.

See also

cudaGraphMemcpyNodeSetParamsFromSymbol_v11010_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaGraphMemcpyNodeSetParamsFromSymbolV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphMemcpyNodeSetParamsFromSymbolV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_memcpy_node_set_params_from_symbol_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphMemcpyNodeSetParamsFromSymbolV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

:

node

int:

offset

int:

ptr

Get the pointer address to the data as Python int.

symbol

int:

class cupti.cupti.CudaGraphMemcpyNodeSetParams1DV11010Params

Bases: object

Empty-initialize an instance of cudaGraphMemcpyNodeSetParams1D_v11010_params.

See also

cudaGraphMemcpyNodeSetParams1D_v11010_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaGraphMemcpyNodeSetParams1DV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphMemcpyNodeSetParams1DV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_memcpy_node_set_params1d_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphMemcpyNodeSetParams1DV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

:

node

int:

ptr

Get the pointer address to the data as Python int.

src

int:

class cupti.cupti.CudaGraphAddMemsetNodeV10000Params

Bases: object

Empty-initialize an instance of cudaGraphAddMemsetNode_v10000_params.

See also

cudaGraphAddMemsetNode_v10000_params

static from_buffer(buffer)

Create an CudaGraphAddMemsetNodeV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddMemsetNodeV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_memset_node_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddMemsetNodeV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

num_dependencies

int:

p_dependencies

int:

p_graph_node

int:

p_memset_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphMemsetNodeGetParamsV10000Params

Bases: object

Empty-initialize an instance of cudaGraphMemsetNodeGetParams_v10000_params.

See also

cudaGraphMemsetNodeGetParams_v10000_params

static from_buffer(buffer)

Create an CudaGraphMemsetNodeGetParamsV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphMemsetNodeGetParamsV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_memset_node_get_params_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphMemsetNodeGetParamsV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

p_node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphMemsetNodeSetParamsV10000Params

Bases: object

Empty-initialize an instance of cudaGraphMemsetNodeSetParams_v10000_params.

See also

cudaGraphMemsetNodeSetParams_v10000_params

static from_buffer(buffer)

Create an CudaGraphMemsetNodeSetParamsV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphMemsetNodeSetParamsV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_memset_node_set_params_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphMemsetNodeSetParamsV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

p_node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphAddHostNodeV10000Params

Bases: object

Empty-initialize an instance of cudaGraphAddHostNode_v10000_params.

See also

cudaGraphAddHostNode_v10000_params

static from_buffer(buffer)

Create an CudaGraphAddHostNodeV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddHostNodeV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_host_node_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddHostNodeV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

num_dependencies

int:

p_dependencies

int:

p_graph_node

int:

p_node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphHostNodeGetParamsV10000Params

Bases: object

Empty-initialize an instance of cudaGraphHostNodeGetParams_v10000_params.

See also

cudaGraphHostNodeGetParams_v10000_params

static from_buffer(buffer)

Create an CudaGraphHostNodeGetParamsV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphHostNodeGetParamsV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_host_node_get_params_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphHostNodeGetParamsV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

p_node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphHostNodeSetParamsV10000Params

Bases: object

Empty-initialize an instance of cudaGraphHostNodeSetParams_v10000_params.

See also

cudaGraphHostNodeSetParams_v10000_params

static from_buffer(buffer)

Create an CudaGraphHostNodeSetParamsV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphHostNodeSetParamsV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_host_node_set_params_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphHostNodeSetParamsV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

p_node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphAddChildGraphNodeV10000Params

Bases: object

Empty-initialize an instance of cudaGraphAddChildGraphNode_v10000_params.

See also

cudaGraphAddChildGraphNode_v10000_params

child_graph

int:

static from_buffer(buffer)

Create an CudaGraphAddChildGraphNodeV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddChildGraphNodeV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_child_graph_node_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddChildGraphNodeV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

num_dependencies

int:

p_dependencies

int:

p_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphChildGraphNodeGetGraphV10000Params

Bases: object

Empty-initialize an instance of cudaGraphChildGraphNodeGetGraph_v10000_params.

See also

cudaGraphChildGraphNodeGetGraph_v10000_params

static from_buffer(buffer)

Create an CudaGraphChildGraphNodeGetGraphV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphChildGraphNodeGetGraphV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_child_graph_node_get_graph_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphChildGraphNodeGetGraphV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

p_graph

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphAddEmptyNodeV10000Params

Bases: object

Empty-initialize an instance of cudaGraphAddEmptyNode_v10000_params.

See also

cudaGraphAddEmptyNode_v10000_params

static from_buffer(buffer)

Create an CudaGraphAddEmptyNodeV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddEmptyNodeV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_empty_node_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddEmptyNodeV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

num_dependencies

int:

p_dependencies

int:

p_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphAddEventRecordNodeV11010Params

Bases: object

Empty-initialize an instance of cudaGraphAddEventRecordNode_v11010_params.

See also

cudaGraphAddEventRecordNode_v11010_params

event

int:

static from_buffer(buffer)

Create an CudaGraphAddEventRecordNodeV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddEventRecordNodeV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_event_record_node_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddEventRecordNodeV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

num_dependencies

int:

p_dependencies

int:

p_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphEventRecordNodeGetEventV11010Params

Bases: object

Empty-initialize an instance of cudaGraphEventRecordNodeGetEvent_v11010_params.

See also

cudaGraphEventRecordNodeGetEvent_v11010_params

event_out

int:

static from_buffer(buffer)

Create an CudaGraphEventRecordNodeGetEventV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphEventRecordNodeGetEventV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_event_record_node_get_event_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphEventRecordNodeGetEventV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphEventRecordNodeSetEventV11010Params

Bases: object

Empty-initialize an instance of cudaGraphEventRecordNodeSetEvent_v11010_params.

See also

cudaGraphEventRecordNodeSetEvent_v11010_params

event

int:

static from_buffer(buffer)

Create an CudaGraphEventRecordNodeSetEventV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphEventRecordNodeSetEventV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_event_record_node_set_event_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphEventRecordNodeSetEventV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphAddEventWaitNodeV11010Params

Bases: object

Empty-initialize an instance of cudaGraphAddEventWaitNode_v11010_params.

See also

cudaGraphAddEventWaitNode_v11010_params

event

int:

static from_buffer(buffer)

Create an CudaGraphAddEventWaitNodeV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddEventWaitNodeV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_event_wait_node_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddEventWaitNodeV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

num_dependencies

int:

p_dependencies

int:

p_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphEventWaitNodeGetEventV11010Params

Bases: object

Empty-initialize an instance of cudaGraphEventWaitNodeGetEvent_v11010_params.

See also

cudaGraphEventWaitNodeGetEvent_v11010_params

event_out

int:

static from_buffer(buffer)

Create an CudaGraphEventWaitNodeGetEventV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphEventWaitNodeGetEventV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_event_wait_node_get_event_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphEventWaitNodeGetEventV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphEventWaitNodeSetEventV11010Params

Bases: object

Empty-initialize an instance of cudaGraphEventWaitNodeSetEvent_v11010_params.

See also

cudaGraphEventWaitNodeSetEvent_v11010_params

event

int:

static from_buffer(buffer)

Create an CudaGraphEventWaitNodeSetEventV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphEventWaitNodeSetEventV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_event_wait_node_set_event_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphEventWaitNodeSetEventV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphAddExternalSemaphoresSignalNodeV11020Params

Bases: object

Empty-initialize an instance of cudaGraphAddExternalSemaphoresSignalNode_v11020_params.

See also

cudaGraphAddExternalSemaphoresSignalNode_v11020_params

static from_buffer(

buffer,

)

Create an CudaGraphAddExternalSemaphoresSignalNodeV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddExternalSemaphoresSignalNodeV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_external_semaphores_signal_node_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddExternalSemaphoresSignalNodeV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

node_params

int:

num_dependencies

int:

p_dependencies

int:

p_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExternalSemaphoresSignalNodeGetParamsV11020Params

Bases: object

Empty-initialize an instance of cudaGraphExternalSemaphoresSignalNodeGetParams_v11020_params.

See also

cudaGraphExternalSemaphoresSignalNodeGetParams_v11020_params

static from_buffer(

buffer,

)

Create an CudaGraphExternalSemaphoresSignalNodeGetParamsV11020Params instance with the memory from the given buffer.

static from_data(

data,

)

Create an CudaGraphExternalSemaphoresSignalNodeGetParamsV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_external_semaphores_signal_node_get_params_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExternalSemaphoresSignalNodeGetParamsV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

params_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExternalSemaphoresSignalNodeSetParamsV11020Params

Bases: object

Empty-initialize an instance of cudaGraphExternalSemaphoresSignalNodeSetParams_v11020_params.

See also

cudaGraphExternalSemaphoresSignalNodeSetParams_v11020_params

static from_buffer(

buffer,

)

Create an CudaGraphExternalSemaphoresSignalNodeSetParamsV11020Params instance with the memory from the given buffer.

static from_data(

data,

)

Create an CudaGraphExternalSemaphoresSignalNodeSetParamsV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_external_semaphores_signal_node_set_params_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExternalSemaphoresSignalNodeSetParamsV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphAddExternalSemaphoresWaitNodeV11020Params

Bases: object

Empty-initialize an instance of cudaGraphAddExternalSemaphoresWaitNode_v11020_params.

See also

cudaGraphAddExternalSemaphoresWaitNode_v11020_params

static from_buffer(buffer)

Create an CudaGraphAddExternalSemaphoresWaitNodeV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddExternalSemaphoresWaitNodeV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_external_semaphores_wait_node_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddExternalSemaphoresWaitNodeV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

node_params

int:

num_dependencies

int:

p_dependencies

int:

p_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExternalSemaphoresWaitNodeGetParamsV11020Params

Bases: object

Empty-initialize an instance of cudaGraphExternalSemaphoresWaitNodeGetParams_v11020_params.

See also

cudaGraphExternalSemaphoresWaitNodeGetParams_v11020_params

static from_buffer(

buffer,

)

Create an CudaGraphExternalSemaphoresWaitNodeGetParamsV11020Params instance with the memory from the given buffer.

static from_data(

data,

)

Create an CudaGraphExternalSemaphoresWaitNodeGetParamsV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_external_semaphores_wait_node_get_params_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExternalSemaphoresWaitNodeGetParamsV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

params_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExternalSemaphoresWaitNodeSetParamsV11020Params

Bases: object

Empty-initialize an instance of cudaGraphExternalSemaphoresWaitNodeSetParams_v11020_params.

See also

cudaGraphExternalSemaphoresWaitNodeSetParams_v11020_params

static from_buffer(

buffer,

)

Create an CudaGraphExternalSemaphoresWaitNodeSetParamsV11020Params instance with the memory from the given buffer.

static from_data(

data,

)

Create an CudaGraphExternalSemaphoresWaitNodeSetParamsV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_external_semaphores_wait_node_set_params_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExternalSemaphoresWaitNodeSetParamsV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphAddMemAllocNodeV11040Params

Bases: object

Empty-initialize an instance of cudaGraphAddMemAllocNode_v11040_params.

See also

cudaGraphAddMemAllocNode_v11040_params

static from_buffer(buffer)

Create an CudaGraphAddMemAllocNodeV11040Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddMemAllocNodeV11040Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_mem_alloc_node_v11040params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddMemAllocNodeV11040Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

node_params

int:

num_dependencies

int:

p_dependencies

int:

p_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphMemAllocNodeGetParamsV11040Params

Bases: object

Empty-initialize an instance of cudaGraphMemAllocNodeGetParams_v11040_params.

See also

cudaGraphMemAllocNodeGetParams_v11040_params

static from_buffer(buffer)

Create an CudaGraphMemAllocNodeGetParamsV11040Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphMemAllocNodeGetParamsV11040Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_mem_alloc_node_get_params_v11040params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphMemAllocNodeGetParamsV11040Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

params_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphAddMemFreeNodeV11040Params

Bases: object

Empty-initialize an instance of cudaGraphAddMemFreeNode_v11040_params.

See also

cudaGraphAddMemFreeNode_v11040_params

dptr

int:

static from_buffer(buffer)

Create an CudaGraphAddMemFreeNodeV11040Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddMemFreeNodeV11040Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_mem_free_node_v11040params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddMemFreeNodeV11040Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

num_dependencies

int:

p_dependencies

int:

p_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphMemFreeNodeGetParamsV11040Params

Bases: object

Empty-initialize an instance of cudaGraphMemFreeNodeGetParams_v11040_params.

See also

cudaGraphMemFreeNodeGetParams_v11040_params

dptr_out

int:

static from_buffer(buffer)

Create an CudaGraphMemFreeNodeGetParamsV11040Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphMemFreeNodeGetParamsV11040Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_mem_free_node_get_params_v11040params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphMemFreeNodeGetParamsV11040Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGraphMemTrimV11040Params

Bases: object

Empty-initialize an instance of cudaDeviceGraphMemTrim_v11040_params.

See also

cudaDeviceGraphMemTrim_v11040_params

device_

int:

static from_buffer(buffer)

Create an CudaDeviceGraphMemTrimV11040Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGraphMemTrimV11040Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_graph_mem_trim_v11040params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGraphMemTrimV11040Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGetGraphMemAttributeV11040Params

Bases: object

Empty-initialize an instance of cudaDeviceGetGraphMemAttribute_v11040_params.

See also

cudaDeviceGetGraphMemAttribute_v11040_params

attr

:

device_

int:

static from_buffer(buffer)

Create an CudaDeviceGetGraphMemAttributeV11040Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetGraphMemAttributeV11040Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_graph_mem_attribute_v11040params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetGraphMemAttributeV11040Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CudaDeviceSetGraphMemAttributeV11040Params

Bases: object

Empty-initialize an instance of cudaDeviceSetGraphMemAttribute_v11040_params.

See also

cudaDeviceSetGraphMemAttribute_v11040_params

attr

:

device_

int:

static from_buffer(buffer)

Create an CudaDeviceSetGraphMemAttributeV11040Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceSetGraphMemAttributeV11040Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_set_graph_mem_attribute_v11040params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceSetGraphMemAttributeV11040Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CudaGraphCloneV10000Params

Bases: object

Empty-initialize an instance of cudaGraphClone_v10000_params.

See also

cudaGraphClone_v10000_params

static from_buffer(buffer)

Create an CudaGraphCloneV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphCloneV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_clone_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphCloneV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

original_graph

int:

p_graph_clone

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphNodeFindInCloneV10000Params

Bases: object

Empty-initialize an instance of cudaGraphNodeFindInClone_v10000_params.

See also

cudaGraphNodeFindInClone_v10000_params

cloned_graph

int:

static from_buffer(buffer)

Create an CudaGraphNodeFindInCloneV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphNodeFindInCloneV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_node_find_in_clone_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphNodeFindInCloneV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

original_node

int:

p_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphNodeGetTypeV10000Params

Bases: object

Empty-initialize an instance of cudaGraphNodeGetType_v10000_params.

See also

cudaGraphNodeGetType_v10000_params

static from_buffer(buffer)

Create an CudaGraphNodeGetTypeV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphNodeGetTypeV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_node_get_type_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphNodeGetTypeV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

p_type

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphGetNodesV10000Params

Bases: object

Empty-initialize an instance of cudaGraphGetNodes_v10000_params.

See also

cudaGraphGetNodes_v10000_params

static from_buffer(buffer)

Create an CudaGraphGetNodesV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphGetNodesV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_get_nodes_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphGetNodesV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

nodes

int:

num_nodes

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphGetRootNodesV10000Params

Bases: object

Empty-initialize an instance of cudaGraphGetRootNodes_v10000_params.

See also

cudaGraphGetRootNodes_v10000_params

static from_buffer(buffer)

Create an CudaGraphGetRootNodesV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphGetRootNodesV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_get_root_nodes_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphGetRootNodesV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

p_num_root_nodes

int:

p_root_nodes

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphDestroyNodeV10000Params

Bases: object

Empty-initialize an instance of cudaGraphDestroyNode_v10000_params.

See also

cudaGraphDestroyNode_v10000_params

static from_buffer(buffer)

Create an CudaGraphDestroyNodeV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphDestroyNodeV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_destroy_node_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphDestroyNodeV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphInstantiateV12000Params

Bases: object

Empty-initialize an instance of cudaGraphInstantiate_v12000_params.

See also

cudaGraphInstantiate_v12000_params

flags_

int:

static from_buffer(buffer)

Create an CudaGraphInstantiateV12000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphInstantiateV12000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_instantiate_v12000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphInstantiateV12000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

p_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphInstantiateWithFlagsV11040Params

Bases: object

Empty-initialize an instance of cudaGraphInstantiateWithFlags_v11040_params.

See also

cudaGraphInstantiateWithFlags_v11040_params

flags_

int:

static from_buffer(buffer)

Create an CudaGraphInstantiateWithFlagsV11040Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphInstantiateWithFlagsV11040Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_instantiate_with_flags_v11040params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphInstantiateWithFlagsV11040Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

p_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphInstantiateWithParamsPtszV12000Params

Bases: object

Empty-initialize an instance of cudaGraphInstantiateWithParams_ptsz_v12000_params.

See also

cudaGraphInstantiateWithParams_ptsz_v12000_params

static from_buffer(buffer)

Create an CudaGraphInstantiateWithParamsPtszV12000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphInstantiateWithParamsPtszV12000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_instantiate_with_params_ptsz_v12000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphInstantiateWithParamsPtszV12000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

instantiate_params

int:

p_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExecGetFlagsV12000Params

Bases: object

Empty-initialize an instance of cudaGraphExecGetFlags_v12000_params.

See also

cudaGraphExecGetFlags_v12000_params

flags_

int:

static from_buffer(buffer)

Create an CudaGraphExecGetFlagsV12000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphExecGetFlagsV12000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_get_flags_v12000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecGetFlagsV12000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_exec

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExecKernelNodeSetParamsV10010Params

Bases: object

Empty-initialize an instance of cudaGraphExecKernelNodeSetParams_v10010_params.

See also

cudaGraphExecKernelNodeSetParams_v10010_params

static from_buffer(buffer)

Create an CudaGraphExecKernelNodeSetParamsV10010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphExecKernelNodeSetParamsV10010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_kernel_node_set_params_v10010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecKernelNodeSetParamsV10010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

node

int:

p_node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExecMemcpyNodeSetParamsV10020Params

Bases: object

Empty-initialize an instance of cudaGraphExecMemcpyNodeSetParams_v10020_params.

See also

cudaGraphExecMemcpyNodeSetParams_v10020_params

static from_buffer(buffer)

Create an CudaGraphExecMemcpyNodeSetParamsV10020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphExecMemcpyNodeSetParamsV10020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_memcpy_node_set_params_v10020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecMemcpyNodeSetParamsV10020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

node

int:

p_node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExecMemcpyNodeSetParamsToSymbolV11010Params

Bases: object

Empty-initialize an instance of cudaGraphExecMemcpyNodeSetParamsToSymbol_v11010_params.

See also

cudaGraphExecMemcpyNodeSetParamsToSymbol_v11010_params

count

int:

static from_buffer(

buffer,

)

Create an CudaGraphExecMemcpyNodeSetParamsToSymbolV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphExecMemcpyNodeSetParamsToSymbolV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_memcpy_node_set_params_to_symbol_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecMemcpyNodeSetParamsToSymbolV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

kind

:

node

int:

offset

int:

ptr

Get the pointer address to the data as Python int.

src

int:

symbol

int:

class cupti.cupti.CudaGraphExecMemcpyNodeSetParamsFromSymbolV11010Params

Bases: object

Empty-initialize an instance of cudaGraphExecMemcpyNodeSetParamsFromSymbol_v11010_params.

See also

cudaGraphExecMemcpyNodeSetParamsFromSymbol_v11010_params

count

int:

dst

int:

static from_buffer(

buffer,

)

Create an CudaGraphExecMemcpyNodeSetParamsFromSymbolV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphExecMemcpyNodeSetParamsFromSymbolV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_memcpy_node_set_params_from_symbol_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecMemcpyNodeSetParamsFromSymbolV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

kind

:

node

int:

offset

int:

ptr

Get the pointer address to the data as Python int.

symbol

int:

class cupti.cupti.CudaGraphExecMemcpyNodeSetParams1DV11010Params

Bases: object

Empty-initialize an instance of cudaGraphExecMemcpyNodeSetParams1D_v11010_params.

See also

cudaGraphExecMemcpyNodeSetParams1D_v11010_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaGraphExecMemcpyNodeSetParams1DV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphExecMemcpyNodeSetParams1DV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_memcpy_node_set_params1d_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecMemcpyNodeSetParams1DV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

kind

:

node

int:

ptr

Get the pointer address to the data as Python int.

src

int:

class cupti.cupti.CudaGraphExecMemsetNodeSetParamsV10020Params

Bases: object

Empty-initialize an instance of cudaGraphExecMemsetNodeSetParams_v10020_params.

See also

cudaGraphExecMemsetNodeSetParams_v10020_params

static from_buffer(buffer)

Create an CudaGraphExecMemsetNodeSetParamsV10020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphExecMemsetNodeSetParamsV10020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_memset_node_set_params_v10020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecMemsetNodeSetParamsV10020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

node

int:

p_node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExecHostNodeSetParamsV10020Params

Bases: object

Empty-initialize an instance of cudaGraphExecHostNodeSetParams_v10020_params.

See also

cudaGraphExecHostNodeSetParams_v10020_params

static from_buffer(buffer)

Create an CudaGraphExecHostNodeSetParamsV10020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphExecHostNodeSetParamsV10020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_host_node_set_params_v10020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecHostNodeSetParamsV10020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

node

int:

p_node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExecChildGraphNodeSetParamsV11010Params

Bases: object

Empty-initialize an instance of cudaGraphExecChildGraphNodeSetParams_v11010_params.

See also

cudaGraphExecChildGraphNodeSetParams_v11010_params

child_graph

int:

static from_buffer(buffer)

Create an CudaGraphExecChildGraphNodeSetParamsV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphExecChildGraphNodeSetParamsV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_child_graph_node_set_params_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecChildGraphNodeSetParamsV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExecEventRecordNodeSetEventV11010Params

Bases: object

Empty-initialize an instance of cudaGraphExecEventRecordNodeSetEvent_v11010_params.

See also

cudaGraphExecEventRecordNodeSetEvent_v11010_params

event

int:

static from_buffer(buffer)

Create an CudaGraphExecEventRecordNodeSetEventV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphExecEventRecordNodeSetEventV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_event_record_node_set_event_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecEventRecordNodeSetEventV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExecEventWaitNodeSetEventV11010Params

Bases: object

Empty-initialize an instance of cudaGraphExecEventWaitNodeSetEvent_v11010_params.

See also

cudaGraphExecEventWaitNodeSetEvent_v11010_params

event

int:

static from_buffer(buffer)

Create an CudaGraphExecEventWaitNodeSetEventV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphExecEventWaitNodeSetEventV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_event_wait_node_set_event_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecEventWaitNodeSetEventV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExecExternalSemaphoresSignalNodeSetParamsV11020Params

Bases: object

Empty-initialize an instance of cudaGraphExecExternalSemaphoresSignalNodeSetParams_v11020_params.

See also

cudaGraphExecExternalSemaphoresSignalNodeSetParams_v11020_params

static from_buffer(

buffer,

)

Create an CudaGraphExecExternalSemaphoresSignalNodeSetParamsV11020Params instance with the memory from the given buffer.

static from_data(

data,

)

Create an CudaGraphExecExternalSemaphoresSignalNodeSetParamsV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_external_semaphores_signal_node_set_params_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecExternalSemaphoresSignalNodeSetParamsV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExecExternalSemaphoresWaitNodeSetParamsV11020Params

Bases: object

Empty-initialize an instance of cudaGraphExecExternalSemaphoresWaitNodeSetParams_v11020_params.

See also

cudaGraphExecExternalSemaphoresWaitNodeSetParams_v11020_params

static from_buffer(

buffer,

)

Create an CudaGraphExecExternalSemaphoresWaitNodeSetParamsV11020Params instance with the memory from the given buffer.

static from_data(

data,

)

Create an CudaGraphExecExternalSemaphoresWaitNodeSetParamsV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_external_semaphores_wait_node_set_params_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecExternalSemaphoresWaitNodeSetParamsV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphNodeSetEnabledV11060Params

Bases: object

Empty-initialize an instance of cudaGraphNodeSetEnabled_v11060_params.

See also

cudaGraphNodeSetEnabled_v11060_params

static from_buffer(buffer)

Create an CudaGraphNodeSetEnabledV11060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphNodeSetEnabledV11060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_node_set_enabled_v11060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphNodeSetEnabledV11060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

is_enabled

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphNodeGetEnabledV11060Params

Bases: object

Empty-initialize an instance of cudaGraphNodeGetEnabled_v11060_params.

See also

cudaGraphNodeGetEnabled_v11060_params

static from_buffer(buffer)

Create an CudaGraphNodeGetEnabledV11060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphNodeGetEnabledV11060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_node_get_enabled_v11060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphNodeGetEnabledV11060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

is_enabled

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExecUpdateV10020Params

Bases: object

Empty-initialize an instance of cudaGraphExecUpdate_v10020_params.

See also

cudaGraphExecUpdate_v10020_params

static from_buffer(buffer)

Create an CudaGraphExecUpdateV10020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphExecUpdateV10020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_update_v10020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecUpdateV10020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

h_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

result_info

int:

class cupti.cupti.CudaGraphUploadPtszV10000Params

Bases: object

Empty-initialize an instance of cudaGraphUpload_ptsz_v10000_params.

See also

cudaGraphUpload_ptsz_v10000_params

static from_buffer(buffer)

Create an CudaGraphUploadPtszV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphUploadPtszV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_upload_ptsz_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphUploadPtszV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_exec

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaGraphLaunchPtszV10000Params

Bases: object

Empty-initialize an instance of cudaGraphLaunch_ptsz_v10000_params.

See also

cudaGraphLaunch_ptsz_v10000_params

static from_buffer(buffer)

Create an CudaGraphLaunchPtszV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphLaunchPtszV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_launch_ptsz_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphLaunchPtszV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_exec

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaGraphExecDestroyV10000Params

Bases: object

Empty-initialize an instance of cudaGraphExecDestroy_v10000_params.

See also

cudaGraphExecDestroy_v10000_params

static from_buffer(buffer)

Create an CudaGraphExecDestroyV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphExecDestroyV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_destroy_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecDestroyV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_exec

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphDestroyV10000Params

Bases: object

Empty-initialize an instance of cudaGraphDestroy_v10000_params.

See also

cudaGraphDestroy_v10000_params

static from_buffer(buffer)

Create an CudaGraphDestroyV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphDestroyV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_destroy_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphDestroyV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphDebugDotPrintV11030Params

Bases: object

Empty-initialize an instance of cudaGraphDebugDotPrint_v11030_params.

See also

cudaGraphDebugDotPrint_v11030_params

flags_

int:

static from_buffer(buffer)

Create an CudaGraphDebugDotPrintV11030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphDebugDotPrintV11030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_debug_dot_print_v11030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphDebugDotPrintV11030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

path

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaUserObjectCreateV11030Params

Bases: object

Empty-initialize an instance of cudaUserObjectCreate_v11030_params.

See also

cudaUserObjectCreate_v11030_params

destroy

int:

flags_

int:

static from_buffer(buffer)

Create an CudaUserObjectCreateV11030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaUserObjectCreateV11030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_user_object_create_v11030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaUserObjectCreateV11030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

initial_refcount

int:

object_out

int:

ptr

Get the pointer address to the data as Python int.

ptr_

int:

class cupti.cupti.CudaUserObjectRetainV11030Params

Bases: object

Empty-initialize an instance of cudaUserObjectRetain_v11030_params.

See also

cudaUserObjectRetain_v11030_params

count

int:

static from_buffer(buffer)

Create an CudaUserObjectRetainV11030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaUserObjectRetainV11030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_user_object_retain_v11030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaUserObjectRetainV11030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

object

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaUserObjectReleaseV11030Params

Bases: object

Empty-initialize an instance of cudaUserObjectRelease_v11030_params.

See also

cudaUserObjectRelease_v11030_params

count

int:

static from_buffer(buffer)

Create an CudaUserObjectReleaseV11030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaUserObjectReleaseV11030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_user_object_release_v11030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaUserObjectReleaseV11030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

object

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphRetainUserObjectV11030Params

Bases: object

Empty-initialize an instance of cudaGraphRetainUserObject_v11030_params.

See also

cudaGraphRetainUserObject_v11030_params

count

int:

flags_

int:

static from_buffer(buffer)

Create an CudaGraphRetainUserObjectV11030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphRetainUserObjectV11030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_retain_user_object_v11030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphRetainUserObjectV11030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

object

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphReleaseUserObjectV11030Params

Bases: object

Empty-initialize an instance of cudaGraphReleaseUserObject_v11030_params.

See also

cudaGraphReleaseUserObject_v11030_params

count

int:

static from_buffer(buffer)

Create an CudaGraphReleaseUserObjectV11030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphReleaseUserObjectV11030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_release_user_object_v11030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphReleaseUserObjectV11030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

object

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphNodeSetParamsV12020Params

Bases: object

Empty-initialize an instance of cudaGraphNodeSetParams_v12020_params.

See also

cudaGraphNodeSetParams_v12020_params

static from_buffer(buffer)

Create an CudaGraphNodeSetParamsV12020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphNodeSetParamsV12020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_node_set_params_v12020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphNodeSetParamsV12020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExecNodeSetParamsV12020Params

Bases: object

Empty-initialize an instance of cudaGraphExecNodeSetParams_v12020_params.

See also

cudaGraphExecNodeSetParams_v12020_params

static from_buffer(buffer)

Create an CudaGraphExecNodeSetParamsV12020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphExecNodeSetParamsV12020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_node_set_params_v12020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecNodeSetParamsV12020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_exec

int:

node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphConditionalHandleCreateV12030Params

Bases: object

Empty-initialize an instance of cudaGraphConditionalHandleCreate_v12030_params.

See also

cudaGraphConditionalHandleCreate_v12030_params

default_launch_value

int:

flags_

int:

static from_buffer(buffer)

Create an CudaGraphConditionalHandleCreateV12030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphConditionalHandleCreateV12030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_conditional_handle_create_v12030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphConditionalHandleCreateV12030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

p_handle_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGetDriverEntryPointPtszV11030Params

Bases: object

Empty-initialize an instance of cudaGetDriverEntryPoint_ptsz_v11030_params.

See also

cudaGetDriverEntryPoint_ptsz_v11030_params

driver_status

int:

flags_

int:

static from_buffer(buffer)

Create an CudaGetDriverEntryPointPtszV11030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetDriverEntryPointPtszV11030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_driver_entry_point_ptsz_v11030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetDriverEntryPointPtszV11030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func_ptr

int:

ptr

Get the pointer address to the data as Python int.

symbol

str:

class cupti.cupti.CudaGetDriverEntryPointByVersionPtszV12050Params

Bases: object

Empty-initialize an instance of cudaGetDriverEntryPointByVersion_ptsz_v12050_params.

See also

cudaGetDriverEntryPointByVersion_ptsz_v12050_params

cuda_version

int:

driver_status

int:

flags_

int:

static from_buffer(buffer)

Create an CudaGetDriverEntryPointByVersionPtszV12050Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetDriverEntryPointByVersionPtszV12050Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_driver_entry_point_by_version_ptsz_v12050params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetDriverEntryPointByVersionPtszV12050Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func_ptr

int:

ptr

Get the pointer address to the data as Python int.

symbol

str:

class cupti.cupti.CudaGetFuncBySymbolV11000Params

Bases: object

Empty-initialize an instance of cudaGetFuncBySymbol_v11000_params.

See also

cudaGetFuncBySymbol_v11000_params

static from_buffer(buffer)

Create an CudaGetFuncBySymbolV11000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetFuncBySymbolV11000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_func_by_symbol_v11000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetFuncBySymbolV11000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

function_ptr

int:

ptr

Get the pointer address to the data as Python int.

symbol_ptr

int:

class cupti.cupti.CudaGetKernelV12000Params

Bases: object

Empty-initialize an instance of cudaGetKernel_v12000_params.

See also

cudaGetKernel_v12000_params

entry_func_addr

int:

static from_buffer(buffer)

Create an CudaGetKernelV12000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetKernelV12000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_kernel_v12000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetKernelV12000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kernel_ptr

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemcpyV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpy_v3020_params.

See also

cudaMemcpy_v3020_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

class cupti.cupti.CudaMemcpyToSymbolV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpyToSymbol_v3020_params.

See also

cudaMemcpyToSymbol_v3020_params

count

int:

static from_buffer(buffer)

Create an CudaMemcpyToSymbolV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyToSymbolV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_to_symbol_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyToSymbolV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

:

offset

int:

ptr

Get the pointer address to the data as Python int.

src

int:

symbol

int:

class cupti.cupti.CudaMemcpyFromSymbolV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpyFromSymbol_v3020_params.

See also

cudaMemcpyFromSymbol_v3020_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyFromSymbolV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyFromSymbolV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_from_symbol_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyFromSymbolV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

:

offset

int:

ptr

Get the pointer address to the data as Python int.

symbol

int:

class cupti.cupti.CudaMemcpy2DV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpy2D_v3020_params.

See also

cudaMemcpy2D_v3020_params

dpitch

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpy2DV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy2DV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy2d_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy2DV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

kind

:

ptr

Get the pointer address to the data as Python int.

spitch

int:

src

int:

width

int:

class cupti.cupti.CudaMemcpyToArrayV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpyToArray_v3020_params.

See also

cudaMemcpyToArray_v3020_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyToArrayV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyToArrayV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_to_array_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyToArrayV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

w_offset

int:

class cupti.cupti.CudaMemcpy2DToArrayV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpy2DToArray_v3020_params.

See also

cudaMemcpy2DToArray_v3020_params

dst

int:

static from_buffer(buffer)

Create an CudaMemcpy2DToArrayV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy2DToArrayV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy2d_to_array_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy2DToArrayV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

height

int:

kind

:

ptr

Get the pointer address to the data as Python int.

spitch

int:

src

int:

w_offset

int:

width

int:

class cupti.cupti.CudaMemcpyFromArrayV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpyFromArray_v3020_params.

See also

cudaMemcpyFromArray_v3020_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyFromArrayV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyFromArrayV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_from_array_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyFromArrayV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

w_offset

int:

class cupti.cupti.CudaMemcpy2DFromArrayV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpy2DFromArray_v3020_params.

See also

cudaMemcpy2DFromArray_v3020_params

dpitch

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpy2DFromArrayV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy2DFromArrayV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy2d_from_array_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy2DFromArrayV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

height

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

w_offset

int:

width

int:

class cupti.cupti.CudaMemcpyArrayToArrayV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpyArrayToArray_v3020_params.

See also

cudaMemcpyArrayToArray_v3020_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyArrayToArrayV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyArrayToArrayV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_array_to_array_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyArrayToArrayV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset_dst

int:

h_offset_src

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

w_offset_dst

int:

w_offset_src

int:

class cupti.cupti.CudaMemcpy2DArrayToArrayV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpy2DArrayToArray_v3020_params.

See also

cudaMemcpy2DArrayToArray_v3020_params

dst

int:

static from_buffer(buffer)

Create an CudaMemcpy2DArrayToArrayV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy2DArrayToArrayV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy2d_array_to_array_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy2DArrayToArrayV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset_dst

int:

h_offset_src

int:

height

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

w_offset_dst

int:

w_offset_src

int:

width

int:

class cupti.cupti.CudaMemcpy3DV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpy3D_v3020_params.

See also

cudaMemcpy3D_v3020_params

static from_buffer(buffer)

Create an CudaMemcpy3DV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy3DV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy3d_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy3DV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemcpy3DPeerV4000Params

Bases: object

Empty-initialize an instance of cudaMemcpy3DPeer_v4000_params.

See also

cudaMemcpy3DPeer_v4000_params

static from_buffer(buffer)

Create an CudaMemcpy3DPeerV4000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy3DPeerV4000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy3d_peer_v4000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy3DPeerV4000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemsetV3020Params

Bases: object

Empty-initialize an instance of cudaMemset_v3020_params.

See also

cudaMemset_v3020_params

count

int:

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMemsetV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemsetV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memset_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemsetV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CudaMemset2DV3020Params

Bases: object

Empty-initialize an instance of cudaMemset2D_v3020_params.

See also

cudaMemset2D_v3020_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMemset2DV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemset2DV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memset2d_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemset2DV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

pitch

int:

ptr

Get the pointer address to the data as Python int.

value

int:

width

int:

class cupti.cupti.CudaMemset3DV3020Params

Bases: object

Empty-initialize an instance of cudaMemset3D_v3020_params.

See also

cudaMemset3D_v3020_params

static from_buffer(buffer)

Create an CudaMemset3DV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemset3DV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memset3d_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemset3DV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CudaMemcpyAsyncV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpyAsync_v3020_params.

See also

cudaMemcpyAsync_v3020_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyAsyncV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyAsyncV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_async_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyAsyncV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

stream

int:

class cupti.cupti.CudaMemcpyToSymbolAsyncV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpyToSymbolAsync_v3020_params.

See also

cudaMemcpyToSymbolAsync_v3020_params

count

int:

static from_buffer(buffer)

Create an CudaMemcpyToSymbolAsyncV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyToSymbolAsyncV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_to_symbol_async_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyToSymbolAsyncV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

:

offset

int:

ptr

Get the pointer address to the data as Python int.

src

int:

stream

int:

symbol

int:

class cupti.cupti.CudaMemcpyFromSymbolAsyncV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpyFromSymbolAsync_v3020_params.

See also

cudaMemcpyFromSymbolAsync_v3020_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyFromSymbolAsyncV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyFromSymbolAsyncV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_from_symbol_async_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyFromSymbolAsyncV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kind

:

offset

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

symbol

int:

class cupti.cupti.CudaMemcpy2DAsyncV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpy2DAsync_v3020_params.

See also

cudaMemcpy2DAsync_v3020_params

dpitch

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpy2DAsyncV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy2DAsyncV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy2_dasync_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy2DAsyncV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

kind

:

ptr

Get the pointer address to the data as Python int.

spitch

int:

src

int:

stream

int:

width

int:

class cupti.cupti.CudaMemcpyToArrayAsyncV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpyToArrayAsync_v3020_params.

See also

cudaMemcpyToArrayAsync_v3020_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyToArrayAsyncV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyToArrayAsyncV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_to_array_async_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyToArrayAsyncV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

stream

int:

w_offset

int:

class cupti.cupti.CudaMemcpy2DToArrayAsyncV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpy2DToArrayAsync_v3020_params.

See also

cudaMemcpy2DToArrayAsync_v3020_params

dst

int:

static from_buffer(buffer)

Create an CudaMemcpy2DToArrayAsyncV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy2DToArrayAsyncV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy2d_to_array_async_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy2DToArrayAsyncV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

height

int:

kind

:

ptr

Get the pointer address to the data as Python int.

spitch

int:

src

int:

stream

int:

w_offset

int:

width

int:

class cupti.cupti.CudaMemcpyFromArrayAsyncV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpyFromArrayAsync_v3020_params.

See also

cudaMemcpyFromArrayAsync_v3020_params

count

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyFromArrayAsyncV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyFromArrayAsyncV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_from_array_async_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyFromArrayAsyncV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

stream

int:

w_offset

int:

class cupti.cupti.CudaMemcpy2DFromArrayAsyncV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpy2DFromArrayAsync_v3020_params.

See also

cudaMemcpy2DFromArrayAsync_v3020_params

dpitch

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpy2DFromArrayAsyncV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy2DFromArrayAsyncV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy2d_from_array_async_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy2DFromArrayAsyncV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_offset

int:

height

int:

kind

:

ptr

Get the pointer address to the data as Python int.

src

int:

stream

int:

w_offset

int:

width

int:

class cupti.cupti.CudaMemcpy3DAsyncV3020Params

Bases: object

Empty-initialize an instance of cudaMemcpy3DAsync_v3020_params.

See also

cudaMemcpy3DAsync_v3020_params

static from_buffer(buffer)

Create an CudaMemcpy3DAsyncV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy3DAsyncV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy3_dasync_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy3DAsyncV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaMemcpy3DPeerAsyncV4000Params

Bases: object

Empty-initialize an instance of cudaMemcpy3DPeerAsync_v4000_params.

See also

cudaMemcpy3DPeerAsync_v4000_params

static from_buffer(buffer)

Create an CudaMemcpy3DPeerAsyncV4000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy3DPeerAsyncV4000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy3d_peer_async_v4000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy3DPeerAsyncV4000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaMemsetAsyncV3020Params

Bases: object

Empty-initialize an instance of cudaMemsetAsync_v3020_params.

See also

cudaMemsetAsync_v3020_params

count

int:

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMemsetAsyncV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemsetAsyncV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memset_async_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemsetAsyncV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CudaMemset2DAsyncV3020Params

Bases: object

Empty-initialize an instance of cudaMemset2DAsync_v3020_params.

See also

cudaMemset2DAsync_v3020_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMemset2DAsyncV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemset2DAsyncV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memset2_dasync_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemset2DAsyncV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

pitch

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

width

int:

class cupti.cupti.CudaMemset3DAsyncV3020Params

Bases: object

Empty-initialize an instance of cudaMemset3DAsync_v3020_params.

See also

cudaMemset3DAsync_v3020_params

static from_buffer(buffer)

Create an CudaMemset3DAsyncV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemset3DAsyncV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memset3_dasync_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemset3DAsyncV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CudaStreamQueryV3020Params

Bases: object

Empty-initialize an instance of cudaStreamQuery_v3020_params.

See also

cudaStreamQuery_v3020_params

static from_buffer(buffer)

Create an CudaStreamQueryV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamQueryV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_query_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamQueryV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamGetDeviceV12080Params

Bases: object

Empty-initialize an instance of cudaStreamGetDevice_v12080_params.

See also

cudaStreamGetDevice_v12080_params

device_

int:

static from_buffer(buffer)

Create an CudaStreamGetDeviceV12080Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamGetDeviceV12080Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_get_device_v12080params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamGetDeviceV12080Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaStreamGetFlagsV5050Params

Bases: object

Empty-initialize an instance of cudaStreamGetFlags_v5050_params.

See also

cudaStreamGetFlags_v5050_params

flags_

int:

static from_buffer(buffer)

Create an CudaStreamGetFlagsV5050Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamGetFlagsV5050Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_get_flags_v5050params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamGetFlagsV5050Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaStreamGetIdV12000Params

Bases: object

Empty-initialize an instance of cudaStreamGetId_v12000_params.

See also

cudaStreamGetId_v12000_params

static from_buffer(buffer)

Create an CudaStreamGetIdV12000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamGetIdV12000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_get_id_v12000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamGetIdV12000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

stream_id

int:

class cupti.cupti.CudaStreamGetPriorityV5050Params

Bases: object

Empty-initialize an instance of cudaStreamGetPriority_v5050_params.

See also

cudaStreamGetPriority_v5050_params

static from_buffer(buffer)

Create an CudaStreamGetPriorityV5050Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamGetPriorityV5050Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_get_priority_v5050params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamGetPriorityV5050Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

priority

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaEventRecordV3020Params

Bases: object

Empty-initialize an instance of cudaEventRecord_v3020_params.

See also

cudaEventRecord_v3020_params

event

int:

static from_buffer(buffer)

Create an CudaEventRecordV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaEventRecordV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_event_record_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaEventRecordV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaEventRecordWithFlagsV11010Params

Bases: object

Empty-initialize an instance of cudaEventRecordWithFlags_v11010_params.

See also

cudaEventRecordWithFlags_v11010_params

event

int:

flags_

int:

static from_buffer(buffer)

Create an CudaEventRecordWithFlagsV11010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaEventRecordWithFlagsV11010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_event_record_with_flags_v11010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaEventRecordWithFlagsV11010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamWaitEventV3020Params

Bases: object

Empty-initialize an instance of cudaStreamWaitEvent_v3020_params.

See also

cudaStreamWaitEvent_v3020_params

event

int:

flags_

int:

static from_buffer(buffer)

Create an CudaStreamWaitEventV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamWaitEventV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_wait_event_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamWaitEventV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamAddCallbackV5000Params

Bases: object

Empty-initialize an instance of cudaStreamAddCallback_v5000_params.

See also

cudaStreamAddCallback_v5000_params

callback

int:

flags_

int:

static from_buffer(buffer)

Create an CudaStreamAddCallbackV5000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamAddCallbackV5000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_add_callback_v5000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamAddCallbackV5000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

user_data

int:

class cupti.cupti.CudaStreamAttachMemAsyncV6000Params

Bases: object

Empty-initialize an instance of cudaStreamAttachMemAsync_v6000_params.

See also

cudaStreamAttachMemAsync_v6000_params

dev_ptr

int:

flags_

int:

static from_buffer(buffer)

Create an CudaStreamAttachMemAsyncV6000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamAttachMemAsyncV6000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_attach_mem_async_v6000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamAttachMemAsyncV6000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

length

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamSynchronizeV3020Params

Bases: object

Empty-initialize an instance of cudaStreamSynchronize_v3020_params.

See also

cudaStreamSynchronize_v3020_params

static from_buffer(buffer)

Create an CudaStreamSynchronizeV3020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamSynchronizeV3020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_synchronize_v3020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamSynchronizeV3020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaLaunchKernelV7000Params

Bases: object

Empty-initialize an instance of cudaLaunchKernel_v7000_params.

See also

cudaLaunchKernel_v7000_params

args

int:

block_dim

:

static from_buffer(buffer)

Create an CudaLaunchKernelV7000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLaunchKernelV7000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_launch_kernel_v7000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLaunchKernelV7000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

grid_dim

:

ptr

Get the pointer address to the data as Python int.

shared_mem

int:

stream

int:

class cupti.cupti.CudaLaunchKernelExCV11060Params

Bases: object

Empty-initialize an instance of cudaLaunchKernelExC_v11060_params.

See also

cudaLaunchKernelExC_v11060_params

args

int:

config

int:

static from_buffer(buffer)

Create an CudaLaunchKernelExCV11060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLaunchKernelExCV11060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_launch_kernel_ex_c_v11060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLaunchKernelExCV11060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLaunchCooperativeKernelV9000Params

Bases: object

Empty-initialize an instance of cudaLaunchCooperativeKernel_v9000_params.

See also

cudaLaunchCooperativeKernel_v9000_params

args

int:

block_dim

:

static from_buffer(buffer)

Create an CudaLaunchCooperativeKernelV9000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLaunchCooperativeKernelV9000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_launch_cooperative_kernel_v9000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLaunchCooperativeKernelV9000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

grid_dim

:

ptr

Get the pointer address to the data as Python int.

shared_mem

int:

stream

int:

class cupti.cupti.CudaLaunchHostFuncV10000Params

Bases: object

Empty-initialize an instance of cudaLaunchHostFunc_v10000_params.

See also

cudaLaunchHostFunc_v10000_params

fn

int:

static from_buffer(buffer)

Create an CudaLaunchHostFuncV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLaunchHostFuncV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_launch_host_func_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLaunchHostFuncV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

user_data

int:

class cupti.cupti.CudaGraphInstantiateWithParamsV12000Params

Bases: object

Empty-initialize an instance of cudaGraphInstantiateWithParams_v12000_params.

See also

cudaGraphInstantiateWithParams_v12000_params

static from_buffer(buffer)

Create an CudaGraphInstantiateWithParamsV12000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphInstantiateWithParamsV12000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_instantiate_with_params_v12000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphInstantiateWithParamsV12000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

instantiate_params

int:

p_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphUploadV10000Params

Bases: object

Empty-initialize an instance of cudaGraphUpload_v10000_params.

See also

cudaGraphUpload_v10000_params

static from_buffer(buffer)

Create an CudaGraphUploadV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphUploadV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_upload_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphUploadV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_exec

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaGraphLaunchV10000Params

Bases: object

Empty-initialize an instance of cudaGraphLaunch_v10000_params.

See also

cudaGraphLaunch_v10000_params

static from_buffer(buffer)

Create an CudaGraphLaunchV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphLaunchV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_launch_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphLaunchV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_exec

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamBeginCaptureV10000Params

Bases: object

Empty-initialize an instance of cudaStreamBeginCapture_v10000_params.

See also

cudaStreamBeginCapture_v10000_params

static from_buffer(buffer)

Create an CudaStreamBeginCaptureV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamBeginCaptureV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_begin_capture_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamBeginCaptureV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mode

:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamBeginCaptureToGraphV12030Params

Bases: object

Empty-initialize an instance of cudaStreamBeginCaptureToGraph_v12030_params.

See also

cudaStreamBeginCaptureToGraph_v12030_params

dependencies

int:

dependency_data

int:

static from_buffer(buffer)

Create an CudaStreamBeginCaptureToGraphV12030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamBeginCaptureToGraphV12030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_begin_capture_to_graph_v12030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamBeginCaptureToGraphV12030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

mode

:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamEndCaptureV10000Params

Bases: object

Empty-initialize an instance of cudaStreamEndCapture_v10000_params.

See also

cudaStreamEndCapture_v10000_params

static from_buffer(buffer)

Create an CudaStreamEndCaptureV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamEndCaptureV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_end_capture_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamEndCaptureV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_graph

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamIsCapturingV10000Params

Bases: object

Empty-initialize an instance of cudaStreamIsCapturing_v10000_params.

See also

cudaStreamIsCapturing_v10000_params

static from_buffer(buffer)

Create an CudaStreamIsCapturingV10000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamIsCapturingV10000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_is_capturing_v10000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamIsCapturingV10000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_capture_status

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamCopyAttributesV11000Params

Bases: object

Empty-initialize an instance of cudaStreamCopyAttributes_v11000_params.

See also

cudaStreamCopyAttributes_v11000_params

dst_stream

int:

static from_buffer(buffer)

Create an CudaStreamCopyAttributesV11000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamCopyAttributesV11000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_copy_attributes_v11000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamCopyAttributesV11000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_stream

int:

class cupti.cupti.CudaStreamGetAttributeV11000Params

Bases: object

Empty-initialize an instance of cudaStreamGetAttribute_v11000_params.

See also

cudaStreamGetAttribute_v11000_params

attr

:

static from_buffer(buffer)

Create an CudaStreamGetAttributeV11000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamGetAttributeV11000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_get_attribute_v11000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamGetAttributeV11000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CudaStreamSetAttributeV11000Params

Bases: object

Empty-initialize an instance of cudaStreamSetAttribute_v11000_params.

See also

cudaStreamSetAttribute_v11000_params

attr

:

static from_buffer(buffer)

Create an CudaStreamSetAttributeV11000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamSetAttributeV11000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_set_attribute_v11000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamSetAttributeV11000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

param

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaMallocAsyncV11020Params

Bases: object

Empty-initialize an instance of cudaMallocAsync_v11020_params.

See also

cudaMallocAsync_v11020_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMallocAsyncV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMallocAsyncV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_malloc_async_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMallocAsyncV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CudaFreeAsyncV11020Params

Bases: object

Empty-initialize an instance of cudaFreeAsync_v11020_params.

See also

cudaFreeAsync_v11020_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaFreeAsyncV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaFreeAsyncV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_free_async_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaFreeAsyncV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMallocFromPoolAsyncV11020Params

Bases: object

Empty-initialize an instance of cudaMallocFromPoolAsync_v11020_params.

See also

cudaMallocFromPoolAsync_v11020_params

static from_buffer(buffer)

Create an CudaMallocFromPoolAsyncV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMallocFromPoolAsyncV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_malloc_from_pool_async_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMallocFromPoolAsyncV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

ptr_

int:

size_

int:

stream

int:

class cupti.cupti.CudaGetDriverEntryPointV11030Params

Bases: object

Empty-initialize an instance of cudaGetDriverEntryPoint_v11030_params.

See also

cudaGetDriverEntryPoint_v11030_params

driver_status

int:

flags_

int:

static from_buffer(buffer)

Create an CudaGetDriverEntryPointV11030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetDriverEntryPointV11030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_driver_entry_point_v11030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetDriverEntryPointV11030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func_ptr

int:

ptr

Get the pointer address to the data as Python int.

symbol

str:

class cupti.cupti.CudaGetDriverEntryPointByVersionV12050Params

Bases: object

Empty-initialize an instance of cudaGetDriverEntryPointByVersion_v12050_params.

See also

cudaGetDriverEntryPointByVersion_v12050_params

cuda_version

int:

driver_status

int:

flags_

int:

static from_buffer(buffer)

Create an CudaGetDriverEntryPointByVersionV12050Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetDriverEntryPointByVersionV12050Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_driver_entry_point_by_version_v12050params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetDriverEntryPointByVersionV12050Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func_ptr

int:

ptr

Get the pointer address to the data as Python int.

symbol

str:

class cupti.cupti.CudaGetDevicePropertiesV12000Params

Bases: object

Empty-initialize an instance of cudaGetDeviceProperties_v12000_params.

See also

cudaGetDeviceProperties_v12000_params

device_

int:

static from_buffer(buffer)

Create an CudaGetDevicePropertiesV12000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetDevicePropertiesV12000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_device_properties_v12000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetDevicePropertiesV12000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

prop

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGetHostAtomicCapabilitiesV13000Params

Bases: object

Empty-initialize an instance of cudaDeviceGetHostAtomicCapabilities_v13000_params.

See also

cudaDeviceGetHostAtomicCapabilities_v13000_params

capabilities

int:

count

int:

device_

int:

static from_buffer(buffer)

Create an CudaDeviceGetHostAtomicCapabilitiesV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetHostAtomicCapabilitiesV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_host_atomic_capabilities_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetHostAtomicCapabilitiesV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

operations

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGetP2PAtomicCapabilitiesV13000Params

Bases: object

Empty-initialize an instance of cudaDeviceGetP2PAtomicCapabilities_v13000_params.

See also

cudaDeviceGetP2PAtomicCapabilities_v13000_params

capabilities

int:

count

int:

dst_device

int:

static from_buffer(buffer)

Create an CudaDeviceGetP2PAtomicCapabilitiesV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetP2PAtomicCapabilitiesV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_p2p_atomic_capabilities_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetP2PAtomicCapabilitiesV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

operations

int:

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CudaStreamGetCaptureInfoPtszV12030Params

Bases: object

Empty-initialize an instance of cudaStreamGetCaptureInfo_ptsz_v12030_params.

See also

cudaStreamGetCaptureInfo_ptsz_v12030_params

capture_status_out

int:

dependencies_out

int:

edge_data_out

int:

static from_buffer(buffer)

Create an CudaStreamGetCaptureInfoPtszV12030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamGetCaptureInfoPtszV12030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_get_capture_info_ptsz_v12030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamGetCaptureInfoPtszV12030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_out

int:

id_out

int:

num_dependencies_out

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamUpdateCaptureDependenciesPtszV12030Params

Bases: object

Empty-initialize an instance of cudaStreamUpdateCaptureDependencies_ptsz_v12030_params.

See also

cudaStreamUpdateCaptureDependencies_ptsz_v12030_params

dependencies

int:

dependency_data

int:

flags_

int:

static from_buffer(

buffer,

)

Create an CudaStreamUpdateCaptureDependenciesPtszV12030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamUpdateCaptureDependenciesPtszV12030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_update_capture_dependencies_ptsz_v12030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamUpdateCaptureDependenciesPtszV12030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaEventElapsedTimeV12080Params

Bases: object

Empty-initialize an instance of cudaEventElapsedTime_v12080_params.

See also

cudaEventElapsedTime_v12080_params

end

int:

static from_buffer(buffer)

Create an CudaEventElapsedTimeV12080Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaEventElapsedTimeV12080Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_event_elapsed_time_v12080params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaEventElapsedTimeV12080Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ms

int:

ptr

Get the pointer address to the data as Python int.

start

int:

class cupti.cupti.CudaSignalExternalSemaphoresAsyncPtszV11020Params

Bases: object

Empty-initialize an instance of cudaSignalExternalSemaphoresAsync_ptsz_v11020_params.

See also

cudaSignalExternalSemaphoresAsync_ptsz_v11020_params

ext_sem_array

int:

static from_buffer(buffer)

Create an CudaSignalExternalSemaphoresAsyncPtszV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaSignalExternalSemaphoresAsyncPtszV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_signal_external_semaphores_async_ptsz_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaSignalExternalSemaphoresAsyncPtszV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_ext_sems

int:

params_array

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaWaitExternalSemaphoresAsyncPtszV11020Params

Bases: object

Empty-initialize an instance of cudaWaitExternalSemaphoresAsync_ptsz_v11020_params.

See also

cudaWaitExternalSemaphoresAsync_ptsz_v11020_params

ext_sem_array

int:

static from_buffer(buffer)

Create an CudaWaitExternalSemaphoresAsyncPtszV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaWaitExternalSemaphoresAsyncPtszV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_wait_external_semaphores_async_ptsz_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaWaitExternalSemaphoresAsyncPtszV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_ext_sems

int:

params_array

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaMemcpyBatchAsyncPtszV13000Params

Bases: object

Empty-initialize an instance of cudaMemcpyBatchAsync_ptsz_v13000_params.

See also

cudaMemcpyBatchAsync_ptsz_v13000_params

attrs

int:

attrs_idxs

int:

count

int:

dsts

int:

static from_buffer(buffer)

Create an CudaMemcpyBatchAsyncPtszV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyBatchAsyncPtszV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_batch_async_ptsz_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyBatchAsyncPtszV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_attrs

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

srcs

int:

stream

int:

class cupti.cupti.CudaMemcpy3DBatchAsyncPtszV13000Params

Bases: object

Empty-initialize an instance of cudaMemcpy3DBatchAsync_ptsz_v13000_params.

See also

cudaMemcpy3DBatchAsync_ptsz_v13000_params

flags_

int:

static from_buffer(buffer)

Create an CudaMemcpy3DBatchAsyncPtszV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy3DBatchAsyncPtszV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy3d_batch_async_ptsz_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy3DBatchAsyncPtszV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_ops

int:

op_list

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaMemPrefetchAsyncPtszV12020Params

Bases: object

Empty-initialize an instance of cudaMemPrefetchAsync_ptsz_v12020_params.

See also

cudaMemPrefetchAsync_ptsz_v12020_params

count

int:

dev_ptr

int:

flags_

int:

static from_buffer(buffer)

Create an CudaMemPrefetchAsyncPtszV12020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemPrefetchAsyncPtszV12020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_prefetch_async_ptsz_v12020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemPrefetchAsyncPtszV12020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaMemPrefetchBatchAsyncPtszV13000Params

Bases: object

Empty-initialize an instance of cudaMemPrefetchBatchAsync_ptsz_v13000_params.

See also

cudaMemPrefetchBatchAsync_ptsz_v13000_params

count

int:

dptrs

int:

flags_

int:

static from_buffer(buffer)

Create an CudaMemPrefetchBatchAsyncPtszV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemPrefetchBatchAsyncPtszV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_prefetch_batch_async_ptsz_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemPrefetchBatchAsyncPtszV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_prefetch_locs

int:

prefetch_loc_idxs

int:

prefetch_locs

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

stream

int:

class cupti.cupti.CudaMemDiscardBatchAsyncPtszV13000Params

Bases: object

Empty-initialize an instance of cudaMemDiscardBatchAsync_ptsz_v13000_params.

See also

cudaMemDiscardBatchAsync_ptsz_v13000_params

count

int:

dptrs

int:

flags_

int:

static from_buffer(buffer)

Create an CudaMemDiscardBatchAsyncPtszV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemDiscardBatchAsyncPtszV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_discard_batch_async_ptsz_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemDiscardBatchAsyncPtszV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

sizes

int:

stream

int:

class cupti.cupti.CudaMemDiscardAndPrefetchBatchAsyncPtszV13000Params

Bases: object

Empty-initialize an instance of cudaMemDiscardAndPrefetchBatchAsync_ptsz_v13000_params.

See also

cudaMemDiscardAndPrefetchBatchAsync_ptsz_v13000_params

count

int:

dptrs

int:

flags_

int:

static from_buffer(

buffer,

)

Create an CudaMemDiscardAndPrefetchBatchAsyncPtszV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemDiscardAndPrefetchBatchAsyncPtszV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_discard_and_prefetch_batch_async_ptsz_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemDiscardAndPrefetchBatchAsyncPtszV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_prefetch_locs

int:

prefetch_loc_idxs

int:

prefetch_locs

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

stream

int:

class cupti.cupti.CudaMemAdviseV12020Params

Bases: object

Empty-initialize an instance of cudaMemAdvise_v12020_params.

See also

cudaMemAdvise_v12020_params

advice

:

count

int:

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMemAdviseV12020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemAdviseV12020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_advise_v12020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemAdviseV12020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemGetDefaultMemPoolV13000Params

Bases: object

Empty-initialize an instance of cudaMemGetDefaultMemPool_v13000_params.

See also

cudaMemGetDefaultMemPool_v13000_params

static from_buffer(buffer)

Create an CudaMemGetDefaultMemPoolV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemGetDefaultMemPoolV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_get_default_mem_pool_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemGetDefaultMemPoolV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

location

int:

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

type

:

class cupti.cupti.CudaMemGetMemPoolV13000Params

Bases: object

Empty-initialize an instance of cudaMemGetMemPool_v13000_params.

See also

cudaMemGetMemPool_v13000_params

static from_buffer(buffer)

Create an CudaMemGetMemPoolV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemGetMemPoolV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_get_mem_pool_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemGetMemPoolV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

location

int:

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

type

:

class cupti.cupti.CudaMemSetMemPoolV13000Params

Bases: object

Empty-initialize an instance of cudaMemSetMemPool_v13000_params.

See also

cudaMemSetMemPool_v13000_params

static from_buffer(buffer)

Create an CudaMemSetMemPoolV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemSetMemPoolV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_set_mem_pool_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemSetMemPoolV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

location

int:

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

type

:

class cupti.cupti.CudaLogsRegisterCallbackV13000Params

Bases: object

Empty-initialize an instance of cudaLogsRegisterCallback_v13000_params.

See also

cudaLogsRegisterCallback_v13000_params

callback_func

int:

callback_out

int:

static from_buffer(buffer)

Create an CudaLogsRegisterCallbackV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLogsRegisterCallbackV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_logs_register_callback_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLogsRegisterCallbackV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

user_data

int:

class cupti.cupti.CudaLogsUnregisterCallbackV13000Params

Bases: object

Empty-initialize an instance of cudaLogsUnregisterCallback_v13000_params.

See also

cudaLogsUnregisterCallback_v13000_params

callback

int:

static from_buffer(buffer)

Create an CudaLogsUnregisterCallbackV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLogsUnregisterCallbackV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_logs_unregister_callback_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLogsUnregisterCallbackV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLogsCurrentV13000Params

Bases: object

Empty-initialize an instance of cudaLogsCurrent_v13000_params.

See also

cudaLogsCurrent_v13000_params

flags_

int:

static from_buffer(buffer)

Create an CudaLogsCurrentV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLogsCurrentV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_logs_current_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLogsCurrentV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

iterator_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLogsDumpToFileV13000Params

Bases: object

Empty-initialize an instance of cudaLogsDumpToFile_v13000_params.

See also

cudaLogsDumpToFile_v13000_params

flags_

int:

static from_buffer(buffer)

Create an CudaLogsDumpToFileV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLogsDumpToFileV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_logs_dump_to_file_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLogsDumpToFileV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

iterator

int:

path_to_file

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLogsDumpToMemoryV13000Params

Bases: object

Empty-initialize an instance of cudaLogsDumpToMemory_v13000_params.

See also

cudaLogsDumpToMemory_v13000_params

buffer

str:

flags_

int:

static from_buffer(buffer)

Create an CudaLogsDumpToMemoryV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLogsDumpToMemoryV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_logs_dump_to_memory_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLogsDumpToMemoryV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

iterator

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CudaGraphGetEdgesV12030Params

Bases: object

Empty-initialize an instance of cudaGraphGetEdges_v12030_params.

See also

cudaGraphGetEdges_v12030_params

edge_data

int:

from_

int:

static from_buffer(buffer)

Create an CudaGraphGetEdgesV12030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphGetEdgesV12030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_get_edges_v12030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphGetEdgesV12030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

num_edges

int:

ptr

Get the pointer address to the data as Python int.

to

int:

class cupti.cupti.CudaGraphNodeGetDependenciesV12030Params

Bases: object

Empty-initialize an instance of cudaGraphNodeGetDependencies_v12030_params.

See also

cudaGraphNodeGetDependencies_v12030_params

edge_data

int:

static from_buffer(buffer)

Create an CudaGraphNodeGetDependenciesV12030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphNodeGetDependenciesV12030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_node_get_dependencies_v12030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphNodeGetDependenciesV12030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

p_dependencies

int:

p_num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphNodeGetDependentNodesV12030Params

Bases: object

Empty-initialize an instance of cudaGraphNodeGetDependentNodes_v12030_params.

See also

cudaGraphNodeGetDependentNodes_v12030_params

edge_data

int:

static from_buffer(buffer)

Create an CudaGraphNodeGetDependentNodesV12030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphNodeGetDependentNodesV12030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_node_get_dependent_nodes_v12030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphNodeGetDependentNodesV12030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

p_dependent_nodes

int:

p_num_dependent_nodes

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphAddDependenciesV12030Params

Bases: object

Empty-initialize an instance of cudaGraphAddDependencies_v12030_params.

See also

cudaGraphAddDependencies_v12030_params

edge_data

int:

from_

int:

static from_buffer(buffer)

Create an CudaGraphAddDependenciesV12030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddDependenciesV12030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_dependencies_v12030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddDependenciesV12030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

to

int:

class cupti.cupti.CudaGraphRemoveDependenciesV12030Params

Bases: object

Empty-initialize an instance of cudaGraphRemoveDependencies_v12030_params.

See also

cudaGraphRemoveDependencies_v12030_params

edge_data

int:

from_

int:

static from_buffer(buffer)

Create an CudaGraphRemoveDependenciesV12030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphRemoveDependenciesV12030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_remove_dependencies_v12030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphRemoveDependenciesV12030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

to

int:

class cupti.cupti.CudaGraphAddNodeV12030Params

Bases: object

Empty-initialize an instance of cudaGraphAddNode_v12030_params.

See also

cudaGraphAddNode_v12030_params

dependency_data

int:

static from_buffer(buffer)

Create an CudaGraphAddNodeV12030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphAddNodeV12030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_add_node_v12030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphAddNodeV12030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

node_params

int:

num_dependencies

int:

p_dependencies

int:

p_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLibraryLoadDataV12060Params

Bases: object

Empty-initialize an instance of cudaLibraryLoadData_v12060_params.

See also

cudaLibraryLoadData_v12060_params

code

int:

static from_buffer(buffer)

Create an CudaLibraryLoadDataV12060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLibraryLoadDataV12060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_library_load_data_v12060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLibraryLoadDataV12060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

jit_options

int:

jit_options_values

int:

library

int:

library_option_values

int:

library_options

int:

num_jit_options

int:

num_library_options

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLibraryLoadFromFileV12060Params

Bases: object

Empty-initialize an instance of cudaLibraryLoadFromFile_v12060_params.

See also

cudaLibraryLoadFromFile_v12060_params

file_name

str:

static from_buffer(buffer)

Create an CudaLibraryLoadFromFileV12060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLibraryLoadFromFileV12060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_library_load_from_file_v12060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLibraryLoadFromFileV12060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

jit_options

int:

jit_options_values

int:

library

int:

library_option_values

int:

library_options

int:

num_jit_options

int:

num_library_options

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLibraryUnloadV12060Params

Bases: object

Empty-initialize an instance of cudaLibraryUnload_v12060_params.

See also

cudaLibraryUnload_v12060_params

static from_buffer(buffer)

Create an CudaLibraryUnloadV12060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLibraryUnloadV12060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_library_unload_v12060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLibraryUnloadV12060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

library

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLibraryGetKernelV12060Params

Bases: object

Empty-initialize an instance of cudaLibraryGetKernel_v12060_params.

See also

cudaLibraryGetKernel_v12060_params

static from_buffer(buffer)

Create an CudaLibraryGetKernelV12060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLibraryGetKernelV12060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_library_get_kernel_v12060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLibraryGetKernelV12060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

library

int:

name

str:

p_kernel

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLibraryGetGlobalV12060Params

Bases: object

Empty-initialize an instance of cudaLibraryGetGlobal_v12060_params.

See also

cudaLibraryGetGlobal_v12060_params

bytes

int:

dptr

int:

static from_buffer(buffer)

Create an CudaLibraryGetGlobalV12060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLibraryGetGlobalV12060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_library_get_global_v12060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLibraryGetGlobalV12060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

library

int:

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLibraryGetManagedV12060Params

Bases: object

Empty-initialize an instance of cudaLibraryGetManaged_v12060_params.

See also

cudaLibraryGetManaged_v12060_params

bytes

int:

dptr

int:

static from_buffer(buffer)

Create an CudaLibraryGetManagedV12060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLibraryGetManagedV12060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_library_get_managed_v12060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLibraryGetManagedV12060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

library

int:

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLibraryGetUnifiedFunctionV12060Params

Bases: object

Empty-initialize an instance of cudaLibraryGetUnifiedFunction_v12060_params.

See also

cudaLibraryGetUnifiedFunction_v12060_params

fptr

int:

static from_buffer(buffer)

Create an CudaLibraryGetUnifiedFunctionV12060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLibraryGetUnifiedFunctionV12060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_library_get_unified_function_v12060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLibraryGetUnifiedFunctionV12060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

library

int:

ptr

Get the pointer address to the data as Python int.

symbol

str:

class cupti.cupti.CudaLibraryGetKernelCountV12060Params

Bases: object

Empty-initialize an instance of cudaLibraryGetKernelCount_v12060_params.

See also

cudaLibraryGetKernelCount_v12060_params

count

int:

static from_buffer(buffer)

Create an CudaLibraryGetKernelCountV12060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLibraryGetKernelCountV12060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_library_get_kernel_count_v12060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLibraryGetKernelCountV12060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

lib

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLibraryEnumerateKernelsV12060Params

Bases: object

Empty-initialize an instance of cudaLibraryEnumerateKernels_v12060_params.

See also

cudaLibraryEnumerateKernels_v12060_params

static from_buffer(buffer)

Create an CudaLibraryEnumerateKernelsV12060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLibraryEnumerateKernelsV12060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_library_enumerate_kernels_v12060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLibraryEnumerateKernelsV12060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kernels

int:

lib

int:

num_kernels

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaKernelSetAttributeForDeviceV12060Params

Bases: object

Empty-initialize an instance of cudaKernelSetAttributeForDevice_v12060_params.

See also

cudaKernelSetAttributeForDevice_v12060_params

attr

:

device_

int:

static from_buffer(buffer)

Create an CudaKernelSetAttributeForDeviceV12060Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaKernelSetAttributeForDeviceV12060Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_kernel_set_attribute_for_device_v12060params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaKernelSetAttributeForDeviceV12060Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kernel

int:

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CudaMemcpyBatchAsyncV13000Params

Bases: object

Empty-initialize an instance of cudaMemcpyBatchAsync_v13000_params.

See also

cudaMemcpyBatchAsync_v13000_params

attrs

int:

attrs_idxs

int:

count

int:

dsts

int:

static from_buffer(buffer)

Create an CudaMemcpyBatchAsyncV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyBatchAsyncV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_batch_async_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyBatchAsyncV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_attrs

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

srcs

int:

stream

int:

class cupti.cupti.CudaMemcpy3DBatchAsyncV13000Params

Bases: object

Empty-initialize an instance of cudaMemcpy3DBatchAsync_v13000_params.

See also

cudaMemcpy3DBatchAsync_v13000_params

flags_

int:

static from_buffer(buffer)

Create an CudaMemcpy3DBatchAsyncV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy3DBatchAsyncV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy3d_batch_async_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy3DBatchAsyncV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_ops

int:

op_list

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaMemPrefetchAsyncV12020Params

Bases: object

Empty-initialize an instance of cudaMemPrefetchAsync_v12020_params.

See also

cudaMemPrefetchAsync_v12020_params

count

int:

dev_ptr

int:

flags_

int:

static from_buffer(buffer)

Create an CudaMemPrefetchAsyncV12020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemPrefetchAsyncV12020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_prefetch_async_v12020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemPrefetchAsyncV12020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaMemPrefetchBatchAsyncV13000Params

Bases: object

Empty-initialize an instance of cudaMemPrefetchBatchAsync_v13000_params.

See also

cudaMemPrefetchBatchAsync_v13000_params

count

int:

dptrs

int:

flags_

int:

static from_buffer(buffer)

Create an CudaMemPrefetchBatchAsyncV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemPrefetchBatchAsyncV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_prefetch_batch_async_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemPrefetchBatchAsyncV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_prefetch_locs

int:

prefetch_loc_idxs

int:

prefetch_locs

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

stream

int:

class cupti.cupti.CudaMemDiscardBatchAsyncV13000Params

Bases: object

Empty-initialize an instance of cudaMemDiscardBatchAsync_v13000_params.

See also

cudaMemDiscardBatchAsync_v13000_params

count

int:

dptrs

int:

flags_

int:

static from_buffer(buffer)

Create an CudaMemDiscardBatchAsyncV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemDiscardBatchAsyncV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_discard_batch_async_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemDiscardBatchAsyncV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

sizes

int:

stream

int:

class cupti.cupti.CudaMemDiscardAndPrefetchBatchAsyncV13000Params

Bases: object

Empty-initialize an instance of cudaMemDiscardAndPrefetchBatchAsync_v13000_params.

See also

cudaMemDiscardAndPrefetchBatchAsync_v13000_params

count

int:

dptrs

int:

flags_

int:

static from_buffer(buffer)

Create an CudaMemDiscardAndPrefetchBatchAsyncV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemDiscardAndPrefetchBatchAsyncV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_discard_and_prefetch_batch_async_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemDiscardAndPrefetchBatchAsyncV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_prefetch_locs

int:

prefetch_loc_idxs

int:

prefetch_locs

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

stream

int:

class cupti.cupti.CudaSignalExternalSemaphoresAsyncV11020Params

Bases: object

Empty-initialize an instance of cudaSignalExternalSemaphoresAsync_v11020_params.

See also

cudaSignalExternalSemaphoresAsync_v11020_params

ext_sem_array

int:

static from_buffer(buffer)

Create an CudaSignalExternalSemaphoresAsyncV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaSignalExternalSemaphoresAsyncV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_signal_external_semaphores_async_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaSignalExternalSemaphoresAsyncV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_ext_sems

int:

params_array

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaWaitExternalSemaphoresAsyncV11020Params

Bases: object

Empty-initialize an instance of cudaWaitExternalSemaphoresAsync_v11020_params.

See also

cudaWaitExternalSemaphoresAsync_v11020_params

ext_sem_array

int:

static from_buffer(buffer)

Create an CudaWaitExternalSemaphoresAsyncV11020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaWaitExternalSemaphoresAsyncV11020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_wait_external_semaphores_async_v11020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaWaitExternalSemaphoresAsyncV11020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_ext_sems

int:

params_array

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamGetCaptureInfoV12030Params

Bases: object

Empty-initialize an instance of cudaStreamGetCaptureInfo_v12030_params.

See also

cudaStreamGetCaptureInfo_v12030_params

capture_status_out

int:

dependencies_out

int:

edge_data_out

int:

static from_buffer(buffer)

Create an CudaStreamGetCaptureInfoV12030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamGetCaptureInfoV12030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_get_capture_info_v12030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamGetCaptureInfoV12030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_out

int:

id_out

int:

num_dependencies_out

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamUpdateCaptureDependenciesV12030Params

Bases: object

Empty-initialize an instance of cudaStreamUpdateCaptureDependencies_v12030_params.

See also

cudaStreamUpdateCaptureDependencies_v12030_params

dependencies

int:

dependency_data

int:

flags_

int:

static from_buffer(buffer)

Create an CudaStreamUpdateCaptureDependenciesV12030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamUpdateCaptureDependenciesV12030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_update_capture_dependencies_v12030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamUpdateCaptureDependenciesV12030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaFuncGetParamCountV13020Params

Bases: object

Empty-initialize an instance of cudaFuncGetParamCount_v13020_params.

See also

cudaFuncGetParamCount_v13020_params

static from_buffer(buffer)

Create an CudaFuncGetParamCountV13020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaFuncGetParamCountV13020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_func_get_param_count_v13020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaFuncGetParamCountV13020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

param_count

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaLaunchHostFuncV2PtszV13020Params

Bases: object

Empty-initialize an instance of cudaLaunchHostFunc_v2_ptsz_v13020_params.

See also

cudaLaunchHostFunc_v2_ptsz_v13020_params

fn

int:

static from_buffer(buffer)

Create an CudaLaunchHostFuncV2PtszV13020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLaunchHostFuncV2PtszV13020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_launch_host_func_v2ptsz_v13020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLaunchHostFuncV2PtszV13020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

sync_mode

int:

user_data

int:

class cupti.cupti.CudaMemcpyWithAttributesAsyncPtszV13020Params

Bases: object

Empty-initialize an instance of cudaMemcpyWithAttributesAsync_ptsz_v13020_params.

See also

cudaMemcpyWithAttributesAsync_ptsz_v13020_params

attr

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyWithAttributesAsyncPtszV13020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyWithAttributesAsyncPtszV13020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_with_attributes_async_ptsz_v13020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyWithAttributesAsyncPtszV13020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

size_

int:

src

int:

stream

int:

class cupti.cupti.CudaMemcpy3DWithAttributesAsyncPtszV13020Params

Bases: object

Empty-initialize an instance of cudaMemcpy3DWithAttributesAsync_ptsz_v13020_params.

See also

cudaMemcpy3DWithAttributesAsync_ptsz_v13020_params

flags_

int:

static from_buffer(buffer)

Create an CudaMemcpy3DWithAttributesAsyncPtszV13020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy3DWithAttributesAsyncPtszV13020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy3d_with_attributes_async_ptsz_v13020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy3DWithAttributesAsyncPtszV13020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

op

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaGraphNodeGetContainingGraphV13010Params

Bases: object

Empty-initialize an instance of cudaGraphNodeGetContainingGraph_v13010_params.

See also

cudaGraphNodeGetContainingGraph_v13010_params

static from_buffer(buffer)

Create an CudaGraphNodeGetContainingGraphV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphNodeGetContainingGraphV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_node_get_containing_graph_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphNodeGetContainingGraphV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ph_graph

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphNodeGetLocalIdV13010Params

Bases: object

Empty-initialize an instance of cudaGraphNodeGetLocalId_v13010_params.

See also

cudaGraphNodeGetLocalId_v13010_params

static from_buffer(buffer)

Create an CudaGraphNodeGetLocalIdV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphNodeGetLocalIdV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_node_get_local_id_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphNodeGetLocalIdV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_id

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphNodeGetToolsIdV13010Params

Bases: object

Empty-initialize an instance of cudaGraphNodeGetToolsId_v13010_params.

See also

cudaGraphNodeGetToolsId_v13010_params

static from_buffer(buffer)

Create an CudaGraphNodeGetToolsIdV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphNodeGetToolsIdV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_node_get_tools_id_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphNodeGetToolsIdV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ptr

Get the pointer address to the data as Python int.

tools_node_id

int:

class cupti.cupti.CudaGraphGetIdV13010Params

Bases: object

Empty-initialize an instance of cudaGraphGetId_v13010_params.

See also

cudaGraphGetId_v13010_params

static from_buffer(buffer)

Create an CudaGraphGetIdV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphGetIdV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_get_id_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphGetIdV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

int:

h_graph

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphExecGetIdV13010Params

Bases: object

Empty-initialize an instance of cudaGraphExecGetId_v13010_params.

See also

cudaGraphExecGetId_v13010_params

static from_buffer(buffer)

Create an CudaGraphExecGetIdV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphExecGetIdV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_exec_get_id_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphExecGetIdV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

int:

h_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphNodeGetParamsV13020Params

Bases: object

Empty-initialize an instance of cudaGraphNodeGetParams_v13020_params.

See also

cudaGraphNodeGetParams_v13020_params

static from_buffer(buffer)

Create an CudaGraphNodeGetParamsV13020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphNodeGetParamsV13020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_node_get_params_v13020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphNodeGetParamsV13020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaGraphConditionalHandleCreateV2V13010Params

Bases: object

Empty-initialize an instance of cudaGraphConditionalHandleCreate_v2_v13010_params.

See also

cudaGraphConditionalHandleCreate_v2_v13010_params

ctx

int:

default_launch_value

int:

flags_

int:

static from_buffer(buffer)

Create an CudaGraphConditionalHandleCreateV2V13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGraphConditionalHandleCreateV2V13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_graph_conditional_handle_create_v2v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGraphConditionalHandleCreateV2V13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

p_handle_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGetDevResourceV13010Params

Bases: object

Empty-initialize an instance of cudaDeviceGetDevResource_v13010_params.

See also

cudaDeviceGetDevResource_v13010_params

device_

int:

static from_buffer(buffer)

Create an CudaDeviceGetDevResourceV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetDevResourceV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_dev_resource_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetDevResourceV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

resource

int:

type

:

class cupti.cupti.CudaDevSmResourceSplitByCountV13010Params

Bases: object

Empty-initialize an instance of cudaDevSmResourceSplitByCount_v13010_params.

See also

cudaDevSmResourceSplitByCount_v13010_params

flags_

int:

static from_buffer(buffer)

Create an CudaDevSmResourceSplitByCountV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDevSmResourceSplitByCountV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_dev_sm_resource_split_by_count_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDevSmResourceSplitByCountV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

input

int:

min_count

int:

nb_groups

int:

ptr

Get the pointer address to the data as Python int.

remaining

int:

result

int:

class cupti.cupti.CudaDevSmResourceSplitV13010Params

Bases: object

Empty-initialize an instance of cudaDevSmResourceSplit_v13010_params.

See also

cudaDevSmResourceSplit_v13010_params

flags_

int:

static from_buffer(buffer)

Create an CudaDevSmResourceSplitV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDevSmResourceSplitV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_dev_sm_resource_split_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDevSmResourceSplitV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

group_params

int:

input

int:

nb_groups

int:

ptr

Get the pointer address to the data as Python int.

remainder

int:

result

int:

class cupti.cupti.CudaDevResourceGenerateDescV13010Params

Bases: object

Empty-initialize an instance of cudaDevResourceGenerateDesc_v13010_params.

See also

cudaDevResourceGenerateDesc_v13010_params

static from_buffer(buffer)

Create an CudaDevResourceGenerateDescV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDevResourceGenerateDescV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_dev_resource_generate_desc_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDevResourceGenerateDescV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

nb_resources

int:

ph_desc

int:

ptr

Get the pointer address to the data as Python int.

resources

int:

class cupti.cupti.CudaGreenCtxCreateV13010Params

Bases: object

Empty-initialize an instance of cudaGreenCtxCreate_v13010_params.

See also

cudaGreenCtxCreate_v13010_params

desc

int:

device_

int:

flags_

int:

static from_buffer(buffer)

Create an CudaGreenCtxCreateV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGreenCtxCreateV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_green_ctx_create_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGreenCtxCreateV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ph_ctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaExecutionCtxDestroyV13010Params

Bases: object

Empty-initialize an instance of cudaExecutionCtxDestroy_v13010_params.

See also

cudaExecutionCtxDestroy_v13010_params

ctx

int:

static from_buffer(buffer)

Create an CudaExecutionCtxDestroyV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaExecutionCtxDestroyV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_execution_ctx_destroy_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaExecutionCtxDestroyV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaExecutionCtxGetDevResourceV13010Params

Bases: object

Empty-initialize an instance of cudaExecutionCtxGetDevResource_v13010_params.

See also

cudaExecutionCtxGetDevResource_v13010_params

ctx

int:

static from_buffer(buffer)

Create an CudaExecutionCtxGetDevResourceV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaExecutionCtxGetDevResourceV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_execution_ctx_get_dev_resource_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaExecutionCtxGetDevResourceV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

resource

int:

type

:

class cupti.cupti.CudaExecutionCtxGetDeviceV13010Params

Bases: object

Empty-initialize an instance of cudaExecutionCtxGetDevice_v13010_params.

See also

cudaExecutionCtxGetDevice_v13010_params

ctx

int:

device_

int:

static from_buffer(buffer)

Create an CudaExecutionCtxGetDeviceV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaExecutionCtxGetDeviceV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_execution_ctx_get_device_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaExecutionCtxGetDeviceV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaExecutionCtxGetIdV13010Params

Bases: object

Empty-initialize an instance of cudaExecutionCtxGetId_v13010_params.

See also

cudaExecutionCtxGetId_v13010_params

ctx

int:

ctx_id

int:

static from_buffer(buffer)

Create an CudaExecutionCtxGetIdV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaExecutionCtxGetIdV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_execution_ctx_get_id_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaExecutionCtxGetIdV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaExecutionCtxStreamCreateV13010Params

Bases: object

Empty-initialize an instance of cudaExecutionCtxStreamCreate_v13010_params.

See also

cudaExecutionCtxStreamCreate_v13010_params

ctx

int:

flags_

int:

static from_buffer(buffer)

Create an CudaExecutionCtxStreamCreateV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaExecutionCtxStreamCreateV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_execution_ctx_stream_create_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaExecutionCtxStreamCreateV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ph_stream

int:

priority

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaExecutionCtxSynchronizeV13010Params

Bases: object

Empty-initialize an instance of cudaExecutionCtxSynchronize_v13010_params.

See also

cudaExecutionCtxSynchronize_v13010_params

ctx

int:

static from_buffer(buffer)

Create an CudaExecutionCtxSynchronizeV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaExecutionCtxSynchronizeV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_execution_ctx_synchronize_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaExecutionCtxSynchronizeV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaStreamGetDevResourcePtszV13010Params

Bases: object

Empty-initialize an instance of cudaStreamGetDevResource_ptsz_v13010_params.

See also

cudaStreamGetDevResource_ptsz_v13010_params

static from_buffer(buffer)

Create an CudaStreamGetDevResourcePtszV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamGetDevResourcePtszV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_get_dev_resource_ptsz_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamGetDevResourcePtszV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

resource

int:

type

:

class cupti.cupti.CudaExecutionCtxRecordEventV13010Params

Bases: object

Empty-initialize an instance of cudaExecutionCtxRecordEvent_v13010_params.

See also

cudaExecutionCtxRecordEvent_v13010_params

ctx

int:

event

int:

static from_buffer(buffer)

Create an CudaExecutionCtxRecordEventV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaExecutionCtxRecordEventV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_execution_ctx_record_event_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaExecutionCtxRecordEventV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaExecutionCtxWaitEventV13010Params

Bases: object

Empty-initialize an instance of cudaExecutionCtxWaitEvent_v13010_params.

See also

cudaExecutionCtxWaitEvent_v13010_params

ctx

int:

event

int:

static from_buffer(buffer)

Create an CudaExecutionCtxWaitEventV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaExecutionCtxWaitEventV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_execution_ctx_wait_event_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaExecutionCtxWaitEventV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaDeviceGetExecutionCtxV13010Params

Bases: object

Empty-initialize an instance of cudaDeviceGetExecutionCtx_v13010_params.

See also

cudaDeviceGetExecutionCtx_v13010_params

ctx

int:

device_

int:

static from_buffer(buffer)

Create an CudaDeviceGetExecutionCtxV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaDeviceGetExecutionCtxV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_device_get_execution_ctx_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaDeviceGetExecutionCtxV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemcpyWithAttributesAsyncV13020Params

Bases: object

Empty-initialize an instance of cudaMemcpyWithAttributesAsync_v13020_params.

See also

cudaMemcpyWithAttributesAsync_v13020_params

attr

int:

dst

int:

static from_buffer(buffer)

Create an CudaMemcpyWithAttributesAsyncV13020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpyWithAttributesAsyncV13020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy_with_attributes_async_v13020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpyWithAttributesAsyncV13020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

size_

int:

src

int:

stream

int:

class cupti.cupti.CudaMemcpy3DWithAttributesAsyncV13020Params

Bases: object

Empty-initialize an instance of cudaMemcpy3DWithAttributesAsync_v13020_params.

See also

cudaMemcpy3DWithAttributesAsync_v13020_params

flags_

int:

static from_buffer(buffer)

Create an CudaMemcpy3DWithAttributesAsyncV13020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemcpy3DWithAttributesAsyncV13020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_memcpy3d_with_attributes_async_v13020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemcpy3DWithAttributesAsyncV13020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

op

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamGetDevResourceV13010Params

Bases: object

Empty-initialize an instance of cudaStreamGetDevResource_v13010_params.

See also

cudaStreamGetDevResource_v13010_params

static from_buffer(buffer)

Create an CudaStreamGetDevResourceV13010Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamGetDevResourceV13010Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_get_dev_resource_v13010params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamGetDevResourceV13010Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

resource

int:

type

:

class cupti.cupti.CudaLaunchHostFuncV2V13020Params

Bases: object

Empty-initialize an instance of cudaLaunchHostFunc_v2_v13020_params.

See also

cudaLaunchHostFunc_v2_v13020_params

fn

int:

static from_buffer(buffer)

Create an CudaLaunchHostFuncV2V13020Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaLaunchHostFuncV2V13020Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_launch_host_func_v2v13020params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaLaunchHostFuncV2V13020Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

sync_mode

int:

user_data

int:

class cupti.cupti.CudaStreamBeginRecaptureToGraphPtszV13030Params

Bases: object

Empty-initialize an instance of cudaStreamBeginRecaptureToGraph_ptsz_v13030_params.

See also

cudaStreamBeginRecaptureToGraph_ptsz_v13030_params

callback_data

int:

static from_buffer(buffer)

Create an CudaStreamBeginRecaptureToGraphPtszV13030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamBeginRecaptureToGraphPtszV13030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_begin_recapture_to_graph_ptsz_v13030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamBeginRecaptureToGraphPtszV13030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

mode

:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaStreamBeginRecaptureToGraphV13030Params

Bases: object

Empty-initialize an instance of cudaStreamBeginRecaptureToGraph_v13030_params.

See also

cudaStreamBeginRecaptureToGraph_v13030_params

callback_data

int:

static from_buffer(buffer)

Create an CudaStreamBeginRecaptureToGraphV13030Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaStreamBeginRecaptureToGraphV13030Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_stream_begin_recapture_to_graph_v13030params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaStreamBeginRecaptureToGraphV13030Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

mode

:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CudaGetExportTableV13000Params

Bases: object

Empty-initialize an instance of cudaGetExportTable_v13000_params.

See also

cudaGetExportTable_v13000_params

static from_buffer(buffer)

Create an CudaGetExportTableV13000Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaGetExportTableV13000Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_get_export_table_v13000params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaGetExportTableV13000Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_export_table_id

int:

pp_export_table

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CudaMemGetLocationInfoV13040Params

Bases: object

Empty-initialize an instance of cudaMemGetLocationInfo_v13040_params.

See also

cudaMemGetLocationInfo_v13040_params

dev_ptr

int:

static from_buffer(buffer)

Create an CudaMemGetLocationInfoV13040Params instance with the memory from the given buffer.

static from_data(data)

Create an CudaMemGetLocationInfoV13040Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cuda_mem_get_location_info_v13040params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CudaMemGetLocationInfoV13040Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

location_out

int:

ptr

Get the pointer address to the data as Python int.

sampling_granularity

int:

size_

int:

summary_granularity

int:

Function Parameter Classes for CUDA Driver APIs

class cupti.cupti.CuGetErrorStringParams

Bases: object

Empty-initialize an instance of cuGetErrorString_params.

See also

cuGetErrorString_params

error

int:

static from_buffer(buffer)

Create an CuGetErrorStringParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGetErrorStringParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_get_error_string_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGetErrorStringParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_str

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGetErrorNameParams

Bases: object

Empty-initialize an instance of cuGetErrorName_params.

See also

cuGetErrorName_params

error

int:

static from_buffer(buffer)

Create an CuGetErrorNameParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGetErrorNameParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_get_error_name_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGetErrorNameParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_str

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuInitParams

Bases: object

Empty-initialize an instance of cuInit_params.

See also

cuInit_params

flags_

int:

static from_buffer(buffer)

Create an CuInitParams instance with the memory from the given buffer.

static from_data(data)

Create an CuInitParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_init_params_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an CuInitParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDriverGetVersionParams

Bases: object

Empty-initialize an instance of cuDriverGetVersion_params.

See also

cuDriverGetVersion_params

driver_version

int:

static from_buffer(buffer)

Create an CuDriverGetVersionParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDriverGetVersionParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_driver_get_version_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDriverGetVersionParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetParams

Bases: object

Empty-initialize an instance of cuDeviceGet_params.

See also

cuDeviceGet_params

device_

int:

static from_buffer(buffer)

Create an CuDeviceGetParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ordinal

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetCountParams

Bases: object

Empty-initialize an instance of cuDeviceGetCount_params.

See also

cuDeviceGetCount_params

count

int:

static from_buffer(buffer)

Create an CuDeviceGetCountParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetCountParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_count_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetCountParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetNameParams

Bases: object

Empty-initialize an instance of cuDeviceGetName_params.

See also

cuDeviceGetName_params

dev

int:

static from_buffer(buffer)

Create an CuDeviceGetNameParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetNameParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_name_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetNameParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

len

int:

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetUuidParams

Bases: object

Empty-initialize an instance of cuDeviceGetUuid_params.

See also

cuDeviceGetUuid_params

dev

int:

static from_buffer(buffer)

Create an CuDeviceGetUuidParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetUuidParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_uuid_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetUuidParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

uuid

int:

class cupti.cupti.CuDeviceGetUuidV2Params

Bases: object

Empty-initialize an instance of cuDeviceGetUuid_v2_params.

See also

cuDeviceGetUuid_v2_params

dev

int:

static from_buffer(buffer)

Create an CuDeviceGetUuidV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetUuidV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_uuid_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetUuidV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

uuid

int:

class cupti.cupti.CuDeviceGetLuidParams

Bases: object

Empty-initialize an instance of cuDeviceGetLuid_params.

See also

cuDeviceGetLuid_params

dev

int:

device_node_mask

int:

static from_buffer(buffer)

Create an CuDeviceGetLuidParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetLuidParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_luid_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetLuidParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

luid

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceTotalMemV2Params

Bases: object

Empty-initialize an instance of cuDeviceTotalMem_v2_params.

See also

cuDeviceTotalMem_v2_params

bytes

int:

dev

int:

static from_buffer(buffer)

Create an CuDeviceTotalMemV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceTotalMemV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_total_mem_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceTotalMemV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetTexture1DLinearMaxWidthParams

Bases: object

Empty-initialize an instance of cuDeviceGetTexture1DLinearMaxWidth_params.

See also

cuDeviceGetTexture1DLinearMaxWidth_params

dev

int:

format

int:

static from_buffer(buffer)

Create an CuDeviceGetTexture1DLinearMaxWidthParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetTexture1DLinearMaxWidthParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_texture1d_linear_max_width_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetTexture1DLinearMaxWidthParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

max_width_in_elements

int:

num_channels

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetAttributeParams

Bases: object

Empty-initialize an instance of cuDeviceGetAttribute_params.

See also

cuDeviceGetAttribute_params

attrib

int:

dev

int:

static from_buffer(buffer)

Create an CuDeviceGetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pi

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetNvSciSyncAttributesParams

Bases: object

Empty-initialize an instance of cuDeviceGetNvSciSyncAttributes_params.

See also

cuDeviceGetNvSciSyncAttributes_params

dev

int:

flags_

int:

static from_buffer(buffer)

Create an CuDeviceGetNvSciSyncAttributesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetNvSciSyncAttributesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_nv_sci_sync_attributes_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetNvSciSyncAttributesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

nv_sci_sync_attr_list

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceSetMemPoolParams

Bases: object

Empty-initialize an instance of cuDeviceSetMemPool_params.

See also

cuDeviceSetMemPool_params

dev

int:

static from_buffer(buffer)

Create an CuDeviceSetMemPoolParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceSetMemPoolParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_set_mem_pool_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceSetMemPoolParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pool

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetMemPoolParams

Bases: object

Empty-initialize an instance of cuDeviceGetMemPool_params.

See also

cuDeviceGetMemPool_params

dev

int:

static from_buffer(buffer)

Create an CuDeviceGetMemPoolParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetMemPoolParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_mem_pool_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetMemPoolParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pool

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetDefaultMemPoolParams

Bases: object

Empty-initialize an instance of cuDeviceGetDefaultMemPool_params.

See also

cuDeviceGetDefaultMemPool_params

dev

int:

static from_buffer(buffer)

Create an CuDeviceGetDefaultMemPoolParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetDefaultMemPoolParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_default_mem_pool_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetDefaultMemPoolParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pool_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetExecAffinitySupportParams

Bases: object

Empty-initialize an instance of cuDeviceGetExecAffinitySupport_params.

See also

cuDeviceGetExecAffinitySupport_params

dev

int:

static from_buffer(buffer)

Create an CuDeviceGetExecAffinitySupportParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetExecAffinitySupportParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_exec_affinity_support_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetExecAffinitySupportParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pi

int:

ptr

Get the pointer address to the data as Python int.

type

int:

class cupti.cupti.CuFlushGPUDirectRDMAWritesParams

Bases: object

Empty-initialize an instance of cuFlushGPUDirectRDMAWrites_params.

See also

cuFlushGPUDirectRDMAWrites_params

static from_buffer(buffer)

Create an CuFlushGPUDirectRDMAWritesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuFlushGPUDirectRDMAWritesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_flush_gpu_direct_rdma_writes_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuFlushGPUDirectRDMAWritesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

scope

int:

target

int:

class cupti.cupti.CuDeviceGetPropertiesParams

Bases: object

Empty-initialize an instance of cuDeviceGetProperties_params.

See also

cuDeviceGetProperties_params

dev

int:

static from_buffer(buffer)

Create an CuDeviceGetPropertiesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetPropertiesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_properties_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetPropertiesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

prop

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceComputeCapabilityParams

Bases: object

Empty-initialize an instance of cuDeviceComputeCapability_params.

See also

cuDeviceComputeCapability_params

dev

int:

static from_buffer(buffer)

Create an CuDeviceComputeCapabilityParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceComputeCapabilityParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_compute_capability_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceComputeCapabilityParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

major

int:

minor

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDevicePrimaryCtxRetainParams

Bases: object

Empty-initialize an instance of cuDevicePrimaryCtxRetain_params.

See also

cuDevicePrimaryCtxRetain_params

dev

int:

static from_buffer(buffer)

Create an CuDevicePrimaryCtxRetainParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDevicePrimaryCtxRetainParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_primary_ctx_retain_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDevicePrimaryCtxRetainParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDevicePrimaryCtxReleaseV2Params

Bases: object

Empty-initialize an instance of cuDevicePrimaryCtxRelease_v2_params.

See also

cuDevicePrimaryCtxRelease_v2_params

dev

int:

static from_buffer(buffer)

Create an CuDevicePrimaryCtxReleaseV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuDevicePrimaryCtxReleaseV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_primary_ctx_release_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDevicePrimaryCtxReleaseV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDevicePrimaryCtxSetFlagsV2Params

Bases: object

Empty-initialize an instance of cuDevicePrimaryCtxSetFlags_v2_params.

See also

cuDevicePrimaryCtxSetFlags_v2_params

dev

int:

flags_

int:

static from_buffer(buffer)

Create an CuDevicePrimaryCtxSetFlagsV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuDevicePrimaryCtxSetFlagsV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_primary_ctx_set_flags_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDevicePrimaryCtxSetFlagsV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDevicePrimaryCtxGetStateParams

Bases: object

Empty-initialize an instance of cuDevicePrimaryCtxGetState_params.

See also

cuDevicePrimaryCtxGetState_params

active

int:

dev

int:

flags_

int:

static from_buffer(buffer)

Create an CuDevicePrimaryCtxGetStateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDevicePrimaryCtxGetStateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_primary_ctx_get_state_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDevicePrimaryCtxGetStateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDevicePrimaryCtxResetV2Params

Bases: object

Empty-initialize an instance of cuDevicePrimaryCtxReset_v2_params.

See also

cuDevicePrimaryCtxReset_v2_params

dev

int:

static from_buffer(buffer)

Create an CuDevicePrimaryCtxResetV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuDevicePrimaryCtxResetV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_primary_ctx_reset_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDevicePrimaryCtxResetV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxCreateV2Params

Bases: object

Empty-initialize an instance of cuCtxCreate_v2_params.

See also

cuCtxCreate_v2_params

dev

int:

flags_

int:

static from_buffer(buffer)

Create an CuCtxCreateV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxCreateV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_create_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxCreateV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxCreateV3Params

Bases: object

Empty-initialize an instance of cuCtxCreate_v3_params.

See also

cuCtxCreate_v3_params

dev

int:

flags_

int:

static from_buffer(buffer)

Create an CuCtxCreateV3Params instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxCreateV3Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_create_v3params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxCreateV3Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_params

int:

params_array

int:

pctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxCreateV4Params

Bases: object

Empty-initialize an instance of cuCtxCreate_v4_params.

See also

cuCtxCreate_v4_params

ctx_create_params

int:

dev

int:

flags_

int:

static from_buffer(buffer)

Create an CuCtxCreateV4Params instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxCreateV4Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_create_v4params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxCreateV4Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxDestroyV2Params

Bases: object

Empty-initialize an instance of cuCtxDestroy_v2_params.

See also

cuCtxDestroy_v2_params

ctx

int:

static from_buffer(buffer)

Create an CuCtxDestroyV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxDestroyV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_destroy_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxDestroyV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxPushCurrentV2Params

Bases: object

Empty-initialize an instance of cuCtxPushCurrent_v2_params.

See also

cuCtxPushCurrent_v2_params

ctx

int:

static from_buffer(buffer)

Create an CuCtxPushCurrentV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxPushCurrentV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_push_current_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxPushCurrentV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxPopCurrentV2Params

Bases: object

Empty-initialize an instance of cuCtxPopCurrent_v2_params.

See also

cuCtxPopCurrent_v2_params

static from_buffer(buffer)

Create an CuCtxPopCurrentV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxPopCurrentV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_pop_current_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxPopCurrentV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxSetCurrentParams

Bases: object

Empty-initialize an instance of cuCtxSetCurrent_params.

See also

cuCtxSetCurrent_params

ctx

int:

static from_buffer(buffer)

Create an CuCtxSetCurrentParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxSetCurrentParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_set_current_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxSetCurrentParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxGetCurrentParams

Bases: object

Empty-initialize an instance of cuCtxGetCurrent_params.

See also

cuCtxGetCurrent_params

static from_buffer(buffer)

Create an CuCtxGetCurrentParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxGetCurrentParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_get_current_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxGetCurrentParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxGetDeviceParams

Bases: object

Empty-initialize an instance of cuCtxGetDevice_params.

See also

cuCtxGetDevice_params

device_

int:

static from_buffer(buffer)

Create an CuCtxGetDeviceParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxGetDeviceParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_get_device_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxGetDeviceParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxGetFlagsParams

Bases: object

Empty-initialize an instance of cuCtxGetFlags_params.

See also

cuCtxGetFlags_params

flags_

int:

static from_buffer(buffer)

Create an CuCtxGetFlagsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxGetFlagsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_get_flags_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxGetFlagsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxSetFlagsParams

Bases: object

Empty-initialize an instance of cuCtxSetFlags_params.

See also

cuCtxSetFlags_params

flags_

int:

static from_buffer(buffer)

Create an CuCtxSetFlagsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxSetFlagsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_set_flags_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxSetFlagsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxGetIdParams

Bases: object

Empty-initialize an instance of cuCtxGetId_params.

See also

cuCtxGetId_params

ctx

int:

ctx_id

int:

static from_buffer(buffer)

Create an CuCtxGetIdParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxGetIdParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_get_id_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxGetIdParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxSetLimitParams

Bases: object

Empty-initialize an instance of cuCtxSetLimit_params.

See also

cuCtxSetLimit_params

static from_buffer(buffer)

Create an CuCtxSetLimitParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxSetLimitParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_set_limit_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxSetLimitParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

limit

int:

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CuCtxGetLimitParams

Bases: object

Empty-initialize an instance of cuCtxGetLimit_params.

See also

cuCtxGetLimit_params

static from_buffer(buffer)

Create an CuCtxGetLimitParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxGetLimitParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_get_limit_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxGetLimitParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

limit

int:

ptr

Get the pointer address to the data as Python int.

pvalue

int:

class cupti.cupti.CuCtxGetCacheConfigParams

Bases: object

Empty-initialize an instance of cuCtxGetCacheConfig_params.

See also

cuCtxGetCacheConfig_params

static from_buffer(buffer)

Create an CuCtxGetCacheConfigParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxGetCacheConfigParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_get_cache_config_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxGetCacheConfigParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pconfig

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxSetCacheConfigParams

Bases: object

Empty-initialize an instance of cuCtxSetCacheConfig_params.

See also

cuCtxSetCacheConfig_params

config

int:

static from_buffer(buffer)

Create an CuCtxSetCacheConfigParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxSetCacheConfigParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_set_cache_config_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxSetCacheConfigParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxGetApiVersionParams

Bases: object

Empty-initialize an instance of cuCtxGetApiVersion_params.

See also

cuCtxGetApiVersion_params

ctx

int:

static from_buffer(buffer)

Create an CuCtxGetApiVersionParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxGetApiVersionParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_get_api_version_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxGetApiVersionParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

version

int:

class cupti.cupti.CuCtxGetStreamPriorityRangeParams

Bases: object

Empty-initialize an instance of cuCtxGetStreamPriorityRange_params.

See also

cuCtxGetStreamPriorityRange_params

static from_buffer(buffer)

Create an CuCtxGetStreamPriorityRangeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxGetStreamPriorityRangeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_get_stream_priority_range_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxGetStreamPriorityRangeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

greatest_priority

int:

least_priority

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxGetExecAffinityParams

Bases: object

Empty-initialize an instance of cuCtxGetExecAffinity_params.

See also

cuCtxGetExecAffinity_params

static from_buffer(buffer)

Create an CuCtxGetExecAffinityParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxGetExecAffinityParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_get_exec_affinity_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxGetExecAffinityParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_exec_affinity

int:

ptr

Get the pointer address to the data as Python int.

type

int:

class cupti.cupti.CuCtxRecordEventParams

Bases: object

Empty-initialize an instance of cuCtxRecordEvent_params.

See also

cuCtxRecordEvent_params

static from_buffer(buffer)

Create an CuCtxRecordEventParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxRecordEventParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_record_event_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxRecordEventParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_ctx

int:

h_event

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxWaitEventParams

Bases: object

Empty-initialize an instance of cuCtxWaitEvent_params.

See also

cuCtxWaitEvent_params

static from_buffer(buffer)

Create an CuCtxWaitEventParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxWaitEventParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_wait_event_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxWaitEventParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_ctx

int:

h_event

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxAttachParams

Bases: object

Empty-initialize an instance of cuCtxAttach_params.

See also

cuCtxAttach_params

flags_

int:

static from_buffer(buffer)

Create an CuCtxAttachParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxAttachParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_attach_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxAttachParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxDetachParams

Bases: object

Empty-initialize an instance of cuCtxDetach_params.

See also

cuCtxDetach_params

ctx

int:

static from_buffer(buffer)

Create an CuCtxDetachParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxDetachParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_detach_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxDetachParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxGetSharedMemConfigParams

Bases: object

Empty-initialize an instance of cuCtxGetSharedMemConfig_params.

See also

cuCtxGetSharedMemConfig_params

static from_buffer(buffer)

Create an CuCtxGetSharedMemConfigParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxGetSharedMemConfigParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_get_shared_mem_config_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxGetSharedMemConfigParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_config

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxSetSharedMemConfigParams

Bases: object

Empty-initialize an instance of cuCtxSetSharedMemConfig_params.

See also

cuCtxSetSharedMemConfig_params

config

int:

static from_buffer(buffer)

Create an CuCtxSetSharedMemConfigParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxSetSharedMemConfigParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_set_shared_mem_config_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxSetSharedMemConfigParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuModuleLoadParams

Bases: object

Empty-initialize an instance of cuModuleLoad_params.

See also

cuModuleLoad_params

fname

str:

static from_buffer(buffer)

Create an CuModuleLoadParams instance with the memory from the given buffer.

static from_data(data)

Create an CuModuleLoadParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_module_load_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuModuleLoadParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

module

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuModuleLoadDataParams

Bases: object

Empty-initialize an instance of cuModuleLoadData_params.

See also

cuModuleLoadData_params

static from_buffer(buffer)

Create an CuModuleLoadDataParams instance with the memory from the given buffer.

static from_data(data)

Create an CuModuleLoadDataParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_module_load_data_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuModuleLoadDataParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

image

int:

module

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuModuleLoadDataExParams

Bases: object

Empty-initialize an instance of cuModuleLoadDataEx_params.

See also

cuModuleLoadDataEx_params

static from_buffer(buffer)

Create an CuModuleLoadDataExParams instance with the memory from the given buffer.

static from_data(data)

Create an CuModuleLoadDataExParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_module_load_data_ex_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuModuleLoadDataExParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

image

int:

module

int:

num_options

int:

option_values

int:

options

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuModuleLoadFatBinaryParams

Bases: object

Empty-initialize an instance of cuModuleLoadFatBinary_params.

See also

cuModuleLoadFatBinary_params

fat_cubin

int:

static from_buffer(buffer)

Create an CuModuleLoadFatBinaryParams instance with the memory from the given buffer.

static from_data(data)

Create an CuModuleLoadFatBinaryParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_module_load_fat_binary_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuModuleLoadFatBinaryParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

module

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuModuleUnloadParams

Bases: object

Empty-initialize an instance of cuModuleUnload_params.

See also

cuModuleUnload_params

static from_buffer(buffer)

Create an CuModuleUnloadParams instance with the memory from the given buffer.

static from_data(data)

Create an CuModuleUnloadParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_module_unload_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuModuleUnloadParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hmod

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuModuleGetLoadingModeParams

Bases: object

Empty-initialize an instance of cuModuleGetLoadingMode_params.

See also

cuModuleGetLoadingMode_params

static from_buffer(buffer)

Create an CuModuleGetLoadingModeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuModuleGetLoadingModeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_module_get_loading_mode_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuModuleGetLoadingModeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mode

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuModuleGetFunctionParams

Bases: object

Empty-initialize an instance of cuModuleGetFunction_params.

See also

cuModuleGetFunction_params

static from_buffer(buffer)

Create an CuModuleGetFunctionParams instance with the memory from the given buffer.

static from_data(data)

Create an CuModuleGetFunctionParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_module_get_function_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuModuleGetFunctionParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hfunc

int:

hmod

int:

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuModuleGetFunctionCountParams

Bases: object

Empty-initialize an instance of cuModuleGetFunctionCount_params.

See also

cuModuleGetFunctionCount_params

count

int:

static from_buffer(buffer)

Create an CuModuleGetFunctionCountParams instance with the memory from the given buffer.

static from_data(data)

Create an CuModuleGetFunctionCountParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_module_get_function_count_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuModuleGetFunctionCountParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mod

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuModuleEnumerateFunctionsParams

Bases: object

Empty-initialize an instance of cuModuleEnumerateFunctions_params.

See also

cuModuleEnumerateFunctions_params

static from_buffer(buffer)

Create an CuModuleEnumerateFunctionsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuModuleEnumerateFunctionsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_module_enumerate_functions_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuModuleEnumerateFunctionsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

functions

int:

mod

int:

num_functions

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuModuleGetGlobalV2Params

Bases: object

Empty-initialize an instance of cuModuleGetGlobal_v2_params.

See also

cuModuleGetGlobal_v2_params

bytes

int:

dptr

int:

static from_buffer(buffer)

Create an CuModuleGetGlobalV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuModuleGetGlobalV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_module_get_global_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuModuleGetGlobalV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hmod

int:

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLinkCreateV2Params

Bases: object

Empty-initialize an instance of cuLinkCreate_v2_params.

See also

cuLinkCreate_v2_params

static from_buffer(buffer)

Create an CuLinkCreateV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuLinkCreateV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_link_create_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLinkCreateV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_options

int:

option_values

int:

options

int:

ptr

Get the pointer address to the data as Python int.

state_out

int:

class cupti.cupti.CuLinkAddDataV2Params

Bases: object

Empty-initialize an instance of cuLinkAddData_v2_params.

See also

cuLinkAddData_v2_params

data_

int:

static from_buffer(buffer)

Create an CuLinkAddDataV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuLinkAddDataV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_link_add_data_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLinkAddDataV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

str:

num_options

int:

option_values

int:

options

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

state

int:

type

int:

class cupti.cupti.CuLinkAddFileV2Params

Bases: object

Empty-initialize an instance of cuLinkAddFile_v2_params.

See also

cuLinkAddFile_v2_params

static from_buffer(buffer)

Create an CuLinkAddFileV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuLinkAddFileV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_link_add_file_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLinkAddFileV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_options

int:

option_values

int:

options

int:

path

str:

ptr

Get the pointer address to the data as Python int.

state

int:

type

int:

class cupti.cupti.CuLinkCompleteParams

Bases: object

Empty-initialize an instance of cuLinkComplete_params.

See also

cuLinkComplete_params

cubin_out

int:

static from_buffer(buffer)

Create an CuLinkCompleteParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLinkCompleteParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_link_complete_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLinkCompleteParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

size_out

int:

state

int:

class cupti.cupti.CuLinkDestroyParams

Bases: object

Empty-initialize an instance of cuLinkDestroy_params.

See also

cuLinkDestroy_params

static from_buffer(buffer)

Create an CuLinkDestroyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLinkDestroyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_link_destroy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLinkDestroyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

state

int:

class cupti.cupti.CuModuleGetTexRefParams

Bases: object

Empty-initialize an instance of cuModuleGetTexRef_params.

See also

cuModuleGetTexRef_params

static from_buffer(buffer)

Create an CuModuleGetTexRefParams instance with the memory from the given buffer.

static from_data(data)

Create an CuModuleGetTexRefParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_module_get_tex_ref_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuModuleGetTexRefParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hmod

int:

name

str:

p_tex_ref

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuModuleGetSurfRefParams

Bases: object

Empty-initialize an instance of cuModuleGetSurfRef_params.

See also

cuModuleGetSurfRef_params

static from_buffer(buffer)

Create an CuModuleGetSurfRefParams instance with the memory from the given buffer.

static from_data(data)

Create an CuModuleGetSurfRefParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_module_get_surf_ref_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuModuleGetSurfRefParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hmod

int:

name

str:

p_surf_ref

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLibraryLoadDataParams

Bases: object

Empty-initialize an instance of cuLibraryLoadData_params.

See also

cuLibraryLoadData_params

code

int:

static from_buffer(buffer)

Create an CuLibraryLoadDataParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLibraryLoadDataParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_library_load_data_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLibraryLoadDataParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

jit_options

int:

jit_options_values

int:

library

int:

library_option_values

int:

library_options

int:

num_jit_options

int:

num_library_options

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLibraryLoadFromFileParams

Bases: object

Empty-initialize an instance of cuLibraryLoadFromFile_params.

See also

cuLibraryLoadFromFile_params

file_name

str:

static from_buffer(buffer)

Create an CuLibraryLoadFromFileParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLibraryLoadFromFileParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_library_load_from_file_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLibraryLoadFromFileParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

jit_options

int:

jit_options_values

int:

library

int:

library_option_values

int:

library_options

int:

num_jit_options

int:

num_library_options

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLibraryUnloadParams

Bases: object

Empty-initialize an instance of cuLibraryUnload_params.

See also

cuLibraryUnload_params

static from_buffer(buffer)

Create an CuLibraryUnloadParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLibraryUnloadParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_library_unload_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLibraryUnloadParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

library

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLibraryGetKernelParams

Bases: object

Empty-initialize an instance of cuLibraryGetKernel_params.

See also

cuLibraryGetKernel_params

static from_buffer(buffer)

Create an CuLibraryGetKernelParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLibraryGetKernelParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_library_get_kernel_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLibraryGetKernelParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

library

int:

name

str:

p_kernel

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLibraryGetKernelCountParams

Bases: object

Empty-initialize an instance of cuLibraryGetKernelCount_params.

See also

cuLibraryGetKernelCount_params

count

int:

static from_buffer(buffer)

Create an CuLibraryGetKernelCountParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLibraryGetKernelCountParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_library_get_kernel_count_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLibraryGetKernelCountParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

lib

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLibraryEnumerateKernelsParams

Bases: object

Empty-initialize an instance of cuLibraryEnumerateKernels_params.

See also

cuLibraryEnumerateKernels_params

static from_buffer(buffer)

Create an CuLibraryEnumerateKernelsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLibraryEnumerateKernelsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_library_enumerate_kernels_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLibraryEnumerateKernelsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kernels

int:

lib

int:

num_kernels

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLibraryGetModuleParams

Bases: object

Empty-initialize an instance of cuLibraryGetModule_params.

See also

cuLibraryGetModule_params

static from_buffer(buffer)

Create an CuLibraryGetModuleParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLibraryGetModuleParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_library_get_module_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLibraryGetModuleParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

library

int:

p_mod

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuKernelGetFunctionParams

Bases: object

Empty-initialize an instance of cuKernelGetFunction_params.

See also

cuKernelGetFunction_params

static from_buffer(buffer)

Create an CuKernelGetFunctionParams instance with the memory from the given buffer.

static from_data(data)

Create an CuKernelGetFunctionParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_kernel_get_function_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuKernelGetFunctionParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kernel

int:

p_func

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuKernelGetLibraryParams

Bases: object

Empty-initialize an instance of cuKernelGetLibrary_params.

See also

cuKernelGetLibrary_params

static from_buffer(buffer)

Create an CuKernelGetLibraryParams instance with the memory from the given buffer.

static from_data(data)

Create an CuKernelGetLibraryParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_kernel_get_library_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuKernelGetLibraryParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kernel

int:

p_lib

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLibraryGetGlobalParams

Bases: object

Empty-initialize an instance of cuLibraryGetGlobal_params.

See also

cuLibraryGetGlobal_params

bytes

int:

dptr

int:

static from_buffer(buffer)

Create an CuLibraryGetGlobalParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLibraryGetGlobalParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_library_get_global_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLibraryGetGlobalParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

library

int:

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLibraryGetManagedParams

Bases: object

Empty-initialize an instance of cuLibraryGetManaged_params.

See also

cuLibraryGetManaged_params

bytes

int:

dptr

int:

static from_buffer(buffer)

Create an CuLibraryGetManagedParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLibraryGetManagedParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_library_get_managed_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLibraryGetManagedParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

library

int:

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLibraryGetUnifiedFunctionParams

Bases: object

Empty-initialize an instance of cuLibraryGetUnifiedFunction_params.

See also

cuLibraryGetUnifiedFunction_params

fptr

int:

static from_buffer(buffer)

Create an CuLibraryGetUnifiedFunctionParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLibraryGetUnifiedFunctionParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_library_get_unified_function_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLibraryGetUnifiedFunctionParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

library

int:

ptr

Get the pointer address to the data as Python int.

symbol

str:

class cupti.cupti.CuKernelGetAttributeParams

Bases: object

Empty-initialize an instance of cuKernelGetAttribute_params.

See also

cuKernelGetAttribute_params

attrib

int:

dev

int:

static from_buffer(buffer)

Create an CuKernelGetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuKernelGetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_kernel_get_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuKernelGetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kernel

int:

pi

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuKernelSetAttributeParams

Bases: object

Empty-initialize an instance of cuKernelSetAttribute_params.

See also

cuKernelSetAttribute_params

attrib

int:

dev

int:

static from_buffer(buffer)

Create an CuKernelSetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuKernelSetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_kernel_set_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuKernelSetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kernel

int:

ptr

Get the pointer address to the data as Python int.

val

int:

class cupti.cupti.CuKernelSetCacheConfigParams

Bases: object

Empty-initialize an instance of cuKernelSetCacheConfig_params.

See also

cuKernelSetCacheConfig_params

config

int:

dev

int:

static from_buffer(buffer)

Create an CuKernelSetCacheConfigParams instance with the memory from the given buffer.

static from_data(data)

Create an CuKernelSetCacheConfigParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_kernel_set_cache_config_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuKernelSetCacheConfigParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kernel

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuKernelGetNameParams

Bases: object

Empty-initialize an instance of cuKernelGetName_params.

See also

cuKernelGetName_params

static from_buffer(buffer)

Create an CuKernelGetNameParams instance with the memory from the given buffer.

static from_data(data)

Create an CuKernelGetNameParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_kernel_get_name_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuKernelGetNameParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hfunc

int:

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuKernelGetParamInfoParams

Bases: object

Empty-initialize an instance of cuKernelGetParamInfo_params.

See also

cuKernelGetParamInfo_params

static from_buffer(buffer)

Create an CuKernelGetParamInfoParams instance with the memory from the given buffer.

static from_data(data)

Create an CuKernelGetParamInfoParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_kernel_get_param_info_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuKernelGetParamInfoParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kernel

int:

param_index

int:

param_offset

int:

param_size

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemGetInfoV2Params

Bases: object

Empty-initialize an instance of cuMemGetInfo_v2_params.

See also

cuMemGetInfo_v2_params

free

int:

static from_buffer(buffer)

Create an CuMemGetInfoV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemGetInfoV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_get_info_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemGetInfoV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

total

int:

class cupti.cupti.CuMemAllocV2Params

Bases: object

Empty-initialize an instance of cuMemAlloc_v2_params.

See also

cuMemAlloc_v2_params

bytesize

int:

dptr

int:

static from_buffer(buffer)

Create an CuMemAllocV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemAllocV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_alloc_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemAllocV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemAllocPitchV2Params

Bases: object

Empty-initialize an instance of cuMemAllocPitch_v2_params.

See also

cuMemAllocPitch_v2_params

dptr

int:

element_size_bytes

int:

static from_buffer(buffer)

Create an CuMemAllocPitchV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemAllocPitchV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_alloc_pitch_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemAllocPitchV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

p_pitch

int:

ptr

Get the pointer address to the data as Python int.

width_in_bytes

int:

class cupti.cupti.CuMemFreeV2Params

Bases: object

Empty-initialize an instance of cuMemFree_v2_params.

See also

cuMemFree_v2_params

dptr

int:

static from_buffer(buffer)

Create an CuMemFreeV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemFreeV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_free_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemFreeV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemGetAddressRangeV2Params

Bases: object

Empty-initialize an instance of cuMemGetAddressRange_v2_params.

See also

cuMemGetAddressRange_v2_params

dptr

int:

static from_buffer(buffer)

Create an CuMemGetAddressRangeV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemGetAddressRangeV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_get_address_range_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemGetAddressRangeV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pbase

int:

psize

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemAllocHostV2Params

Bases: object

Empty-initialize an instance of cuMemAllocHost_v2_params.

See also

cuMemAllocHost_v2_params

bytesize

int:

static from_buffer(buffer)

Create an CuMemAllocHostV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemAllocHostV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_alloc_host_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemAllocHostV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pp

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemFreeHostParams

Bases: object

Empty-initialize an instance of cuMemFreeHost_params.

See also

cuMemFreeHost_params

static from_buffer(buffer)

Create an CuMemFreeHostParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemFreeHostParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_free_host_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemFreeHostParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemHostAllocParams

Bases: object

Empty-initialize an instance of cuMemHostAlloc_params.

See also

cuMemHostAlloc_params

bytesize

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemHostAllocParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemHostAllocParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_host_alloc_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemHostAllocParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pp

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemHostGetDevicePointerV2Params

Bases: object

Empty-initialize an instance of cuMemHostGetDevicePointer_v2_params.

See also

cuMemHostGetDevicePointer_v2_params

flags_

int:

static from_buffer(buffer)

Create an CuMemHostGetDevicePointerV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemHostGetDevicePointerV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_host_get_device_pointer_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemHostGetDevicePointerV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p

int:

pdptr

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemHostGetFlagsParams

Bases: object

Empty-initialize an instance of cuMemHostGetFlags_params.

See also

cuMemHostGetFlags_params

static from_buffer(buffer)

Create an CuMemHostGetFlagsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemHostGetFlagsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_host_get_flags_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemHostGetFlagsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p

int:

p_flags

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemAllocManagedParams

Bases: object

Empty-initialize an instance of cuMemAllocManaged_params.

See also

cuMemAllocManaged_params

bytesize

int:

dptr

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemAllocManagedParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemAllocManagedParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_alloc_managed_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemAllocManagedParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceRegisterAsyncNotificationParams

Bases: object

Empty-initialize an instance of cuDeviceRegisterAsyncNotification_params.

See also

cuDeviceRegisterAsyncNotification_params

callback

int:

callback_func

int:

device_

int:

static from_buffer(buffer)

Create an CuDeviceRegisterAsyncNotificationParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceRegisterAsyncNotificationParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_register_async_notification_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceRegisterAsyncNotificationParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

user_data

int:

class cupti.cupti.CuDeviceUnregisterAsyncNotificationParams

Bases: object

Empty-initialize an instance of cuDeviceUnregisterAsyncNotification_params.

See also

cuDeviceUnregisterAsyncNotification_params

callback

int:

device_

int:

static from_buffer(buffer)

Create an CuDeviceUnregisterAsyncNotificationParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceUnregisterAsyncNotificationParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_unregister_async_notification_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceUnregisterAsyncNotificationParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetByPCIBusIdParams

Bases: object

Empty-initialize an instance of cuDeviceGetByPCIBusId_params.

See also

cuDeviceGetByPCIBusId_params

dev

int:

static from_buffer(buffer)

Create an CuDeviceGetByPCIBusIdParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetByPCIBusIdParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_by_pci_bus_id_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetByPCIBusIdParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pci_bus_id

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetPCIBusIdParams

Bases: object

Empty-initialize an instance of cuDeviceGetPCIBusId_params.

See also

cuDeviceGetPCIBusId_params

dev

int:

static from_buffer(buffer)

Create an CuDeviceGetPCIBusIdParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetPCIBusIdParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_pci_bus_id_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetPCIBusIdParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

len

int:

pci_bus_id

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuIpcGetEventHandleParams

Bases: object

Empty-initialize an instance of cuIpcGetEventHandle_params.

See also

cuIpcGetEventHandle_params

event

int:

static from_buffer(buffer)

Create an CuIpcGetEventHandleParams instance with the memory from the given buffer.

static from_data(data)

Create an CuIpcGetEventHandleParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ipc_get_event_handle_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuIpcGetEventHandleParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_handle

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuIpcGetMemHandleParams

Bases: object

Empty-initialize an instance of cuIpcGetMemHandle_params.

See also

cuIpcGetMemHandle_params

dptr

int:

static from_buffer(buffer)

Create an CuIpcGetMemHandleParams instance with the memory from the given buffer.

static from_data(data)

Create an CuIpcGetMemHandleParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ipc_get_mem_handle_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuIpcGetMemHandleParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_handle

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuIpcCloseMemHandleParams

Bases: object

Empty-initialize an instance of cuIpcCloseMemHandle_params.

See also

cuIpcCloseMemHandle_params

dptr

int:

static from_buffer(buffer)

Create an CuIpcCloseMemHandleParams instance with the memory from the given buffer.

static from_data(data)

Create an CuIpcCloseMemHandleParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ipc_close_mem_handle_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuIpcCloseMemHandleParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemHostRegisterV2Params

Bases: object

Empty-initialize an instance of cuMemHostRegister_v2_params.

See also

cuMemHostRegister_v2_params

bytesize

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemHostRegisterV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemHostRegisterV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_host_register_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemHostRegisterV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemHostUnregisterParams

Bases: object

Empty-initialize an instance of cuMemHostUnregister_params.

See also

cuMemHostUnregister_params

static from_buffer(buffer)

Create an CuMemHostUnregisterParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemHostUnregisterParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_host_unregister_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemHostUnregisterParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpyPtdsParams

Bases: object

Empty-initialize an instance of cuMemcpy_ptds_params.

See also

cuMemcpy_ptds_params

byte_count

int:

dst

int:

static from_buffer(buffer)

Create an CuMemcpyPtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyPtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyPtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src

int:

class cupti.cupti.CuMemcpyPeerPtdsParams

Bases: object

Empty-initialize an instance of cuMemcpyPeer_ptds_params.

See also

cuMemcpyPeer_ptds_params

byte_count

int:

dst_context

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyPeerPtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyPeerPtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_peer_ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyPeerPtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_context

int:

src_device

int:

class cupti.cupti.CuMemcpyHtoDV2PtdsParams

Bases: object

Empty-initialize an instance of cuMemcpyHtoD_v2_ptds_params.

See also

cuMemcpyHtoD_v2_ptds_params

byte_count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyHtoDV2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyHtoDV2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_hto_d_v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyHtoDV2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_host

int:

class cupti.cupti.CuMemcpyDtoHV2PtdsParams

Bases: object

Empty-initialize an instance of cuMemcpyDtoH_v2_ptds_params.

See also

cuMemcpyDtoH_v2_ptds_params

byte_count

int:

dst_host

int:

static from_buffer(buffer)

Create an CuMemcpyDtoHV2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyDtoHV2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_dto_h_v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyDtoHV2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuMemcpyDtoDV2PtdsParams

Bases: object

Empty-initialize an instance of cuMemcpyDtoD_v2_ptds_params.

See also

cuMemcpyDtoD_v2_ptds_params

byte_count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyDtoDV2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyDtoDV2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_dto_d_v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyDtoDV2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuMemcpyDtoAV2PtdsParams

Bases: object

Empty-initialize an instance of cuMemcpyDtoA_v2_ptds_params.

See also

cuMemcpyDtoA_v2_ptds_params

byte_count

int:

dst_array

int:

dst_offset

int:

static from_buffer(buffer)

Create an CuMemcpyDtoAV2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyDtoAV2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_dto_a_v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyDtoAV2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuMemcpyAtoDV2PtdsParams

Bases: object

Empty-initialize an instance of cuMemcpyAtoD_v2_ptds_params.

See also

cuMemcpyAtoD_v2_ptds_params

byte_count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyAtoDV2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyAtoDV2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_ato_d_v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyAtoDV2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_array

int:

src_offset

int:

class cupti.cupti.CuMemcpyHtoAV2PtdsParams

Bases: object

Empty-initialize an instance of cuMemcpyHtoA_v2_ptds_params.

See also

cuMemcpyHtoA_v2_ptds_params

byte_count

int:

dst_array

int:

dst_offset

int:

static from_buffer(buffer)

Create an CuMemcpyHtoAV2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyHtoAV2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_hto_a_v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyHtoAV2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_host

int:

class cupti.cupti.CuMemcpyAtoHV2PtdsParams

Bases: object

Empty-initialize an instance of cuMemcpyAtoH_v2_ptds_params.

See also

cuMemcpyAtoH_v2_ptds_params

byte_count

int:

dst_host

int:

static from_buffer(buffer)

Create an CuMemcpyAtoHV2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyAtoHV2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_ato_h_v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyAtoHV2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_array

int:

src_offset

int:

class cupti.cupti.CuMemcpyAtoAV2PtdsParams

Bases: object

Empty-initialize an instance of cuMemcpyAtoA_v2_ptds_params.

See also

cuMemcpyAtoA_v2_ptds_params

byte_count

int:

dst_array

int:

dst_offset

int:

static from_buffer(buffer)

Create an CuMemcpyAtoAV2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyAtoAV2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_ato_a_v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyAtoAV2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_array

int:

src_offset

int:

class cupti.cupti.CuMemcpy2DV2PtdsParams

Bases: object

Empty-initialize an instance of cuMemcpy2D_v2_ptds_params.

See also

cuMemcpy2D_v2_ptds_params

static from_buffer(buffer)

Create an CuMemcpy2DV2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy2DV2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy2d_v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy2DV2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpy2DUnalignedV2PtdsParams

Bases: object

Empty-initialize an instance of cuMemcpy2DUnaligned_v2_ptds_params.

See also

cuMemcpy2DUnaligned_v2_ptds_params

static from_buffer(buffer)

Create an CuMemcpy2DUnalignedV2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy2DUnalignedV2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy2d_unaligned_v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy2DUnalignedV2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpy3DV2PtdsParams

Bases: object

Empty-initialize an instance of cuMemcpy3D_v2_ptds_params.

See also

cuMemcpy3D_v2_ptds_params

static from_buffer(buffer)

Create an CuMemcpy3DV2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DV2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3d_v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DV2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpy3DPeerPtdsParams

Bases: object

Empty-initialize an instance of cuMemcpy3DPeer_ptds_params.

See also

cuMemcpy3DPeer_ptds_params

static from_buffer(buffer)

Create an CuMemcpy3DPeerPtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DPeerPtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3d_peer_ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DPeerPtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpyAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemcpyAsync_ptsz_params.

See also

cuMemcpyAsync_ptsz_params

byte_count

int:

dst

int:

static from_buffer(buffer)

Create an CuMemcpyAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src

int:

class cupti.cupti.CuMemcpyPeerAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemcpyPeerAsync_ptsz_params.

See also

cuMemcpyPeerAsync_ptsz_params

byte_count

int:

dst_context

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyPeerAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyPeerAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_peer_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyPeerAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_context

int:

src_device

int:

class cupti.cupti.CuMemcpyHtoDAsyncV2PtszParams

Bases: object

Empty-initialize an instance of cuMemcpyHtoDAsync_v2_ptsz_params.

See also

cuMemcpyHtoDAsync_v2_ptsz_params

byte_count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyHtoDAsyncV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyHtoDAsyncV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_hto_dasync_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyHtoDAsyncV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_host

int:

class cupti.cupti.CuMemcpyDtoHAsyncV2PtszParams

Bases: object

Empty-initialize an instance of cuMemcpyDtoHAsync_v2_ptsz_params.

See also

cuMemcpyDtoHAsync_v2_ptsz_params

byte_count

int:

dst_host

int:

static from_buffer(buffer)

Create an CuMemcpyDtoHAsyncV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyDtoHAsyncV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_dto_hasync_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyDtoHAsyncV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuMemcpyDtoDAsyncV2PtszParams

Bases: object

Empty-initialize an instance of cuMemcpyDtoDAsync_v2_ptsz_params.

See also

cuMemcpyDtoDAsync_v2_ptsz_params

byte_count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyDtoDAsyncV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyDtoDAsyncV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_dto_dasync_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyDtoDAsyncV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuMemcpyHtoAAsyncV2PtszParams

Bases: object

Empty-initialize an instance of cuMemcpyHtoAAsync_v2_ptsz_params.

See also

cuMemcpyHtoAAsync_v2_ptsz_params

byte_count

int:

dst_array

int:

dst_offset

int:

static from_buffer(buffer)

Create an CuMemcpyHtoAAsyncV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyHtoAAsyncV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_hto_aasync_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyHtoAAsyncV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_host

int:

class cupti.cupti.CuMemcpyAtoHAsyncV2PtszParams

Bases: object

Empty-initialize an instance of cuMemcpyAtoHAsync_v2_ptsz_params.

See also

cuMemcpyAtoHAsync_v2_ptsz_params

byte_count

int:

dst_host

int:

static from_buffer(buffer)

Create an CuMemcpyAtoHAsyncV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyAtoHAsyncV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_ato_hasync_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyAtoHAsyncV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_array

int:

src_offset

int:

class cupti.cupti.CuMemcpy2DAsyncV2PtszParams

Bases: object

Empty-initialize an instance of cuMemcpy2DAsync_v2_ptsz_params.

See also

cuMemcpy2DAsync_v2_ptsz_params

static from_buffer(buffer)

Create an CuMemcpy2DAsyncV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy2DAsyncV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy2_dasync_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy2DAsyncV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpy3DAsyncV2PtszParams

Bases: object

Empty-initialize an instance of cuMemcpy3DAsync_v2_ptsz_params.

See also

cuMemcpy3DAsync_v2_ptsz_params

static from_buffer(buffer)

Create an CuMemcpy3DAsyncV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DAsyncV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3_dasync_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DAsyncV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpy3DPeerAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemcpy3DPeerAsync_ptsz_params.

See also

cuMemcpy3DPeerAsync_ptsz_params

static from_buffer(buffer)

Create an CuMemcpy3DPeerAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DPeerAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3d_peer_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DPeerAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpyBatchAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemcpyBatchAsync_ptsz_params.

See also

cuMemcpyBatchAsync_ptsz_params

attrs

int:

attrs_idxs

int:

count

int:

dsts

int:

fail_idx

int:

static from_buffer(buffer)

Create an CuMemcpyBatchAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyBatchAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_batch_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyBatchAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_attrs

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

srcs

int:

class cupti.cupti.CuMemcpy3DBatchAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemcpy3DBatchAsync_ptsz_params.

See also

cuMemcpy3DBatchAsync_ptsz_params

fail_idx

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemcpy3DBatchAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DBatchAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3d_batch_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DBatchAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_ops

int:

op_list

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemsetD8V2PtdsParams

Bases: object

Empty-initialize an instance of cuMemsetD8_v2_ptds_params.

See also

cuMemsetD8_v2_ptds_params

dst_device

int:

static from_buffer(buffer)

Create an CuMemsetD8V2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD8V2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d8v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD8V2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

n

int:

ptr

Get the pointer address to the data as Python int.

uc

int:

class cupti.cupti.CuMemsetD16V2PtdsParams

Bases: object

Empty-initialize an instance of cuMemsetD16_v2_ptds_params.

See also

cuMemsetD16_v2_ptds_params

dst_device

int:

static from_buffer(buffer)

Create an CuMemsetD16V2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD16V2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d16v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD16V2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

n

int:

ptr

Get the pointer address to the data as Python int.

us

int:

class cupti.cupti.CuMemsetD32V2PtdsParams

Bases: object

Empty-initialize an instance of cuMemsetD32_v2_ptds_params.

See also

cuMemsetD32_v2_ptds_params

dst_device

int:

static from_buffer(buffer)

Create an CuMemsetD32V2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD32V2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d32v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD32V2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

n

int:

ptr

Get the pointer address to the data as Python int.

ui

int:

class cupti.cupti.CuMemsetD2D8V2PtdsParams

Bases: object

Empty-initialize an instance of cuMemsetD2D8_v2_ptds_params.

See also

cuMemsetD2D8_v2_ptds_params

dst_device

int:

dst_pitch

int:

static from_buffer(buffer)

Create an CuMemsetD2D8V2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD2D8V2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d2d8v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD2D8V2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

ptr

Get the pointer address to the data as Python int.

uc

int:

width

int:

class cupti.cupti.CuMemsetD2D16V2PtdsParams

Bases: object

Empty-initialize an instance of cuMemsetD2D16_v2_ptds_params.

See also

cuMemsetD2D16_v2_ptds_params

dst_device

int:

dst_pitch

int:

static from_buffer(buffer)

Create an CuMemsetD2D16V2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD2D16V2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d2d16v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD2D16V2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

ptr

Get the pointer address to the data as Python int.

us

int:

width

int:

class cupti.cupti.CuMemsetD2D32V2PtdsParams

Bases: object

Empty-initialize an instance of cuMemsetD2D32_v2_ptds_params.

See also

cuMemsetD2D32_v2_ptds_params

dst_device

int:

dst_pitch

int:

static from_buffer(buffer)

Create an CuMemsetD2D32V2PtdsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD2D32V2PtdsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d2d32v2ptds_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD2D32V2PtdsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

ptr

Get the pointer address to the data as Python int.

ui

int:

width

int:

class cupti.cupti.CuMemsetD8AsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemsetD8Async_ptsz_params.

See also

cuMemsetD8Async_ptsz_params

dst_device

int:

static from_buffer(buffer)

Create an CuMemsetD8AsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD8AsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d8_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD8AsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

n

int:

ptr

Get the pointer address to the data as Python int.

uc

int:

class cupti.cupti.CuMemsetD16AsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemsetD16Async_ptsz_params.

See also

cuMemsetD16Async_ptsz_params

dst_device

int:

static from_buffer(buffer)

Create an CuMemsetD16AsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD16AsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d16_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD16AsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

n

int:

ptr

Get the pointer address to the data as Python int.

us

int:

class cupti.cupti.CuMemsetD32AsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemsetD32Async_ptsz_params.

See also

cuMemsetD32Async_ptsz_params

dst_device

int:

static from_buffer(buffer)

Create an CuMemsetD32AsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD32AsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d32_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD32AsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

n

int:

ptr

Get the pointer address to the data as Python int.

ui

int:

class cupti.cupti.CuMemsetD2D8AsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemsetD2D8Async_ptsz_params.

See also

cuMemsetD2D8Async_ptsz_params

dst_device

int:

dst_pitch

int:

static from_buffer(buffer)

Create an CuMemsetD2D8AsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD2D8AsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d2d8_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD2D8AsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

height

int:

ptr

Get the pointer address to the data as Python int.

uc

int:

width

int:

class cupti.cupti.CuMemsetD2D16AsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemsetD2D16Async_ptsz_params.

See also

cuMemsetD2D16Async_ptsz_params

dst_device

int:

dst_pitch

int:

static from_buffer(buffer)

Create an CuMemsetD2D16AsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD2D16AsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d2d16_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD2D16AsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

height

int:

ptr

Get the pointer address to the data as Python int.

us

int:

width

int:

class cupti.cupti.CuMemsetD2D32AsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemsetD2D32Async_ptsz_params.

See also

cuMemsetD2D32Async_ptsz_params

dst_device

int:

dst_pitch

int:

static from_buffer(buffer)

Create an CuMemsetD2D32AsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD2D32AsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d2d32_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD2D32AsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

height

int:

ptr

Get the pointer address to the data as Python int.

ui

int:

width

int:

class cupti.cupti.CuArrayCreateV2Params

Bases: object

Empty-initialize an instance of cuArrayCreate_v2_params.

See also

cuArrayCreate_v2_params

static from_buffer(buffer)

Create an CuArrayCreateV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuArrayCreateV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_array_create_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuArrayCreateV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_allocate_array

int:

p_handle

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuArrayGetDescriptorV2Params

Bases: object

Empty-initialize an instance of cuArrayGetDescriptor_v2_params.

See also

cuArrayGetDescriptor_v2_params

static from_buffer(buffer)

Create an CuArrayGetDescriptorV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuArrayGetDescriptorV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_array_get_descriptor_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuArrayGetDescriptorV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_array

int:

p_array_descriptor

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuArrayGetSparsePropertiesParams

Bases: object

Empty-initialize an instance of cuArrayGetSparseProperties_params.

See also

cuArrayGetSparseProperties_params

array

int:

static from_buffer(buffer)

Create an CuArrayGetSparsePropertiesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuArrayGetSparsePropertiesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_array_get_sparse_properties_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuArrayGetSparsePropertiesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

sparse_properties

int:

class cupti.cupti.CuMipmappedArrayGetSparsePropertiesParams

Bases: object

Empty-initialize an instance of cuMipmappedArrayGetSparseProperties_params.

See also

cuMipmappedArrayGetSparseProperties_params

static from_buffer(buffer)

Create an CuMipmappedArrayGetSparsePropertiesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMipmappedArrayGetSparsePropertiesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mipmapped_array_get_sparse_properties_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMipmappedArrayGetSparsePropertiesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mipmap

int:

ptr

Get the pointer address to the data as Python int.

sparse_properties

int:

class cupti.cupti.CuArrayGetMemoryRequirementsParams

Bases: object

Empty-initialize an instance of cuArrayGetMemoryRequirements_params.

See also

cuArrayGetMemoryRequirements_params

array

int:

device_

int:

static from_buffer(buffer)

Create an CuArrayGetMemoryRequirementsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuArrayGetMemoryRequirementsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_array_get_memory_requirements_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuArrayGetMemoryRequirementsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

memory_requirements

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMipmappedArrayGetMemoryRequirementsParams

Bases: object

Empty-initialize an instance of cuMipmappedArrayGetMemoryRequirements_params.

See also

cuMipmappedArrayGetMemoryRequirements_params

device_

int:

static from_buffer(buffer)

Create an CuMipmappedArrayGetMemoryRequirementsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMipmappedArrayGetMemoryRequirementsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mipmapped_array_get_memory_requirements_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMipmappedArrayGetMemoryRequirementsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

memory_requirements

int:

mipmap

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuArrayGetPlaneParams

Bases: object

Empty-initialize an instance of cuArrayGetPlane_params.

See also

cuArrayGetPlane_params

static from_buffer(buffer)

Create an CuArrayGetPlaneParams instance with the memory from the given buffer.

static from_data(data)

Create an CuArrayGetPlaneParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_array_get_plane_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuArrayGetPlaneParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_array

int:

p_plane_array

int:

plane_idx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuArrayDestroyParams

Bases: object

Empty-initialize an instance of cuArrayDestroy_params.

See also

cuArrayDestroy_params

static from_buffer(buffer)

Create an CuArrayDestroyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuArrayDestroyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_array_destroy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuArrayDestroyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_array

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuArray3DCreateV2Params

Bases: object

Empty-initialize an instance of cuArray3DCreate_v2_params.

See also

cuArray3DCreate_v2_params

static from_buffer(buffer)

Create an CuArray3DCreateV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuArray3DCreateV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_array3d_create_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuArray3DCreateV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_allocate_array

int:

p_handle

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuArray3DGetDescriptorV2Params

Bases: object

Empty-initialize an instance of cuArray3DGetDescriptor_v2_params.

See also

cuArray3DGetDescriptor_v2_params

static from_buffer(buffer)

Create an CuArray3DGetDescriptorV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuArray3DGetDescriptorV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_array3d_get_descriptor_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuArray3DGetDescriptorV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_array

int:

p_array_descriptor

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMipmappedArrayCreateParams

Bases: object

Empty-initialize an instance of cuMipmappedArrayCreate_params.

See also

cuMipmappedArrayCreate_params

static from_buffer(buffer)

Create an CuMipmappedArrayCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMipmappedArrayCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mipmapped_array_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMipmappedArrayCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_mipmap_levels

int:

p_handle

int:

p_mipmapped_array_desc

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMipmappedArrayGetLevelParams

Bases: object

Empty-initialize an instance of cuMipmappedArrayGetLevel_params.

See also

cuMipmappedArrayGetLevel_params

static from_buffer(buffer)

Create an CuMipmappedArrayGetLevelParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMipmappedArrayGetLevelParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mipmapped_array_get_level_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMipmappedArrayGetLevelParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_mipmapped_array

int:

level

int:

p_level_array

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMipmappedArrayDestroyParams

Bases: object

Empty-initialize an instance of cuMipmappedArrayDestroy_params.

See also

cuMipmappedArrayDestroy_params

static from_buffer(buffer)

Create an CuMipmappedArrayDestroyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMipmappedArrayDestroyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mipmapped_array_destroy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMipmappedArrayDestroyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_mipmapped_array

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemGetHandleForAddressRangeParams

Bases: object

Empty-initialize an instance of cuMemGetHandleForAddressRange_params.

See also

cuMemGetHandleForAddressRange_params

dptr

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemGetHandleForAddressRangeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemGetHandleForAddressRangeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_get_handle_for_address_range_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemGetHandleForAddressRangeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

int:

handle_type

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CuMemBatchDecompressAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemBatchDecompressAsync_ptsz_params.

See also

cuMemBatchDecompressAsync_ptsz_params

count

int:

error_index

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemBatchDecompressAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemBatchDecompressAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_batch_decompress_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemBatchDecompressAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

params_array

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CuMemAddressReserveParams

Bases: object

Empty-initialize an instance of cuMemAddressReserve_params.

See also

cuMemAddressReserve_params

addr

int:

alignment

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemAddressReserveParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemAddressReserveParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_address_reserve_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemAddressReserveParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

ptr_

int:

size_

int:

class cupti.cupti.CuMemAddressFreeParams

Bases: object

Empty-initialize an instance of cuMemAddressFree_params.

See also

cuMemAddressFree_params

static from_buffer(buffer)

Create an CuMemAddressFreeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemAddressFreeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_address_free_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemAddressFreeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

ptr_

int:

size_

int:

class cupti.cupti.CuMemCreateParams

Bases: object

Empty-initialize an instance of cuMemCreate_params.

See also

cuMemCreate_params

flags_

int:

static from_buffer(buffer)

Create an CuMemCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

int:

prop

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CuMemReleaseParams

Bases: object

Empty-initialize an instance of cuMemRelease_params.

See also

cuMemRelease_params

static from_buffer(buffer)

Create an CuMemReleaseParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemReleaseParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_release_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemReleaseParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemMapParams

Bases: object

Empty-initialize an instance of cuMemMap_params.

See also

cuMemMap_params

flags_

int:

static from_buffer(buffer)

Create an CuMemMapParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemMapParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_map_params_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an CuMemMapParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

int:

offset

int:

ptr

Get the pointer address to the data as Python int.

ptr_

int:

size_

int:

class cupti.cupti.CuMemMapArrayAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemMapArrayAsync_ptsz_params.

See also

cuMemMapArrayAsync_ptsz_params

count

int:

static from_buffer(buffer)

Create an CuMemMapArrayAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemMapArrayAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_map_array_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemMapArrayAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

map_info_list

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemUnmapParams

Bases: object

Empty-initialize an instance of cuMemUnmap_params.

See also

cuMemUnmap_params

static from_buffer(buffer)

Create an CuMemUnmapParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemUnmapParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_unmap_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemUnmapParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

ptr_

int:

size_

int:

class cupti.cupti.CuMemSetAccessParams

Bases: object

Empty-initialize an instance of cuMemSetAccess_params.

See also

cuMemSetAccess_params

count

int:

desc

int:

static from_buffer(buffer)

Create an CuMemSetAccessParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemSetAccessParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_set_access_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemSetAccessParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

ptr_

int:

size_

int:

class cupti.cupti.CuMemGetAccessParams

Bases: object

Empty-initialize an instance of cuMemGetAccess_params.

See also

cuMemGetAccess_params

flags_

int:

static from_buffer(buffer)

Create an CuMemGetAccessParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemGetAccessParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_get_access_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemGetAccessParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

location

int:

ptr

Get the pointer address to the data as Python int.

ptr_

int:

class cupti.cupti.CuMemExportToShareableHandleParams

Bases: object

Empty-initialize an instance of cuMemExportToShareableHandle_params.

See also

cuMemExportToShareableHandle_params

flags_

int:

static from_buffer(buffer)

Create an CuMemExportToShareableHandleParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemExportToShareableHandleParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_export_to_shareable_handle_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemExportToShareableHandleParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

int:

handle_type

int:

ptr

Get the pointer address to the data as Python int.

shareable_handle

int:

class cupti.cupti.CuMemImportFromShareableHandleParams

Bases: object

Empty-initialize an instance of cuMemImportFromShareableHandle_params.

See also

cuMemImportFromShareableHandle_params

static from_buffer(buffer)

Create an CuMemImportFromShareableHandleParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemImportFromShareableHandleParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_import_from_shareable_handle_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemImportFromShareableHandleParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

int:

os_handle

int:

ptr

Get the pointer address to the data as Python int.

sh_handle_type

int:

class cupti.cupti.CuMemGetAllocationGranularityParams

Bases: object

Empty-initialize an instance of cuMemGetAllocationGranularity_params.

See also

cuMemGetAllocationGranularity_params

static from_buffer(buffer)

Create an CuMemGetAllocationGranularityParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemGetAllocationGranularityParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_get_allocation_granularity_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemGetAllocationGranularityParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

granularity

int:

option

int:

prop

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemGetAllocationPropertiesFromHandleParams

Bases: object

Empty-initialize an instance of cuMemGetAllocationPropertiesFromHandle_params.

See also

cuMemGetAllocationPropertiesFromHandle_params

static from_buffer(buffer)

Create an CuMemGetAllocationPropertiesFromHandleParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemGetAllocationPropertiesFromHandleParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_get_allocation_properties_from_handle_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemGetAllocationPropertiesFromHandleParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

int:

prop

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemRetainAllocationHandleParams

Bases: object

Empty-initialize an instance of cuMemRetainAllocationHandle_params.

See also

cuMemRetainAllocationHandle_params

addr

int:

static from_buffer(buffer)

Create an CuMemRetainAllocationHandleParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemRetainAllocationHandleParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_retain_allocation_handle_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemRetainAllocationHandleParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemFreeAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemFreeAsync_ptsz_params.

See also

cuMemFreeAsync_ptsz_params

dptr

int:

static from_buffer(buffer)

Create an CuMemFreeAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemFreeAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_free_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemFreeAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemAllocAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemAllocAsync_ptsz_params.

See also

cuMemAllocAsync_ptsz_params

bytesize

int:

dptr

int:

static from_buffer(buffer)

Create an CuMemAllocAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemAllocAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_alloc_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemAllocAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemPoolTrimToParams

Bases: object

Empty-initialize an instance of cuMemPoolTrimTo_params.

See also

cuMemPoolTrimTo_params

static from_buffer(buffer)

Create an CuMemPoolTrimToParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPoolTrimToParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_pool_trim_to_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPoolTrimToParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

min_bytes_to_keep

int:

pool

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemPoolSetAttributeParams

Bases: object

Empty-initialize an instance of cuMemPoolSetAttribute_params.

See also

cuMemPoolSetAttribute_params

attr

int:

static from_buffer(buffer)

Create an CuMemPoolSetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPoolSetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_pool_set_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPoolSetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pool

int:

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CuMemPoolGetAttributeParams

Bases: object

Empty-initialize an instance of cuMemPoolGetAttribute_params.

See also

cuMemPoolGetAttribute_params

attr

int:

static from_buffer(buffer)

Create an CuMemPoolGetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPoolGetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_pool_get_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPoolGetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pool

int:

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CuMemPoolSetAccessParams

Bases: object

Empty-initialize an instance of cuMemPoolSetAccess_params.

See also

cuMemPoolSetAccess_params

count

int:

static from_buffer(buffer)

Create an CuMemPoolSetAccessParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPoolSetAccessParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_pool_set_access_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPoolSetAccessParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

map

int:

pool

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemPoolGetAccessParams

Bases: object

Empty-initialize an instance of cuMemPoolGetAccess_params.

See also

cuMemPoolGetAccess_params

flags_

int:

static from_buffer(buffer)

Create an CuMemPoolGetAccessParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPoolGetAccessParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_pool_get_access_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPoolGetAccessParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

location

int:

mem_pool

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemPoolCreateParams

Bases: object

Empty-initialize an instance of cuMemPoolCreate_params.

See also

cuMemPoolCreate_params

static from_buffer(buffer)

Create an CuMemPoolCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPoolCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_pool_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPoolCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pool

int:

pool_props

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemPoolDestroyParams

Bases: object

Empty-initialize an instance of cuMemPoolDestroy_params.

See also

cuMemPoolDestroy_params

static from_buffer(buffer)

Create an CuMemPoolDestroyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPoolDestroyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_pool_destroy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPoolDestroyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pool

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemAllocFromPoolAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemAllocFromPoolAsync_ptsz_params.

See also

cuMemAllocFromPoolAsync_ptsz_params

bytesize

int:

dptr

int:

static from_buffer(buffer)

Create an CuMemAllocFromPoolAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemAllocFromPoolAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_alloc_from_pool_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemAllocFromPoolAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

pool

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemPoolExportToShareableHandleParams

Bases: object

Empty-initialize an instance of cuMemPoolExportToShareableHandle_params.

See also

cuMemPoolExportToShareableHandle_params

flags_

int:

static from_buffer(buffer)

Create an CuMemPoolExportToShareableHandleParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPoolExportToShareableHandleParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_pool_export_to_shareable_handle_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPoolExportToShareableHandleParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle_out

int:

handle_type

int:

pool

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemPoolImportFromShareableHandleParams

Bases: object

Empty-initialize an instance of cuMemPoolImportFromShareableHandle_params.

See also

cuMemPoolImportFromShareableHandle_params

flags_

int:

static from_buffer(buffer)

Create an CuMemPoolImportFromShareableHandleParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPoolImportFromShareableHandleParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_pool_import_from_shareable_handle_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPoolImportFromShareableHandleParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

int:

handle_type

int:

pool_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemPoolExportPointerParams

Bases: object

Empty-initialize an instance of cuMemPoolExportPointer_params.

See also

cuMemPoolExportPointer_params

static from_buffer(buffer)

Create an CuMemPoolExportPointerParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPoolExportPointerParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_pool_export_pointer_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPoolExportPointerParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

ptr_

int:

share_data_out

int:

class cupti.cupti.CuMemPoolImportPointerParams

Bases: object

Empty-initialize an instance of cuMemPoolImportPointer_params.

See also

cuMemPoolImportPointer_params

static from_buffer(buffer)

Create an CuMemPoolImportPointerParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPoolImportPointerParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_pool_import_pointer_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPoolImportPointerParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pool

int:

ptr

Get the pointer address to the data as Python int.

ptr_out

int:

share_data

int:

class cupti.cupti.CuMulticastCreateParams

Bases: object

Empty-initialize an instance of cuMulticastCreate_params.

See also

cuMulticastCreate_params

static from_buffer(buffer)

Create an CuMulticastCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMulticastCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_multicast_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMulticastCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mc_handle

int:

prop

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMulticastAddDeviceParams

Bases: object

Empty-initialize an instance of cuMulticastAddDevice_params.

See also

cuMulticastAddDevice_params

dev

int:

static from_buffer(buffer)

Create an CuMulticastAddDeviceParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMulticastAddDeviceParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_multicast_add_device_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMulticastAddDeviceParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mc_handle

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMulticastBindMemParams

Bases: object

Empty-initialize an instance of cuMulticastBindMem_params.

See also

cuMulticastBindMem_params

flags_

int:

static from_buffer(buffer)

Create an CuMulticastBindMemParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMulticastBindMemParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_multicast_bind_mem_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMulticastBindMemParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mc_handle

int:

mc_offset

int:

mem_handle

int:

mem_offset

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CuMulticastBindAddrParams

Bases: object

Empty-initialize an instance of cuMulticastBindAddr_params.

See also

cuMulticastBindAddr_params

flags_

int:

static from_buffer(buffer)

Create an CuMulticastBindAddrParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMulticastBindAddrParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_multicast_bind_addr_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMulticastBindAddrParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mc_handle

int:

mc_offset

int:

memptr

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CuMulticastUnbindParams

Bases: object

Empty-initialize an instance of cuMulticastUnbind_params.

See also

cuMulticastUnbind_params

dev

int:

static from_buffer(buffer)

Create an CuMulticastUnbindParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMulticastUnbindParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_multicast_unbind_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMulticastUnbindParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mc_handle

int:

mc_offset

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CuMulticastGetGranularityParams

Bases: object

Empty-initialize an instance of cuMulticastGetGranularity_params.

See also

cuMulticastGetGranularity_params

static from_buffer(buffer)

Create an CuMulticastGetGranularityParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMulticastGetGranularityParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_multicast_get_granularity_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMulticastGetGranularityParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

granularity

int:

option

int:

prop

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuPointerGetAttributeParams

Bases: object

Empty-initialize an instance of cuPointerGetAttribute_params.

See also

cuPointerGetAttribute_params

attribute

int:

data_

int:

static from_buffer(buffer)

Create an CuPointerGetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuPointerGetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_pointer_get_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuPointerGetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

ptr_

int:

class cupti.cupti.CuMemPrefetchAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemPrefetchAsync_ptsz_params.

See also

cuMemPrefetchAsync_ptsz_params

count

int:

dev_ptr

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemPrefetchAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPrefetchAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_prefetch_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPrefetchAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemAdviseParams

Bases: object

Empty-initialize an instance of cuMemAdvise_params.

See also

cuMemAdvise_params

advice

int:

count

int:

dev_ptr

int:

device_

int:

static from_buffer(buffer)

Create an CuMemAdviseParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemAdviseParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_advise_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemAdviseParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemRangeGetAttributeParams

Bases: object

Empty-initialize an instance of cuMemRangeGetAttribute_params.

See also

cuMemRangeGetAttribute_params

attribute

int:

count

int:

data_

int:

data_size

int:

dev_ptr

int:

static from_buffer(buffer)

Create an CuMemRangeGetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemRangeGetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_range_get_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemRangeGetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemRangeGetAttributesParams

Bases: object

Empty-initialize an instance of cuMemRangeGetAttributes_params.

See also

cuMemRangeGetAttributes_params

attributes

int:

count

int:

data_

int:

data_sizes

int:

dev_ptr

int:

static from_buffer(buffer)

Create an CuMemRangeGetAttributesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemRangeGetAttributesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_range_get_attributes_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemRangeGetAttributesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_attributes

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuPointerSetAttributeParams

Bases: object

Empty-initialize an instance of cuPointerSetAttribute_params.

See also

cuPointerSetAttribute_params

attribute

int:

static from_buffer(buffer)

Create an CuPointerSetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuPointerSetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_pointer_set_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuPointerSetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

ptr_

int:

value

int:

class cupti.cupti.CuPointerGetAttributesParams

Bases: object

Empty-initialize an instance of cuPointerGetAttributes_params.

See also

cuPointerGetAttributes_params

attributes

int:

data_

int:

static from_buffer(buffer)

Create an CuPointerGetAttributesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuPointerGetAttributesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_pointer_get_attributes_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuPointerGetAttributesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_attributes

int:

ptr

Get the pointer address to the data as Python int.

ptr_

int:

class cupti.cupti.CuStreamCreateParams

Bases: object

Empty-initialize an instance of cuStreamCreate_params.

See also

cuStreamCreate_params

flags_

int:

static from_buffer(buffer)

Create an CuStreamCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ph_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamCreateWithPriorityParams

Bases: object

Empty-initialize an instance of cuStreamCreateWithPriority_params.

See also

cuStreamCreateWithPriority_params

flags_

int:

static from_buffer(buffer)

Create an CuStreamCreateWithPriorityParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamCreateWithPriorityParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_create_with_priority_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamCreateWithPriorityParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ph_stream

int:

priority

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetPriorityPtszParams

Bases: object

Empty-initialize an instance of cuStreamGetPriority_ptsz_params.

See also

cuStreamGetPriority_ptsz_params

static from_buffer(buffer)

Create an CuStreamGetPriorityPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetPriorityPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_priority_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetPriorityPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

priority

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetDevicePtszParams

Bases: object

Empty-initialize an instance of cuStreamGetDevice_ptsz_params.

See also

cuStreamGetDevice_ptsz_params

device_

int:

static from_buffer(buffer)

Create an CuStreamGetDevicePtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetDevicePtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_device_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetDevicePtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetFlagsPtszParams

Bases: object

Empty-initialize an instance of cuStreamGetFlags_ptsz_params.

See also

cuStreamGetFlags_ptsz_params

flags_

int:

static from_buffer(buffer)

Create an CuStreamGetFlagsPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetFlagsPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_flags_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetFlagsPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetIdPtszParams

Bases: object

Empty-initialize an instance of cuStreamGetId_ptsz_params.

See also

cuStreamGetId_ptsz_params

static from_buffer(buffer)

Create an CuStreamGetIdPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetIdPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_id_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetIdPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

stream_id

int:

class cupti.cupti.CuStreamGetCtxPtszParams

Bases: object

Empty-initialize an instance of cuStreamGetCtx_ptsz_params.

See also

cuStreamGetCtx_ptsz_params

static from_buffer(buffer)

Create an CuStreamGetCtxPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetCtxPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_ctx_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetCtxPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

pctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetCtxV2PtszParams

Bases: object

Empty-initialize an instance of cuStreamGetCtx_v2_ptsz_params.

See also

cuStreamGetCtx_v2_ptsz_params

static from_buffer(buffer)

Create an CuStreamGetCtxV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetCtxV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_ctx_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetCtxV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

p_ctx

int:

p_green_ctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamWaitEventPtszParams

Bases: object

Empty-initialize an instance of cuStreamWaitEvent_ptsz_params.

See also

cuStreamWaitEvent_ptsz_params

flags_

int:

static from_buffer(buffer)

Create an CuStreamWaitEventPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWaitEventPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_wait_event_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWaitEventPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_event

int:

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamAddCallbackPtszParams

Bases: object

Empty-initialize an instance of cuStreamAddCallback_ptsz_params.

See also

cuStreamAddCallback_ptsz_params

callback

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamAddCallbackPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamAddCallbackPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_add_callback_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamAddCallbackPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

user_data

int:

class cupti.cupti.CuStreamBeginCaptureV2PtszParams

Bases: object

Empty-initialize an instance of cuStreamBeginCapture_v2_ptsz_params.

See also

cuStreamBeginCapture_v2_ptsz_params

static from_buffer(buffer)

Create an CuStreamBeginCaptureV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamBeginCaptureV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_begin_capture_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamBeginCaptureV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

mode

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamBeginCaptureToGraphPtszParams

Bases: object

Empty-initialize an instance of cuStreamBeginCaptureToGraph_ptsz_params.

See also

cuStreamBeginCaptureToGraph_ptsz_params

dependencies

int:

dependency_data

int:

static from_buffer(buffer)

Create an CuStreamBeginCaptureToGraphPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamBeginCaptureToGraphPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_begin_capture_to_graph_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamBeginCaptureToGraphPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

h_stream

int:

mode

int:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuThreadExchangeStreamCaptureModeParams

Bases: object

Empty-initialize an instance of cuThreadExchangeStreamCaptureMode_params.

See also

cuThreadExchangeStreamCaptureMode_params

static from_buffer(buffer)

Create an CuThreadExchangeStreamCaptureModeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuThreadExchangeStreamCaptureModeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_thread_exchange_stream_capture_mode_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuThreadExchangeStreamCaptureModeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mode

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamEndCapturePtszParams

Bases: object

Empty-initialize an instance of cuStreamEndCapture_ptsz_params.

See also

cuStreamEndCapture_ptsz_params

static from_buffer(buffer)

Create an CuStreamEndCapturePtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamEndCapturePtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_end_capture_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamEndCapturePtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ph_graph

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamIsCapturingPtszParams

Bases: object

Empty-initialize an instance of cuStreamIsCapturing_ptsz_params.

See also

cuStreamIsCapturing_ptsz_params

capture_status

int:

static from_buffer(buffer)

Create an CuStreamIsCapturingPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamIsCapturingPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_is_capturing_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamIsCapturingPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetCaptureInfoV2PtszParams

Bases: object

Empty-initialize an instance of cuStreamGetCaptureInfo_v2_ptsz_params.

See also

cuStreamGetCaptureInfo_v2_ptsz_params

capture_status_out

int:

dependencies_out

int:

static from_buffer(buffer)

Create an CuStreamGetCaptureInfoV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetCaptureInfoV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_capture_info_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetCaptureInfoV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_out

int:

h_stream

int:

id_out

int:

num_dependencies_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetCaptureInfoV3PtszParams

Bases: object

Empty-initialize an instance of cuStreamGetCaptureInfo_v3_ptsz_params.

See also

cuStreamGetCaptureInfo_v3_ptsz_params

capture_status_out

int:

dependencies_out

int:

edge_data_out

int:

static from_buffer(buffer)

Create an CuStreamGetCaptureInfoV3PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetCaptureInfoV3PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_capture_info_v3ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetCaptureInfoV3PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_out

int:

h_stream

int:

id_out

int:

num_dependencies_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamUpdateCaptureDependenciesPtszParams

Bases: object

Empty-initialize an instance of cuStreamUpdateCaptureDependencies_ptsz_params.

See also

cuStreamUpdateCaptureDependencies_ptsz_params

dependencies

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamUpdateCaptureDependenciesPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamUpdateCaptureDependenciesPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_update_capture_dependencies_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamUpdateCaptureDependenciesPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamUpdateCaptureDependenciesV2PtszParams

Bases: object

Empty-initialize an instance of cuStreamUpdateCaptureDependencies_v2_ptsz_params.

See also

cuStreamUpdateCaptureDependencies_v2_ptsz_params

dependencies

int:

dependency_data

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamUpdateCaptureDependenciesV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamUpdateCaptureDependenciesV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_update_capture_dependencies_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamUpdateCaptureDependenciesV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamAttachMemAsyncPtszParams

Bases: object

Empty-initialize an instance of cuStreamAttachMemAsync_ptsz_params.

See also

cuStreamAttachMemAsync_ptsz_params

dptr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamAttachMemAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamAttachMemAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_attach_mem_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamAttachMemAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

length

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamQueryPtszParams

Bases: object

Empty-initialize an instance of cuStreamQuery_ptsz_params.

See also

cuStreamQuery_ptsz_params

static from_buffer(buffer)

Create an CuStreamQueryPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamQueryPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_query_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamQueryPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamSynchronizePtszParams

Bases: object

Empty-initialize an instance of cuStreamSynchronize_ptsz_params.

See also

cuStreamSynchronize_ptsz_params

static from_buffer(buffer)

Create an CuStreamSynchronizePtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamSynchronizePtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_synchronize_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamSynchronizePtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamDestroyV2Params

Bases: object

Empty-initialize an instance of cuStreamDestroy_v2_params.

See also

cuStreamDestroy_v2_params

static from_buffer(buffer)

Create an CuStreamDestroyV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamDestroyV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_destroy_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamDestroyV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamCopyAttributesPtszParams

Bases: object

Empty-initialize an instance of cuStreamCopyAttributes_ptsz_params.

See also

cuStreamCopyAttributes_ptsz_params

dst

int:

static from_buffer(buffer)

Create an CuStreamCopyAttributesPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamCopyAttributesPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_copy_attributes_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamCopyAttributesPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src

int:

class cupti.cupti.CuStreamGetAttributePtszParams

Bases: object

Empty-initialize an instance of cuStreamGetAttribute_ptsz_params.

See also

cuStreamGetAttribute_ptsz_params

attr

int:

static from_buffer(buffer)

Create an CuStreamGetAttributePtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetAttributePtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_attribute_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetAttributePtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

value_out

int:

class cupti.cupti.CuStreamSetAttributePtszParams

Bases: object

Empty-initialize an instance of cuStreamSetAttribute_ptsz_params.

See also

cuStreamSetAttribute_ptsz_params

attr

int:

static from_buffer(buffer)

Create an CuStreamSetAttributePtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamSetAttributePtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_set_attribute_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamSetAttributePtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CuEventCreateParams

Bases: object

Empty-initialize an instance of cuEventCreate_params.

See also

cuEventCreate_params

flags_

int:

static from_buffer(buffer)

Create an CuEventCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuEventCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_event_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuEventCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ph_event

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuEventRecordPtszParams

Bases: object

Empty-initialize an instance of cuEventRecord_ptsz_params.

See also

cuEventRecord_ptsz_params

static from_buffer(buffer)

Create an CuEventRecordPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuEventRecordPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_event_record_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuEventRecordPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_event

int:

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuEventRecordWithFlagsPtszParams

Bases: object

Empty-initialize an instance of cuEventRecordWithFlags_ptsz_params.

See also

cuEventRecordWithFlags_ptsz_params

flags_

int:

static from_buffer(buffer)

Create an CuEventRecordWithFlagsPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuEventRecordWithFlagsPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_event_record_with_flags_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuEventRecordWithFlagsPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_event

int:

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuEventQueryParams

Bases: object

Empty-initialize an instance of cuEventQuery_params.

See also

cuEventQuery_params

static from_buffer(buffer)

Create an CuEventQueryParams instance with the memory from the given buffer.

static from_data(data)

Create an CuEventQueryParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_event_query_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuEventQueryParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_event

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuEventSynchronizeParams

Bases: object

Empty-initialize an instance of cuEventSynchronize_params.

See also

cuEventSynchronize_params

static from_buffer(buffer)

Create an CuEventSynchronizeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuEventSynchronizeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_event_synchronize_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuEventSynchronizeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_event

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuEventDestroyV2Params

Bases: object

Empty-initialize an instance of cuEventDestroy_v2_params.

See also

cuEventDestroy_v2_params

static from_buffer(buffer)

Create an CuEventDestroyV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuEventDestroyV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_event_destroy_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuEventDestroyV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_event

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuEventElapsedTimeParams

Bases: object

Empty-initialize an instance of cuEventElapsedTime_params.

See also

cuEventElapsedTime_params

static from_buffer(buffer)

Create an CuEventElapsedTimeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuEventElapsedTimeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_event_elapsed_time_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuEventElapsedTimeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_end

int:

h_start

int:

p_milliseconds

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuEventElapsedTimeV2Params

Bases: object

Empty-initialize an instance of cuEventElapsedTime_v2_params.

See also

cuEventElapsedTime_v2_params

static from_buffer(buffer)

Create an CuEventElapsedTimeV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuEventElapsedTimeV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_event_elapsed_time_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuEventElapsedTimeV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_end

int:

h_start

int:

p_milliseconds

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuImportExternalMemoryParams

Bases: object

Empty-initialize an instance of cuImportExternalMemory_params.

See also

cuImportExternalMemory_params

ext_mem_out

int:

static from_buffer(buffer)

Create an CuImportExternalMemoryParams instance with the memory from the given buffer.

static from_data(data)

Create an CuImportExternalMemoryParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_import_external_memory_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuImportExternalMemoryParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mem_handle_desc

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuExternalMemoryGetMappedBufferParams

Bases: object

Empty-initialize an instance of cuExternalMemoryGetMappedBuffer_params.

See also

cuExternalMemoryGetMappedBuffer_params

buffer_desc

int:

dev_ptr

int:

ext_mem

int:

static from_buffer(buffer)

Create an CuExternalMemoryGetMappedBufferParams instance with the memory from the given buffer.

static from_data(data)

Create an CuExternalMemoryGetMappedBufferParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_external_memory_get_mapped_buffer_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuExternalMemoryGetMappedBufferParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuExternalMemoryGetMappedMipmappedArrayParams

Bases: object

Empty-initialize an instance of cuExternalMemoryGetMappedMipmappedArray_params.

See also

cuExternalMemoryGetMappedMipmappedArray_params

ext_mem

int:

static from_buffer(buffer)

Create an CuExternalMemoryGetMappedMipmappedArrayParams instance with the memory from the given buffer.

static from_data(data)

Create an CuExternalMemoryGetMappedMipmappedArrayParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_external_memory_get_mapped_mipmapped_array_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuExternalMemoryGetMappedMipmappedArrayParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mipmap

int:

mipmap_desc

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDestroyExternalMemoryParams

Bases: object

Empty-initialize an instance of cuDestroyExternalMemory_params.

See also

cuDestroyExternalMemory_params

ext_mem

int:

static from_buffer(buffer)

Create an CuDestroyExternalMemoryParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDestroyExternalMemoryParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_destroy_external_memory_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDestroyExternalMemoryParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuImportExternalSemaphoreParams

Bases: object

Empty-initialize an instance of cuImportExternalSemaphore_params.

See also

cuImportExternalSemaphore_params

ext_sem_out

int:

static from_buffer(buffer)

Create an CuImportExternalSemaphoreParams instance with the memory from the given buffer.

static from_data(data)

Create an CuImportExternalSemaphoreParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_import_external_semaphore_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuImportExternalSemaphoreParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

sem_handle_desc

int:

class cupti.cupti.CuSignalExternalSemaphoresAsyncPtszParams

Bases: object

Empty-initialize an instance of cuSignalExternalSemaphoresAsync_ptsz_params.

See also

cuSignalExternalSemaphoresAsync_ptsz_params

ext_sem_array

int:

static from_buffer(buffer)

Create an CuSignalExternalSemaphoresAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuSignalExternalSemaphoresAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_signal_external_semaphores_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuSignalExternalSemaphoresAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_ext_sems

int:

params_array

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CuWaitExternalSemaphoresAsyncPtszParams

Bases: object

Empty-initialize an instance of cuWaitExternalSemaphoresAsync_ptsz_params.

See also

cuWaitExternalSemaphoresAsync_ptsz_params

ext_sem_array

int:

static from_buffer(buffer)

Create an CuWaitExternalSemaphoresAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuWaitExternalSemaphoresAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_wait_external_semaphores_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuWaitExternalSemaphoresAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_ext_sems

int:

params_array

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CuDestroyExternalSemaphoreParams

Bases: object

Empty-initialize an instance of cuDestroyExternalSemaphore_params.

See also

cuDestroyExternalSemaphore_params

ext_sem

int:

static from_buffer(buffer)

Create an CuDestroyExternalSemaphoreParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDestroyExternalSemaphoreParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_destroy_external_semaphore_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDestroyExternalSemaphoreParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamWaitValue32V2PtszParams

Bases: object

Empty-initialize an instance of cuStreamWaitValue32_v2_ptsz_params.

See also

cuStreamWaitValue32_v2_ptsz_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWaitValue32V2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWaitValue32V2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_wait_value32v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWaitValue32V2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamWaitValue64V2PtszParams

Bases: object

Empty-initialize an instance of cuStreamWaitValue64_v2_ptsz_params.

See also

cuStreamWaitValue64_v2_ptsz_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWaitValue64V2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWaitValue64V2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_wait_value64v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWaitValue64V2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamWriteValue32V2PtszParams

Bases: object

Empty-initialize an instance of cuStreamWriteValue32_v2_ptsz_params.

See also

cuStreamWriteValue32_v2_ptsz_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWriteValue32V2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWriteValue32V2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_write_value32v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWriteValue32V2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamWriteValue64V2PtszParams

Bases: object

Empty-initialize an instance of cuStreamWriteValue64_v2_ptsz_params.

See also

cuStreamWriteValue64_v2_ptsz_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWriteValue64V2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWriteValue64V2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_write_value64v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWriteValue64V2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamBatchMemOpV2PtszParams

Bases: object

Empty-initialize an instance of cuStreamBatchMemOp_v2_ptsz_params.

See also

cuStreamBatchMemOp_v2_ptsz_params

count

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamBatchMemOpV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamBatchMemOpV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_batch_mem_op_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamBatchMemOpV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

param_array

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CuFuncGetAttributeParams

Bases: object

Empty-initialize an instance of cuFuncGetAttribute_params.

See also

cuFuncGetAttribute_params

attrib

int:

static from_buffer(buffer)

Create an CuFuncGetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuFuncGetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_func_get_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuFuncGetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hfunc

int:

pi

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuFuncSetAttributeParams

Bases: object

Empty-initialize an instance of cuFuncSetAttribute_params.

See also

cuFuncSetAttribute_params

attrib

int:

static from_buffer(buffer)

Create an CuFuncSetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuFuncSetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_func_set_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuFuncSetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hfunc

int:

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CuFuncSetCacheConfigParams

Bases: object

Empty-initialize an instance of cuFuncSetCacheConfig_params.

See also

cuFuncSetCacheConfig_params

config

int:

static from_buffer(buffer)

Create an CuFuncSetCacheConfigParams instance with the memory from the given buffer.

static from_data(data)

Create an CuFuncSetCacheConfigParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_func_set_cache_config_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuFuncSetCacheConfigParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hfunc

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuFuncGetModuleParams

Bases: object

Empty-initialize an instance of cuFuncGetModule_params.

See also

cuFuncGetModule_params

static from_buffer(buffer)

Create an CuFuncGetModuleParams instance with the memory from the given buffer.

static from_data(data)

Create an CuFuncGetModuleParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_func_get_module_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuFuncGetModuleParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hfunc

int:

hmod

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuFuncGetNameParams

Bases: object

Empty-initialize an instance of cuFuncGetName_params.

See also

cuFuncGetName_params

static from_buffer(buffer)

Create an CuFuncGetNameParams instance with the memory from the given buffer.

static from_data(data)

Create an CuFuncGetNameParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_func_get_name_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuFuncGetNameParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hfunc

int:

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuFuncGetParamInfoParams

Bases: object

Empty-initialize an instance of cuFuncGetParamInfo_params.

See also

cuFuncGetParamInfo_params

static from_buffer(buffer)

Create an CuFuncGetParamInfoParams instance with the memory from the given buffer.

static from_data(data)

Create an CuFuncGetParamInfoParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_func_get_param_info_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuFuncGetParamInfoParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

param_index

int:

param_offset

int:

param_size

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuFuncIsLoadedParams

Bases: object

Empty-initialize an instance of cuFuncIsLoaded_params.

See also

cuFuncIsLoaded_params

static from_buffer(buffer)

Create an CuFuncIsLoadedParams instance with the memory from the given buffer.

static from_data(data)

Create an CuFuncIsLoadedParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_func_is_loaded_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuFuncIsLoadedParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

function

int:

ptr

Get the pointer address to the data as Python int.

state

int:

class cupti.cupti.CuFuncLoadParams

Bases: object

Empty-initialize an instance of cuFuncLoad_params.

See also

cuFuncLoad_params

static from_buffer(buffer)

Create an CuFuncLoadParams instance with the memory from the given buffer.

static from_data(data)

Create an CuFuncLoadParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_func_load_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuFuncLoadParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

function

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLaunchKernelPtszParams

Bases: object

Empty-initialize an instance of cuLaunchKernel_ptsz_params.

See also

cuLaunchKernel_ptsz_params

block_dim_x

int:

block_dim_y

int:

block_dim_z

int:

extra

int:

f

int:

static from_buffer(buffer)

Create an CuLaunchKernelPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLaunchKernelPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_launch_kernel_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLaunchKernelPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

grid_dim_x

int:

grid_dim_y

int:

grid_dim_z

int:

h_stream

int:

kernel_params

int:

ptr

Get the pointer address to the data as Python int.

shared_mem_bytes

int:

class cupti.cupti.CuLaunchKernelExPtszParams

Bases: object

Empty-initialize an instance of cuLaunchKernelEx_ptsz_params.

See also

cuLaunchKernelEx_ptsz_params

config

int:

extra

int:

f

int:

static from_buffer(buffer)

Create an CuLaunchKernelExPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLaunchKernelExPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_launch_kernel_ex_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLaunchKernelExPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kernel_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLaunchCooperativeKernelPtszParams

Bases: object

Empty-initialize an instance of cuLaunchCooperativeKernel_ptsz_params.

See also

cuLaunchCooperativeKernel_ptsz_params

block_dim_x

int:

block_dim_y

int:

block_dim_z

int:

f

int:

static from_buffer(buffer)

Create an CuLaunchCooperativeKernelPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLaunchCooperativeKernelPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_launch_cooperative_kernel_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLaunchCooperativeKernelPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

grid_dim_x

int:

grid_dim_y

int:

grid_dim_z

int:

h_stream

int:

kernel_params

int:

ptr

Get the pointer address to the data as Python int.

shared_mem_bytes

int:

class cupti.cupti.CuLaunchCooperativeKernelMultiDeviceParams

Bases: object

Empty-initialize an instance of cuLaunchCooperativeKernelMultiDevice_params.

See also

cuLaunchCooperativeKernelMultiDevice_params

flags_

int:

static from_buffer(buffer)

Create an CuLaunchCooperativeKernelMultiDeviceParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLaunchCooperativeKernelMultiDeviceParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_launch_cooperative_kernel_multi_device_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLaunchCooperativeKernelMultiDeviceParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

launch_params_list

int:

num_devices

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLaunchHostFuncPtszParams

Bases: object

Empty-initialize an instance of cuLaunchHostFunc_ptsz_params.

See also

cuLaunchHostFunc_ptsz_params

fn

int:

static from_buffer(buffer)

Create an CuLaunchHostFuncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLaunchHostFuncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_launch_host_func_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLaunchHostFuncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

user_data

int:

class cupti.cupti.CuFuncSetBlockShapeParams

Bases: object

Empty-initialize an instance of cuFuncSetBlockShape_params.

See also

cuFuncSetBlockShape_params

static from_buffer(buffer)

Create an CuFuncSetBlockShapeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuFuncSetBlockShapeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_func_set_block_shape_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuFuncSetBlockShapeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hfunc

int:

ptr

Get the pointer address to the data as Python int.

x

int:

y

int:

z

int:

class cupti.cupti.CuFuncSetSharedSizeParams

Bases: object

Empty-initialize an instance of cuFuncSetSharedSize_params.

See also

cuFuncSetSharedSize_params

bytes

int:

static from_buffer(buffer)

Create an CuFuncSetSharedSizeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuFuncSetSharedSizeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_func_set_shared_size_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuFuncSetSharedSizeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hfunc

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuParamSetSizeParams

Bases: object

Empty-initialize an instance of cuParamSetSize_params.

See also

cuParamSetSize_params

static from_buffer(buffer)

Create an CuParamSetSizeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuParamSetSizeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_param_set_size_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuParamSetSizeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hfunc

int:

numbytes

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuParamSetiParams

Bases: object

Empty-initialize an instance of cuParamSeti_params.

See also

cuParamSeti_params

static from_buffer(buffer)

Create an CuParamSetiParams instance with the memory from the given buffer.

static from_data(data)

Create an CuParamSetiParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_param_seti_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuParamSetiParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hfunc

int:

offset

int:

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CuParamSetfParams

Bases: object

Empty-initialize an instance of cuParamSetf_params.

See also

cuParamSetf_params

static from_buffer(buffer)

Create an CuParamSetfParams instance with the memory from the given buffer.

static from_data(data)

Create an CuParamSetfParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_param_setf_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuParamSetfParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hfunc

int:

offset

int:

ptr

Get the pointer address to the data as Python int.

value

float:

class cupti.cupti.CuParamSetvParams

Bases: object

Empty-initialize an instance of cuParamSetv_params.

See also

cuParamSetv_params

static from_buffer(buffer)

Create an CuParamSetvParams instance with the memory from the given buffer.

static from_data(data)

Create an CuParamSetvParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_param_setv_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuParamSetvParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hfunc

int:

numbytes

int:

offset

int:

ptr

Get the pointer address to the data as Python int.

ptr_

int:

class cupti.cupti.CuLaunchParams

Bases: object

Empty-initialize an instance of cuLaunch_params.

See also

cuLaunch_params

f

int:

static from_buffer(buffer)

Create an CuLaunchParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLaunchParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_launch_params_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an CuLaunchParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLaunchGridParams

Bases: object

Empty-initialize an instance of cuLaunchGrid_params.

See also

cuLaunchGrid_params

f

int:

static from_buffer(buffer)

Create an CuLaunchGridParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLaunchGridParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_launch_grid_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLaunchGridParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

grid_height

int:

grid_width

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLaunchGridAsyncParams

Bases: object

Empty-initialize an instance of cuLaunchGridAsync_params.

See also

cuLaunchGridAsync_params

f

int:

static from_buffer(buffer)

Create an CuLaunchGridAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLaunchGridAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_launch_grid_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLaunchGridAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

grid_height

int:

grid_width

int:

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuParamSetTexRefParams

Bases: object

Empty-initialize an instance of cuParamSetTexRef_params.

See also

cuParamSetTexRef_params

static from_buffer(buffer)

Create an CuParamSetTexRefParams instance with the memory from the given buffer.

static from_data(data)

Create an CuParamSetTexRefParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_param_set_tex_ref_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuParamSetTexRefParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

hfunc

int:

ptr

Get the pointer address to the data as Python int.

texunit

int:

class cupti.cupti.CuFuncSetSharedMemConfigParams

Bases: object

Empty-initialize an instance of cuFuncSetSharedMemConfig_params.

See also

cuFuncSetSharedMemConfig_params

config

int:

static from_buffer(buffer)

Create an CuFuncSetSharedMemConfigParams instance with the memory from the given buffer.

static from_data(data)

Create an CuFuncSetSharedMemConfigParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_func_set_shared_mem_config_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuFuncSetSharedMemConfigParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hfunc

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphCreateParams

Bases: object

Empty-initialize an instance of cuGraphCreate_params.

See also

cuGraphCreate_params

flags_

int:

static from_buffer(buffer)

Create an CuGraphCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ph_graph

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddKernelNodeV2Params

Bases: object

Empty-initialize an instance of cuGraphAddKernelNode_v2_params.

See also

cuGraphAddKernelNode_v2_params

dependencies

int:

static from_buffer(buffer)

Create an CuGraphAddKernelNodeV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddKernelNodeV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_kernel_node_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddKernelNodeV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

node_params

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphKernelNodeGetParamsV2Params

Bases: object

Empty-initialize an instance of cuGraphKernelNodeGetParams_v2_params.

See also

cuGraphKernelNodeGetParams_v2_params

static from_buffer(buffer)

Create an CuGraphKernelNodeGetParamsV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphKernelNodeGetParamsV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_kernel_node_get_params_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphKernelNodeGetParamsV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphKernelNodeSetParamsV2Params

Bases: object

Empty-initialize an instance of cuGraphKernelNodeSetParams_v2_params.

See also

cuGraphKernelNodeSetParams_v2_params

static from_buffer(buffer)

Create an CuGraphKernelNodeSetParamsV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphKernelNodeSetParamsV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_kernel_node_set_params_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphKernelNodeSetParamsV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddMemcpyNodeParams

Bases: object

Empty-initialize an instance of cuGraphAddMemcpyNode_params.

See also

cuGraphAddMemcpyNode_params

copy_params

int:

ctx

int:

dependencies

int:

static from_buffer(buffer)

Create an CuGraphAddMemcpyNodeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddMemcpyNodeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_memcpy_node_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddMemcpyNodeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphMemcpyNodeGetParamsParams

Bases: object

Empty-initialize an instance of cuGraphMemcpyNodeGetParams_params.

See also

cuGraphMemcpyNodeGetParams_params

static from_buffer(buffer)

Create an CuGraphMemcpyNodeGetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphMemcpyNodeGetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_memcpy_node_get_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphMemcpyNodeGetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphMemcpyNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphMemcpyNodeSetParams_params.

See also

cuGraphMemcpyNodeSetParams_params

static from_buffer(buffer)

Create an CuGraphMemcpyNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphMemcpyNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_memcpy_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphMemcpyNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddMemsetNodeParams

Bases: object

Empty-initialize an instance of cuGraphAddMemsetNode_params.

See also

cuGraphAddMemsetNode_params

ctx

int:

dependencies

int:

static from_buffer(buffer)

Create an CuGraphAddMemsetNodeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddMemsetNodeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_memset_node_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddMemsetNodeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

memset_params

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphMemsetNodeGetParamsParams

Bases: object

Empty-initialize an instance of cuGraphMemsetNodeGetParams_params.

See also

cuGraphMemsetNodeGetParams_params

static from_buffer(buffer)

Create an CuGraphMemsetNodeGetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphMemsetNodeGetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_memset_node_get_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphMemsetNodeGetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphMemsetNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphMemsetNodeSetParams_params.

See also

cuGraphMemsetNodeSetParams_params

static from_buffer(buffer)

Create an CuGraphMemsetNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphMemsetNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_memset_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphMemsetNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddHostNodeParams

Bases: object

Empty-initialize an instance of cuGraphAddHostNode_params.

See also

cuGraphAddHostNode_params

dependencies

int:

static from_buffer(buffer)

Create an CuGraphAddHostNodeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddHostNodeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_host_node_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddHostNodeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

node_params

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphHostNodeGetParamsParams

Bases: object

Empty-initialize an instance of cuGraphHostNodeGetParams_params.

See also

cuGraphHostNodeGetParams_params

static from_buffer(buffer)

Create an CuGraphHostNodeGetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphHostNodeGetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_host_node_get_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphHostNodeGetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphHostNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphHostNodeSetParams_params.

See also

cuGraphHostNodeSetParams_params

static from_buffer(buffer)

Create an CuGraphHostNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphHostNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_host_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphHostNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddChildGraphNodeParams

Bases: object

Empty-initialize an instance of cuGraphAddChildGraphNode_params.

See also

cuGraphAddChildGraphNode_params

child_graph

int:

dependencies

int:

static from_buffer(buffer)

Create an CuGraphAddChildGraphNodeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddChildGraphNodeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_child_graph_node_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddChildGraphNodeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphChildGraphNodeGetGraphParams

Bases: object

Empty-initialize an instance of cuGraphChildGraphNodeGetGraph_params.

See also

cuGraphChildGraphNodeGetGraph_params

static from_buffer(buffer)

Create an CuGraphChildGraphNodeGetGraphParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphChildGraphNodeGetGraphParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_child_graph_node_get_graph_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphChildGraphNodeGetGraphParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ph_graph

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddEmptyNodeParams

Bases: object

Empty-initialize an instance of cuGraphAddEmptyNode_params.

See also

cuGraphAddEmptyNode_params

dependencies

int:

static from_buffer(buffer)

Create an CuGraphAddEmptyNodeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddEmptyNodeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_empty_node_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddEmptyNodeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddEventRecordNodeParams

Bases: object

Empty-initialize an instance of cuGraphAddEventRecordNode_params.

See also

cuGraphAddEventRecordNode_params

dependencies

int:

event

int:

static from_buffer(buffer)

Create an CuGraphAddEventRecordNodeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddEventRecordNodeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_event_record_node_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddEventRecordNodeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphEventRecordNodeGetEventParams

Bases: object

Empty-initialize an instance of cuGraphEventRecordNodeGetEvent_params.

See also

cuGraphEventRecordNodeGetEvent_params

event_out

int:

static from_buffer(buffer)

Create an CuGraphEventRecordNodeGetEventParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphEventRecordNodeGetEventParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_event_record_node_get_event_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphEventRecordNodeGetEventParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphEventRecordNodeSetEventParams

Bases: object

Empty-initialize an instance of cuGraphEventRecordNodeSetEvent_params.

See also

cuGraphEventRecordNodeSetEvent_params

event

int:

static from_buffer(buffer)

Create an CuGraphEventRecordNodeSetEventParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphEventRecordNodeSetEventParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_event_record_node_set_event_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphEventRecordNodeSetEventParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddEventWaitNodeParams

Bases: object

Empty-initialize an instance of cuGraphAddEventWaitNode_params.

See also

cuGraphAddEventWaitNode_params

dependencies

int:

event

int:

static from_buffer(buffer)

Create an CuGraphAddEventWaitNodeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddEventWaitNodeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_event_wait_node_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddEventWaitNodeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphEventWaitNodeGetEventParams

Bases: object

Empty-initialize an instance of cuGraphEventWaitNodeGetEvent_params.

See also

cuGraphEventWaitNodeGetEvent_params

event_out

int:

static from_buffer(buffer)

Create an CuGraphEventWaitNodeGetEventParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphEventWaitNodeGetEventParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_event_wait_node_get_event_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphEventWaitNodeGetEventParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphEventWaitNodeSetEventParams

Bases: object

Empty-initialize an instance of cuGraphEventWaitNodeSetEvent_params.

See also

cuGraphEventWaitNodeSetEvent_params

event

int:

static from_buffer(buffer)

Create an CuGraphEventWaitNodeSetEventParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphEventWaitNodeSetEventParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_event_wait_node_set_event_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphEventWaitNodeSetEventParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddExternalSemaphoresSignalNodeParams

Bases: object

Empty-initialize an instance of cuGraphAddExternalSemaphoresSignalNode_params.

See also

cuGraphAddExternalSemaphoresSignalNode_params

dependencies

int:

static from_buffer(buffer)

Create an CuGraphAddExternalSemaphoresSignalNodeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddExternalSemaphoresSignalNodeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_external_semaphores_signal_node_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddExternalSemaphoresSignalNodeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

node_params

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExternalSemaphoresSignalNodeGetParamsParams

Bases: object

Empty-initialize an instance of cuGraphExternalSemaphoresSignalNodeGetParams_params.

See also

cuGraphExternalSemaphoresSignalNodeGetParams_params

static from_buffer(buffer)

Create an CuGraphExternalSemaphoresSignalNodeGetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExternalSemaphoresSignalNodeGetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_external_semaphores_signal_node_get_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExternalSemaphoresSignalNodeGetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

params_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExternalSemaphoresSignalNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphExternalSemaphoresSignalNodeSetParams_params.

See also

cuGraphExternalSemaphoresSignalNodeSetParams_params

static from_buffer(buffer)

Create an CuGraphExternalSemaphoresSignalNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExternalSemaphoresSignalNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_external_semaphores_signal_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExternalSemaphoresSignalNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddExternalSemaphoresWaitNodeParams

Bases: object

Empty-initialize an instance of cuGraphAddExternalSemaphoresWaitNode_params.

See also

cuGraphAddExternalSemaphoresWaitNode_params

dependencies

int:

static from_buffer(buffer)

Create an CuGraphAddExternalSemaphoresWaitNodeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddExternalSemaphoresWaitNodeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_external_semaphores_wait_node_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddExternalSemaphoresWaitNodeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

node_params

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExternalSemaphoresWaitNodeGetParamsParams

Bases: object

Empty-initialize an instance of cuGraphExternalSemaphoresWaitNodeGetParams_params.

See also

cuGraphExternalSemaphoresWaitNodeGetParams_params

static from_buffer(buffer)

Create an CuGraphExternalSemaphoresWaitNodeGetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExternalSemaphoresWaitNodeGetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_external_semaphores_wait_node_get_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExternalSemaphoresWaitNodeGetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

params_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExternalSemaphoresWaitNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphExternalSemaphoresWaitNodeSetParams_params.

See also

cuGraphExternalSemaphoresWaitNodeSetParams_params

static from_buffer(buffer)

Create an CuGraphExternalSemaphoresWaitNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExternalSemaphoresWaitNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_external_semaphores_wait_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExternalSemaphoresWaitNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddBatchMemOpNodeParams

Bases: object

Empty-initialize an instance of cuGraphAddBatchMemOpNode_params.

See also

cuGraphAddBatchMemOpNode_params

dependencies

int:

static from_buffer(buffer)

Create an CuGraphAddBatchMemOpNodeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddBatchMemOpNodeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_batch_mem_op_node_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddBatchMemOpNodeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

node_params

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphBatchMemOpNodeGetParamsParams

Bases: object

Empty-initialize an instance of cuGraphBatchMemOpNodeGetParams_params.

See also

cuGraphBatchMemOpNodeGetParams_params

static from_buffer(buffer)

Create an CuGraphBatchMemOpNodeGetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphBatchMemOpNodeGetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_batch_mem_op_node_get_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphBatchMemOpNodeGetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphBatchMemOpNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphBatchMemOpNodeSetParams_params.

See also

cuGraphBatchMemOpNodeSetParams_params

static from_buffer(buffer)

Create an CuGraphBatchMemOpNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphBatchMemOpNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_batch_mem_op_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphBatchMemOpNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecBatchMemOpNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphExecBatchMemOpNodeSetParams_params.

See also

cuGraphExecBatchMemOpNodeSetParams_params

static from_buffer(buffer)

Create an CuGraphExecBatchMemOpNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecBatchMemOpNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_batch_mem_op_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecBatchMemOpNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddMemAllocNodeParams

Bases: object

Empty-initialize an instance of cuGraphAddMemAllocNode_params.

See also

cuGraphAddMemAllocNode_params

dependencies

int:

static from_buffer(buffer)

Create an CuGraphAddMemAllocNodeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddMemAllocNodeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_mem_alloc_node_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddMemAllocNodeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

node_params

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphMemAllocNodeGetParamsParams

Bases: object

Empty-initialize an instance of cuGraphMemAllocNodeGetParams_params.

See also

cuGraphMemAllocNodeGetParams_params

static from_buffer(buffer)

Create an CuGraphMemAllocNodeGetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphMemAllocNodeGetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_mem_alloc_node_get_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphMemAllocNodeGetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

params_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddMemFreeNodeParams

Bases: object

Empty-initialize an instance of cuGraphAddMemFreeNode_params.

See also

cuGraphAddMemFreeNode_params

dependencies

int:

dptr

int:

static from_buffer(buffer)

Create an CuGraphAddMemFreeNodeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddMemFreeNodeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_mem_free_node_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddMemFreeNodeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphMemFreeNodeGetParamsParams

Bases: object

Empty-initialize an instance of cuGraphMemFreeNodeGetParams_params.

See also

cuGraphMemFreeNodeGetParams_params

dptr_out

int:

static from_buffer(buffer)

Create an CuGraphMemFreeNodeGetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphMemFreeNodeGetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_mem_free_node_get_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphMemFreeNodeGetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGraphMemTrimParams

Bases: object

Empty-initialize an instance of cuDeviceGraphMemTrim_params.

See also

cuDeviceGraphMemTrim_params

device_

int:

static from_buffer(buffer)

Create an CuDeviceGraphMemTrimParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGraphMemTrimParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_graph_mem_trim_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGraphMemTrimParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetGraphMemAttributeParams

Bases: object

Empty-initialize an instance of cuDeviceGetGraphMemAttribute_params.

See also

cuDeviceGetGraphMemAttribute_params

attr

int:

device_

int:

static from_buffer(buffer)

Create an CuDeviceGetGraphMemAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetGraphMemAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_graph_mem_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetGraphMemAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CuDeviceSetGraphMemAttributeParams

Bases: object

Empty-initialize an instance of cuDeviceSetGraphMemAttribute_params.

See also

cuDeviceSetGraphMemAttribute_params

attr

int:

device_

int:

static from_buffer(buffer)

Create an CuDeviceSetGraphMemAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceSetGraphMemAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_set_graph_mem_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceSetGraphMemAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CuGraphCloneParams

Bases: object

Empty-initialize an instance of cuGraphClone_params.

See also

cuGraphClone_params

static from_buffer(buffer)

Create an CuGraphCloneParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphCloneParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_clone_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphCloneParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

original_graph

int:

ph_graph_clone

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphNodeFindInCloneParams

Bases: object

Empty-initialize an instance of cuGraphNodeFindInClone_params.

See also

cuGraphNodeFindInClone_params

static from_buffer(buffer)

Create an CuGraphNodeFindInCloneParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphNodeFindInCloneParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_node_find_in_clone_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphNodeFindInCloneParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_cloned_graph

int:

h_original_node

int:

ph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphNodeGetTypeParams

Bases: object

Empty-initialize an instance of cuGraphNodeGetType_params.

See also

cuGraphNodeGetType_params

static from_buffer(buffer)

Create an CuGraphNodeGetTypeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphNodeGetTypeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_node_get_type_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphNodeGetTypeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ptr

Get the pointer address to the data as Python int.

type

int:

class cupti.cupti.CuGraphGetNodesParams

Bases: object

Empty-initialize an instance of cuGraphGetNodes_params.

See also

cuGraphGetNodes_params

static from_buffer(buffer)

Create an CuGraphGetNodesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphGetNodesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_get_nodes_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphGetNodesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

nodes

int:

num_nodes

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphGetRootNodesParams

Bases: object

Empty-initialize an instance of cuGraphGetRootNodes_params.

See also

cuGraphGetRootNodes_params

static from_buffer(buffer)

Create an CuGraphGetRootNodesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphGetRootNodesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_get_root_nodes_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphGetRootNodesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

num_root_nodes

int:

ptr

Get the pointer address to the data as Python int.

root_nodes

int:

class cupti.cupti.CuGraphGetEdgesParams

Bases: object

Empty-initialize an instance of cuGraphGetEdges_params.

See also

cuGraphGetEdges_params

from_

int:

static from_buffer(buffer)

Create an CuGraphGetEdgesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphGetEdgesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_get_edges_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphGetEdgesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

num_edges

int:

ptr

Get the pointer address to the data as Python int.

to

int:

class cupti.cupti.CuGraphGetEdgesV2Params

Bases: object

Empty-initialize an instance of cuGraphGetEdges_v2_params.

See also

cuGraphGetEdges_v2_params

edge_data

int:

from_

int:

static from_buffer(buffer)

Create an CuGraphGetEdgesV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphGetEdgesV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_get_edges_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphGetEdgesV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

num_edges

int:

ptr

Get the pointer address to the data as Python int.

to

int:

class cupti.cupti.CuGraphNodeGetDependenciesParams

Bases: object

Empty-initialize an instance of cuGraphNodeGetDependencies_params.

See also

cuGraphNodeGetDependencies_params

dependencies

int:

static from_buffer(buffer)

Create an CuGraphNodeGetDependenciesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphNodeGetDependenciesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_node_get_dependencies_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphNodeGetDependenciesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphNodeGetDependenciesV2Params

Bases: object

Empty-initialize an instance of cuGraphNodeGetDependencies_v2_params.

See also

cuGraphNodeGetDependencies_v2_params

dependencies

int:

edge_data

int:

static from_buffer(buffer)

Create an CuGraphNodeGetDependenciesV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphNodeGetDependenciesV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_node_get_dependencies_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphNodeGetDependenciesV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphNodeGetDependentNodesParams

Bases: object

Empty-initialize an instance of cuGraphNodeGetDependentNodes_params.

See also

cuGraphNodeGetDependentNodes_params

dependent_nodes

int:

static from_buffer(buffer)

Create an CuGraphNodeGetDependentNodesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphNodeGetDependentNodesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_node_get_dependent_nodes_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphNodeGetDependentNodesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

num_dependent_nodes

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphNodeGetDependentNodesV2Params

Bases: object

Empty-initialize an instance of cuGraphNodeGetDependentNodes_v2_params.

See also

cuGraphNodeGetDependentNodes_v2_params

dependent_nodes

int:

edge_data

int:

static from_buffer(buffer)

Create an CuGraphNodeGetDependentNodesV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphNodeGetDependentNodesV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_node_get_dependent_nodes_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphNodeGetDependentNodesV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

num_dependent_nodes

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddDependenciesParams

Bases: object

Empty-initialize an instance of cuGraphAddDependencies_params.

See also

cuGraphAddDependencies_params

from_

int:

static from_buffer(buffer)

Create an CuGraphAddDependenciesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddDependenciesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_dependencies_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddDependenciesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

to

int:

class cupti.cupti.CuGraphAddDependenciesV2Params

Bases: object

Empty-initialize an instance of cuGraphAddDependencies_v2_params.

See also

cuGraphAddDependencies_v2_params

edge_data

int:

from_

int:

static from_buffer(buffer)

Create an CuGraphAddDependenciesV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddDependenciesV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_dependencies_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddDependenciesV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

to

int:

class cupti.cupti.CuGraphRemoveDependenciesParams

Bases: object

Empty-initialize an instance of cuGraphRemoveDependencies_params.

See also

cuGraphRemoveDependencies_params

from_

int:

static from_buffer(buffer)

Create an CuGraphRemoveDependenciesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphRemoveDependenciesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_remove_dependencies_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphRemoveDependenciesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

to

int:

class cupti.cupti.CuGraphRemoveDependenciesV2Params

Bases: object

Empty-initialize an instance of cuGraphRemoveDependencies_v2_params.

See also

cuGraphRemoveDependencies_v2_params

edge_data

int:

from_

int:

static from_buffer(buffer)

Create an CuGraphRemoveDependenciesV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphRemoveDependenciesV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_remove_dependencies_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphRemoveDependenciesV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

to

int:

class cupti.cupti.CuGraphDestroyNodeParams

Bases: object

Empty-initialize an instance of cuGraphDestroyNode_params.

See also

cuGraphDestroyNode_params

static from_buffer(buffer)

Create an CuGraphDestroyNodeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphDestroyNodeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_destroy_node_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphDestroyNodeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphInstantiateWithFlagsParams

Bases: object

Empty-initialize an instance of cuGraphInstantiateWithFlags_params.

See also

cuGraphInstantiateWithFlags_params

flags_

int:

static from_buffer(buffer)

Create an CuGraphInstantiateWithFlagsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphInstantiateWithFlagsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_instantiate_with_flags_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphInstantiateWithFlagsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

ph_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphInstantiateWithParamsPtszParams

Bases: object

Empty-initialize an instance of cuGraphInstantiateWithParams_ptsz_params.

See also

cuGraphInstantiateWithParams_ptsz_params

static from_buffer(buffer)

Create an CuGraphInstantiateWithParamsPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphInstantiateWithParamsPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_instantiate_with_params_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphInstantiateWithParamsPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

instantiate_params

int:

ph_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecGetFlagsParams

Bases: object

Empty-initialize an instance of cuGraphExecGetFlags_params.

See also

cuGraphExecGetFlags_params

flags_

int:

static from_buffer(buffer)

Create an CuGraphExecGetFlagsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecGetFlagsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_get_flags_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecGetFlagsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecKernelNodeSetParamsV2Params

Bases: object

Empty-initialize an instance of cuGraphExecKernelNodeSetParams_v2_params.

See also

cuGraphExecKernelNodeSetParams_v2_params

static from_buffer(buffer)

Create an CuGraphExecKernelNodeSetParamsV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecKernelNodeSetParamsV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_kernel_node_set_params_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecKernelNodeSetParamsV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecMemcpyNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphExecMemcpyNodeSetParams_params.

See also

cuGraphExecMemcpyNodeSetParams_params

copy_params

int:

ctx

int:

static from_buffer(buffer)

Create an CuGraphExecMemcpyNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecMemcpyNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_memcpy_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecMemcpyNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecMemsetNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphExecMemsetNodeSetParams_params.

See also

cuGraphExecMemsetNodeSetParams_params

ctx

int:

static from_buffer(buffer)

Create an CuGraphExecMemsetNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecMemsetNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_memset_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecMemsetNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

memset_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecHostNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphExecHostNodeSetParams_params.

See also

cuGraphExecHostNodeSetParams_params

static from_buffer(buffer)

Create an CuGraphExecHostNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecHostNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_host_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecHostNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecChildGraphNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphExecChildGraphNodeSetParams_params.

See also

cuGraphExecChildGraphNodeSetParams_params

child_graph

int:

static from_buffer(buffer)

Create an CuGraphExecChildGraphNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecChildGraphNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_child_graph_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecChildGraphNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecEventRecordNodeSetEventParams

Bases: object

Empty-initialize an instance of cuGraphExecEventRecordNodeSetEvent_params.

See also

cuGraphExecEventRecordNodeSetEvent_params

event

int:

static from_buffer(buffer)

Create an CuGraphExecEventRecordNodeSetEventParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecEventRecordNodeSetEventParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_event_record_node_set_event_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecEventRecordNodeSetEventParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecEventWaitNodeSetEventParams

Bases: object

Empty-initialize an instance of cuGraphExecEventWaitNodeSetEvent_params.

See also

cuGraphExecEventWaitNodeSetEvent_params

event

int:

static from_buffer(buffer)

Create an CuGraphExecEventWaitNodeSetEventParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecEventWaitNodeSetEventParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_event_wait_node_set_event_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecEventWaitNodeSetEventParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecExternalSemaphoresSignalNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphExecExternalSemaphoresSignalNodeSetParams_params.

See also

cuGraphExecExternalSemaphoresSignalNodeSetParams_params

static from_buffer(

buffer,

)

Create an CuGraphExecExternalSemaphoresSignalNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecExternalSemaphoresSignalNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_external_semaphores_signal_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecExternalSemaphoresSignalNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecExternalSemaphoresWaitNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphExecExternalSemaphoresWaitNodeSetParams_params.

See also

cuGraphExecExternalSemaphoresWaitNodeSetParams_params

static from_buffer(

buffer,

)

Create an CuGraphExecExternalSemaphoresWaitNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecExternalSemaphoresWaitNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_external_semaphores_wait_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecExternalSemaphoresWaitNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphNodeSetEnabledParams

Bases: object

Empty-initialize an instance of cuGraphNodeSetEnabled_params.

See also

cuGraphNodeSetEnabled_params

static from_buffer(buffer)

Create an CuGraphNodeSetEnabledParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphNodeSetEnabledParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_node_set_enabled_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphNodeSetEnabledParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

is_enabled

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphNodeGetEnabledParams

Bases: object

Empty-initialize an instance of cuGraphNodeGetEnabled_params.

See also

cuGraphNodeGetEnabled_params

static from_buffer(buffer)

Create an CuGraphNodeGetEnabledParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphNodeGetEnabledParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_node_get_enabled_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphNodeGetEnabledParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

is_enabled

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphUploadPtszParams

Bases: object

Empty-initialize an instance of cuGraphUpload_ptsz_params.

See also

cuGraphUpload_ptsz_params

static from_buffer(buffer)

Create an CuGraphUploadPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphUploadPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_upload_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphUploadPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphLaunchPtszParams

Bases: object

Empty-initialize an instance of cuGraphLaunch_ptsz_params.

See also

cuGraphLaunch_ptsz_params

static from_buffer(buffer)

Create an CuGraphLaunchPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphLaunchPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_launch_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphLaunchPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecDestroyParams

Bases: object

Empty-initialize an instance of cuGraphExecDestroy_params.

See also

cuGraphExecDestroy_params

static from_buffer(buffer)

Create an CuGraphExecDestroyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecDestroyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_destroy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecDestroyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphDestroyParams

Bases: object

Empty-initialize an instance of cuGraphDestroy_params.

See also

cuGraphDestroy_params

static from_buffer(buffer)

Create an CuGraphDestroyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphDestroyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_destroy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphDestroyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecUpdateV2Params

Bases: object

Empty-initialize an instance of cuGraphExecUpdate_v2_params.

See also

cuGraphExecUpdate_v2_params

static from_buffer(buffer)

Create an CuGraphExecUpdateV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecUpdateV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_update_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecUpdateV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

h_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

result_info

int:

class cupti.cupti.CuGraphKernelNodeCopyAttributesParams

Bases: object

Empty-initialize an instance of cuGraphKernelNodeCopyAttributes_params.

See also

cuGraphKernelNodeCopyAttributes_params

dst

int:

static from_buffer(buffer)

Create an CuGraphKernelNodeCopyAttributesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphKernelNodeCopyAttributesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_kernel_node_copy_attributes_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphKernelNodeCopyAttributesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src

int:

class cupti.cupti.CuGraphKernelNodeGetAttributeParams

Bases: object

Empty-initialize an instance of cuGraphKernelNodeGetAttribute_params.

See also

cuGraphKernelNodeGetAttribute_params

attr

int:

static from_buffer(buffer)

Create an CuGraphKernelNodeGetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphKernelNodeGetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_kernel_node_get_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphKernelNodeGetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ptr

Get the pointer address to the data as Python int.

value_out

int:

class cupti.cupti.CuGraphKernelNodeSetAttributeParams

Bases: object

Empty-initialize an instance of cuGraphKernelNodeSetAttribute_params.

See also

cuGraphKernelNodeSetAttribute_params

attr

int:

static from_buffer(buffer)

Create an CuGraphKernelNodeSetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphKernelNodeSetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_kernel_node_set_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphKernelNodeSetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CuGraphDebugDotPrintParams

Bases: object

Empty-initialize an instance of cuGraphDebugDotPrint_params.

See also

cuGraphDebugDotPrint_params

flags_

int:

static from_buffer(buffer)

Create an CuGraphDebugDotPrintParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphDebugDotPrintParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_debug_dot_print_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphDebugDotPrintParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

path

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuUserObjectCreateParams

Bases: object

Empty-initialize an instance of cuUserObjectCreate_params.

See also

cuUserObjectCreate_params

destroy

int:

flags_

int:

static from_buffer(buffer)

Create an CuUserObjectCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuUserObjectCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_user_object_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuUserObjectCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

initial_refcount

int:

object_out

int:

ptr

Get the pointer address to the data as Python int.

ptr_

int:

class cupti.cupti.CuUserObjectRetainParams

Bases: object

Empty-initialize an instance of cuUserObjectRetain_params.

See also

cuUserObjectRetain_params

count

int:

static from_buffer(buffer)

Create an CuUserObjectRetainParams instance with the memory from the given buffer.

static from_data(data)

Create an CuUserObjectRetainParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_user_object_retain_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuUserObjectRetainParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

object

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuUserObjectReleaseParams

Bases: object

Empty-initialize an instance of cuUserObjectRelease_params.

See also

cuUserObjectRelease_params

count

int:

static from_buffer(buffer)

Create an CuUserObjectReleaseParams instance with the memory from the given buffer.

static from_data(data)

Create an CuUserObjectReleaseParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_user_object_release_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuUserObjectReleaseParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

object

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphRetainUserObjectParams

Bases: object

Empty-initialize an instance of cuGraphRetainUserObject_params.

See also

cuGraphRetainUserObject_params

count

int:

flags_

int:

static from_buffer(buffer)

Create an CuGraphRetainUserObjectParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphRetainUserObjectParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_retain_user_object_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphRetainUserObjectParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

object

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphReleaseUserObjectParams

Bases: object

Empty-initialize an instance of cuGraphReleaseUserObject_params.

See also

cuGraphReleaseUserObject_params

count

int:

static from_buffer(buffer)

Create an CuGraphReleaseUserObjectParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphReleaseUserObjectParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_release_user_object_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphReleaseUserObjectParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph

int:

object

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddNodeParams

Bases: object

Empty-initialize an instance of cuGraphAddNode_params.

See also

cuGraphAddNode_params

dependencies

int:

static from_buffer(buffer)

Create an CuGraphAddNodeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddNodeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_node_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddNodeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

node_params

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddNodeV2Params

Bases: object

Empty-initialize an instance of cuGraphAddNode_v2_params.

See also

cuGraphAddNode_v2_params

dependencies

int:

dependency_data

int:

static from_buffer(buffer)

Create an CuGraphAddNodeV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddNodeV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_node_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddNodeV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

node_params

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphNodeSetParams_params.

See also

cuGraphNodeSetParams_params

static from_buffer(buffer)

Create an CuGraphNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphExecNodeSetParams_params.

See also

cuGraphExecNodeSetParams_params

static from_buffer(buffer)

Create an CuGraphExecNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphConditionalHandleCreateParams

Bases: object

Empty-initialize an instance of cuGraphConditionalHandleCreate_params.

See also

cuGraphConditionalHandleCreate_params

ctx

int:

default_launch_value

int:

flags_

int:

static from_buffer(buffer)

Create an CuGraphConditionalHandleCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphConditionalHandleCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_conditional_handle_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphConditionalHandleCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

p_handle_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuOccupancyMaxActiveBlocksPerMultiprocessorParams

Bases: object

Empty-initialize an instance of cuOccupancyMaxActiveBlocksPerMultiprocessor_params.

See also

cuOccupancyMaxActiveBlocksPerMultiprocessor_params

block_size

int:

dynamic_s_mem_size

int:

static from_buffer(buffer)

Create an CuOccupancyMaxActiveBlocksPerMultiprocessorParams instance with the memory from the given buffer.

static from_data(data)

Create an CuOccupancyMaxActiveBlocksPerMultiprocessorParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_occupancy_max_active_blocks_per_multiprocessor_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuOccupancyMaxActiveBlocksPerMultiprocessorParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

num_blocks

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuOccupancyMaxActiveBlocksPerMultiprocessorWithFlagsParams

Bases: object

Empty-initialize an instance of cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags_params.

See also

cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags_params

block_size

int:

dynamic_s_mem_size

int:

flags_

int:

static from_buffer(

buffer,

)

Create an CuOccupancyMaxActiveBlocksPerMultiprocessorWithFlagsParams instance with the memory from the given buffer.

static from_data(

data,

)

Create an CuOccupancyMaxActiveBlocksPerMultiprocessorWithFlagsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_occupancy_max_active_blocks_per_multiprocessor_with_flags_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuOccupancyMaxActiveBlocksPerMultiprocessorWithFlagsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

num_blocks

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuOccupancyMaxPotentialBlockSizeParams

Bases: object

Empty-initialize an instance of cuOccupancyMaxPotentialBlockSize_params.

See also

cuOccupancyMaxPotentialBlockSize_params

block_size

int:

block_size_limit

int:

block_size_to_dynamic_s_mem_size

int:

dynamic_s_mem_size

int:

static from_buffer(buffer)

Create an CuOccupancyMaxPotentialBlockSizeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuOccupancyMaxPotentialBlockSizeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_occupancy_max_potential_block_size_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuOccupancyMaxPotentialBlockSizeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

min_grid_size

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuOccupancyMaxPotentialBlockSizeWithFlagsParams

Bases: object

Empty-initialize an instance of cuOccupancyMaxPotentialBlockSizeWithFlags_params.

See also

cuOccupancyMaxPotentialBlockSizeWithFlags_params

block_size

int:

block_size_limit

int:

block_size_to_dynamic_s_mem_size

int:

dynamic_s_mem_size

int:

flags_

int:

static from_buffer(buffer)

Create an CuOccupancyMaxPotentialBlockSizeWithFlagsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuOccupancyMaxPotentialBlockSizeWithFlagsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_occupancy_max_potential_block_size_with_flags_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuOccupancyMaxPotentialBlockSizeWithFlagsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

min_grid_size

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuOccupancyAvailableDynamicSMemPerBlockParams

Bases: object

Empty-initialize an instance of cuOccupancyAvailableDynamicSMemPerBlock_params.

See also

cuOccupancyAvailableDynamicSMemPerBlock_params

block_size

int:

dynamic_smem_size

int:

static from_buffer(buffer)

Create an CuOccupancyAvailableDynamicSMemPerBlockParams instance with the memory from the given buffer.

static from_data(data)

Create an CuOccupancyAvailableDynamicSMemPerBlockParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_occupancy_available_dynamic_s_mem_per_block_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuOccupancyAvailableDynamicSMemPerBlockParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

num_blocks

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuOccupancyMaxPotentialClusterSizeParams

Bases: object

Empty-initialize an instance of cuOccupancyMaxPotentialClusterSize_params.

See also

cuOccupancyMaxPotentialClusterSize_params

cluster_size

int:

config

int:

static from_buffer(buffer)

Create an CuOccupancyMaxPotentialClusterSizeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuOccupancyMaxPotentialClusterSizeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_occupancy_max_potential_cluster_size_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuOccupancyMaxPotentialClusterSizeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuOccupancyMaxActiveClustersParams

Bases: object

Empty-initialize an instance of cuOccupancyMaxActiveClusters_params.

See also

cuOccupancyMaxActiveClusters_params

config

int:

static from_buffer(buffer)

Create an CuOccupancyMaxActiveClustersParams instance with the memory from the given buffer.

static from_data(data)

Create an CuOccupancyMaxActiveClustersParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_occupancy_max_active_clusters_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuOccupancyMaxActiveClustersParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

num_clusters

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefSetArrayParams

Bases: object

Empty-initialize an instance of cuTexRefSetArray_params.

See also

cuTexRefSetArray_params

flags_

int:

static from_buffer(buffer)

Create an CuTexRefSetArrayParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetArrayParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_array_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetArrayParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_array

int:

h_tex_ref

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefSetMipmappedArrayParams

Bases: object

Empty-initialize an instance of cuTexRefSetMipmappedArray_params.

See also

cuTexRefSetMipmappedArray_params

flags_

int:

static from_buffer(buffer)

Create an CuTexRefSetMipmappedArrayParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetMipmappedArrayParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_mipmapped_array_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetMipmappedArrayParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_mipmapped_array

int:

h_tex_ref

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefSetAddressV2Params

Bases: object

Empty-initialize an instance of cuTexRefSetAddress_v2_params.

See also

cuTexRefSetAddress_v2_params

byte_offset

int:

bytes

int:

dptr

int:

static from_buffer(buffer)

Create an CuTexRefSetAddressV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetAddressV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_address_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetAddressV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefSetAddress2DV3Params

Bases: object

Empty-initialize an instance of cuTexRefSetAddress2D_v3_params.

See also

cuTexRefSetAddress2D_v3_params

desc

int:

dptr

int:

static from_buffer(buffer)

Create an CuTexRefSetAddress2DV3Params instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetAddress2DV3Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_address2d_v3params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetAddress2DV3Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

pitch

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefSetFormatParams

Bases: object

Empty-initialize an instance of cuTexRefSetFormat_params.

See also

cuTexRefSetFormat_params

fmt

int:

static from_buffer(buffer)

Create an CuTexRefSetFormatParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetFormatParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_format_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetFormatParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

num_packed_components

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefSetAddressModeParams

Bases: object

Empty-initialize an instance of cuTexRefSetAddressMode_params.

See also

cuTexRefSetAddressMode_params

am

int:

dim

int:

static from_buffer(buffer)

Create an CuTexRefSetAddressModeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetAddressModeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_address_mode_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetAddressModeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefSetFilterModeParams

Bases: object

Empty-initialize an instance of cuTexRefSetFilterMode_params.

See also

cuTexRefSetFilterMode_params

fm

int:

static from_buffer(buffer)

Create an CuTexRefSetFilterModeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetFilterModeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_filter_mode_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetFilterModeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefSetMipmapFilterModeParams

Bases: object

Empty-initialize an instance of cuTexRefSetMipmapFilterMode_params.

See also

cuTexRefSetMipmapFilterMode_params

fm

int:

static from_buffer(buffer)

Create an CuTexRefSetMipmapFilterModeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetMipmapFilterModeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_mipmap_filter_mode_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetMipmapFilterModeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefSetMipmapLevelBiasParams

Bases: object

Empty-initialize an instance of cuTexRefSetMipmapLevelBias_params.

See also

cuTexRefSetMipmapLevelBias_params

bias

float:

static from_buffer(buffer)

Create an CuTexRefSetMipmapLevelBiasParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetMipmapLevelBiasParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_mipmap_level_bias_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetMipmapLevelBiasParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefSetMipmapLevelClampParams

Bases: object

Empty-initialize an instance of cuTexRefSetMipmapLevelClamp_params.

See also

cuTexRefSetMipmapLevelClamp_params

static from_buffer(buffer)

Create an CuTexRefSetMipmapLevelClampParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetMipmapLevelClampParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_mipmap_level_clamp_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetMipmapLevelClampParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

max_mipmap_level_clamp

float:

min_mipmap_level_clamp

float:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefSetMaxAnisotropyParams

Bases: object

Empty-initialize an instance of cuTexRefSetMaxAnisotropy_params.

See also

cuTexRefSetMaxAnisotropy_params

static from_buffer(buffer)

Create an CuTexRefSetMaxAnisotropyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetMaxAnisotropyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_max_anisotropy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetMaxAnisotropyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

max_aniso

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefSetBorderColorParams

Bases: object

Empty-initialize an instance of cuTexRefSetBorderColor_params.

See also

cuTexRefSetBorderColor_params

static from_buffer(buffer)

Create an CuTexRefSetBorderColorParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetBorderColorParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_border_color_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetBorderColorParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

p_border_color

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefSetFlagsParams

Bases: object

Empty-initialize an instance of cuTexRefSetFlags_params.

See also

cuTexRefSetFlags_params

flags_

int:

static from_buffer(buffer)

Create an CuTexRefSetFlagsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetFlagsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_flags_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetFlagsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefGetAddressV2Params

Bases: object

Empty-initialize an instance of cuTexRefGetAddress_v2_params.

See also

cuTexRefGetAddress_v2_params

static from_buffer(buffer)

Create an CuTexRefGetAddressV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefGetAddressV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_get_address_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefGetAddressV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

pdptr

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefGetArrayParams

Bases: object

Empty-initialize an instance of cuTexRefGetArray_params.

See also

cuTexRefGetArray_params

static from_buffer(buffer)

Create an CuTexRefGetArrayParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefGetArrayParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_get_array_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefGetArrayParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

ph_array

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefGetMipmappedArrayParams

Bases: object

Empty-initialize an instance of cuTexRefGetMipmappedArray_params.

See also

cuTexRefGetMipmappedArray_params

static from_buffer(buffer)

Create an CuTexRefGetMipmappedArrayParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefGetMipmappedArrayParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_get_mipmapped_array_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefGetMipmappedArrayParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

ph_mipmapped_array

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefGetAddressModeParams

Bases: object

Empty-initialize an instance of cuTexRefGetAddressMode_params.

See also

cuTexRefGetAddressMode_params

dim

int:

static from_buffer(buffer)

Create an CuTexRefGetAddressModeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefGetAddressModeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_get_address_mode_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefGetAddressModeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

pam

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefGetFilterModeParams

Bases: object

Empty-initialize an instance of cuTexRefGetFilterMode_params.

See also

cuTexRefGetFilterMode_params

static from_buffer(buffer)

Create an CuTexRefGetFilterModeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefGetFilterModeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_get_filter_mode_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefGetFilterModeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

pfm

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefGetFormatParams

Bases: object

Empty-initialize an instance of cuTexRefGetFormat_params.

See also

cuTexRefGetFormat_params

static from_buffer(buffer)

Create an CuTexRefGetFormatParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefGetFormatParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_get_format_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefGetFormatParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

p_format

int:

p_num_channels

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefGetMipmapFilterModeParams

Bases: object

Empty-initialize an instance of cuTexRefGetMipmapFilterMode_params.

See also

cuTexRefGetMipmapFilterMode_params

static from_buffer(buffer)

Create an CuTexRefGetMipmapFilterModeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefGetMipmapFilterModeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_get_mipmap_filter_mode_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefGetMipmapFilterModeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

pfm

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefGetMipmapLevelBiasParams

Bases: object

Empty-initialize an instance of cuTexRefGetMipmapLevelBias_params.

See also

cuTexRefGetMipmapLevelBias_params

static from_buffer(buffer)

Create an CuTexRefGetMipmapLevelBiasParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefGetMipmapLevelBiasParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_get_mipmap_level_bias_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefGetMipmapLevelBiasParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

pbias

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefGetMipmapLevelClampParams

Bases: object

Empty-initialize an instance of cuTexRefGetMipmapLevelClamp_params.

See also

cuTexRefGetMipmapLevelClamp_params

static from_buffer(buffer)

Create an CuTexRefGetMipmapLevelClampParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefGetMipmapLevelClampParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_get_mipmap_level_clamp_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefGetMipmapLevelClampParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

pmax_mipmap_level_clamp

int:

pmin_mipmap_level_clamp

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefGetMaxAnisotropyParams

Bases: object

Empty-initialize an instance of cuTexRefGetMaxAnisotropy_params.

See also

cuTexRefGetMaxAnisotropy_params

static from_buffer(buffer)

Create an CuTexRefGetMaxAnisotropyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefGetMaxAnisotropyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_get_max_anisotropy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefGetMaxAnisotropyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

pmax_aniso

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefGetBorderColorParams

Bases: object

Empty-initialize an instance of cuTexRefGetBorderColor_params.

See also

cuTexRefGetBorderColor_params

static from_buffer(buffer)

Create an CuTexRefGetBorderColorParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefGetBorderColorParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_get_border_color_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefGetBorderColorParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

p_border_color

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefGetFlagsParams

Bases: object

Empty-initialize an instance of cuTexRefGetFlags_params.

See also

cuTexRefGetFlags_params

static from_buffer(buffer)

Create an CuTexRefGetFlagsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefGetFlagsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_get_flags_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefGetFlagsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

p_flags

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefCreateParams

Bases: object

Empty-initialize an instance of cuTexRefCreate_params.

See also

cuTexRefCreate_params

static from_buffer(buffer)

Create an CuTexRefCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_tex_ref

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefDestroyParams

Bases: object

Empty-initialize an instance of cuTexRefDestroy_params.

See also

cuTexRefDestroy_params

static from_buffer(buffer)

Create an CuTexRefDestroyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefDestroyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_destroy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefDestroyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuSurfRefSetArrayParams

Bases: object

Empty-initialize an instance of cuSurfRefSetArray_params.

See also

cuSurfRefSetArray_params

flags_

int:

static from_buffer(buffer)

Create an CuSurfRefSetArrayParams instance with the memory from the given buffer.

static from_data(data)

Create an CuSurfRefSetArrayParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_surf_ref_set_array_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuSurfRefSetArrayParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_array

int:

h_surf_ref

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuSurfRefGetArrayParams

Bases: object

Empty-initialize an instance of cuSurfRefGetArray_params.

See also

cuSurfRefGetArray_params

static from_buffer(buffer)

Create an CuSurfRefGetArrayParams instance with the memory from the given buffer.

static from_data(data)

Create an CuSurfRefGetArrayParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_surf_ref_get_array_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuSurfRefGetArrayParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_surf_ref

int:

ph_array

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexObjectCreateParams

Bases: object

Empty-initialize an instance of cuTexObjectCreate_params.

See also

cuTexObjectCreate_params

static from_buffer(buffer)

Create an CuTexObjectCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexObjectCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_object_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexObjectCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_res_desc

int:

p_res_view_desc

int:

p_tex_desc

int:

p_tex_object

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexObjectDestroyParams

Bases: object

Empty-initialize an instance of cuTexObjectDestroy_params.

See also

cuTexObjectDestroy_params

static from_buffer(buffer)

Create an CuTexObjectDestroyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexObjectDestroyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_object_destroy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexObjectDestroyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

tex_object

int:

class cupti.cupti.CuTexObjectGetResourceDescParams

Bases: object

Empty-initialize an instance of cuTexObjectGetResourceDesc_params.

See also

cuTexObjectGetResourceDesc_params

static from_buffer(buffer)

Create an CuTexObjectGetResourceDescParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexObjectGetResourceDescParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_object_get_resource_desc_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexObjectGetResourceDescParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_res_desc

int:

ptr

Get the pointer address to the data as Python int.

tex_object

int:

class cupti.cupti.CuTexObjectGetTextureDescParams

Bases: object

Empty-initialize an instance of cuTexObjectGetTextureDesc_params.

See also

cuTexObjectGetTextureDesc_params

static from_buffer(buffer)

Create an CuTexObjectGetTextureDescParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexObjectGetTextureDescParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_object_get_texture_desc_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexObjectGetTextureDescParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_tex_desc

int:

ptr

Get the pointer address to the data as Python int.

tex_object

int:

class cupti.cupti.CuTexObjectGetResourceViewDescParams

Bases: object

Empty-initialize an instance of cuTexObjectGetResourceViewDesc_params.

See also

cuTexObjectGetResourceViewDesc_params

static from_buffer(buffer)

Create an CuTexObjectGetResourceViewDescParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexObjectGetResourceViewDescParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_object_get_resource_view_desc_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexObjectGetResourceViewDescParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_res_view_desc

int:

ptr

Get the pointer address to the data as Python int.

tex_object

int:

class cupti.cupti.CuSurfObjectCreateParams

Bases: object

Empty-initialize an instance of cuSurfObjectCreate_params.

See also

cuSurfObjectCreate_params

static from_buffer(buffer)

Create an CuSurfObjectCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuSurfObjectCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_surf_object_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuSurfObjectCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_res_desc

int:

p_surf_object

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuSurfObjectDestroyParams

Bases: object

Empty-initialize an instance of cuSurfObjectDestroy_params.

See also

cuSurfObjectDestroy_params

static from_buffer(buffer)

Create an CuSurfObjectDestroyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuSurfObjectDestroyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_surf_object_destroy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuSurfObjectDestroyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

surf_object

int:

class cupti.cupti.CuSurfObjectGetResourceDescParams

Bases: object

Empty-initialize an instance of cuSurfObjectGetResourceDesc_params.

See also

cuSurfObjectGetResourceDesc_params

static from_buffer(buffer)

Create an CuSurfObjectGetResourceDescParams instance with the memory from the given buffer.

static from_data(data)

Create an CuSurfObjectGetResourceDescParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_surf_object_get_resource_desc_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuSurfObjectGetResourceDescParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_res_desc

int:

ptr

Get the pointer address to the data as Python int.

surf_object

int:

class cupti.cupti.CuTensorMapEncodeTiledParams

Bases: object

Empty-initialize an instance of cuTensorMapEncodeTiled_params.

See also

cuTensorMapEncodeTiled_params

box_dim

int:

element_strides

int:

static from_buffer(buffer)

Create an CuTensorMapEncodeTiledParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTensorMapEncodeTiledParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tensor_map_encode_tiled_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTensorMapEncodeTiledParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

global_address

int:

global_dim

int:

global_strides

int:

interleave

int:

l2promotion

int:

oob_fill

int:

ptr

Get the pointer address to the data as Python int.

swizzle

int:

tensor_data_type

int:

tensor_map

int:

tensor_rank

int:

class cupti.cupti.CuTensorMapEncodeIm2colParams

Bases: object

Empty-initialize an instance of cuTensorMapEncodeIm2col_params.

See also

cuTensorMapEncodeIm2col_params

channels_per_pixel

int:

element_strides

int:

static from_buffer(buffer)

Create an CuTensorMapEncodeIm2colParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTensorMapEncodeIm2colParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tensor_map_encode_im2col_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTensorMapEncodeIm2colParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

global_address

int:

global_dim

int:

global_strides

int:

interleave

int:

l2promotion

int:

oob_fill

int:

pixel_box_lower_corner

int:

pixel_box_upper_corner

int:

pixels_per_column

int:

ptr

Get the pointer address to the data as Python int.

swizzle

int:

tensor_data_type

int:

tensor_map

int:

tensor_rank

int:

class cupti.cupti.CuTensorMapReplaceAddressParams

Bases: object

Empty-initialize an instance of cuTensorMapReplaceAddress_params.

See also

cuTensorMapReplaceAddress_params

static from_buffer(buffer)

Create an CuTensorMapReplaceAddressParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTensorMapReplaceAddressParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tensor_map_replace_address_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTensorMapReplaceAddressParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

global_address

int:

ptr

Get the pointer address to the data as Python int.

tensor_map

int:

class cupti.cupti.CuDeviceCanAccessPeerParams

Bases: object

Empty-initialize an instance of cuDeviceCanAccessPeer_params.

See also

cuDeviceCanAccessPeer_params

can_access_peer

int:

dev

int:

static from_buffer(buffer)

Create an CuDeviceCanAccessPeerParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceCanAccessPeerParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_can_access_peer_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceCanAccessPeerParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

peer_dev

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxEnablePeerAccessParams

Bases: object

Empty-initialize an instance of cuCtxEnablePeerAccess_params.

See also

cuCtxEnablePeerAccess_params

flags_

int:

static from_buffer(buffer)

Create an CuCtxEnablePeerAccessParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxEnablePeerAccessParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_enable_peer_access_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxEnablePeerAccessParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

peer_context

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxDisablePeerAccessParams

Bases: object

Empty-initialize an instance of cuCtxDisablePeerAccess_params.

See also

cuCtxDisablePeerAccess_params

static from_buffer(buffer)

Create an CuCtxDisablePeerAccessParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxDisablePeerAccessParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_disable_peer_access_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxDisablePeerAccessParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

peer_context

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetP2PAttributeParams

Bases: object

Empty-initialize an instance of cuDeviceGetP2PAttribute_params.

See also

cuDeviceGetP2PAttribute_params

attrib

int:

dst_device

int:

static from_buffer(buffer)

Create an CuDeviceGetP2PAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetP2PAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_p2p_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetP2PAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_device

int:

value

int:

class cupti.cupti.CuGraphicsUnregisterResourceParams

Bases: object

Empty-initialize an instance of cuGraphicsUnregisterResource_params.

See also

cuGraphicsUnregisterResource_params

static from_buffer(buffer)

Create an CuGraphicsUnregisterResourceParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphicsUnregisterResourceParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graphics_unregister_resource_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphicsUnregisterResourceParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

resource

int:

class cupti.cupti.CuGraphicsSubResourceGetMappedArrayParams

Bases: object

Empty-initialize an instance of cuGraphicsSubResourceGetMappedArray_params.

See also

cuGraphicsSubResourceGetMappedArray_params

array_index

int:

static from_buffer(buffer)

Create an CuGraphicsSubResourceGetMappedArrayParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphicsSubResourceGetMappedArrayParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graphics_sub_resource_get_mapped_array_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphicsSubResourceGetMappedArrayParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mip_level

int:

p_array

int:

ptr

Get the pointer address to the data as Python int.

resource

int:

class cupti.cupti.CuGraphicsResourceGetMappedMipmappedArrayParams

Bases: object

Empty-initialize an instance of cuGraphicsResourceGetMappedMipmappedArray_params.

See also

cuGraphicsResourceGetMappedMipmappedArray_params

static from_buffer(buffer)

Create an CuGraphicsResourceGetMappedMipmappedArrayParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphicsResourceGetMappedMipmappedArrayParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graphics_resource_get_mapped_mipmapped_array_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphicsResourceGetMappedMipmappedArrayParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_mipmapped_array

int:

ptr

Get the pointer address to the data as Python int.

resource

int:

class cupti.cupti.CuGraphicsResourceGetMappedPointerV2Params

Bases: object

Empty-initialize an instance of cuGraphicsResourceGetMappedPointer_v2_params.

See also

cuGraphicsResourceGetMappedPointer_v2_params

static from_buffer(buffer)

Create an CuGraphicsResourceGetMappedPointerV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphicsResourceGetMappedPointerV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graphics_resource_get_mapped_pointer_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphicsResourceGetMappedPointerV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_dev_ptr

int:

p_size

int:

ptr

Get the pointer address to the data as Python int.

resource

int:

class cupti.cupti.CuGraphicsResourceSetMapFlagsV2Params

Bases: object

Empty-initialize an instance of cuGraphicsResourceSetMapFlags_v2_params.

See also

cuGraphicsResourceSetMapFlags_v2_params

flags_

int:

static from_buffer(buffer)

Create an CuGraphicsResourceSetMapFlagsV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphicsResourceSetMapFlagsV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graphics_resource_set_map_flags_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphicsResourceSetMapFlagsV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

resource

int:

class cupti.cupti.CuGraphicsMapResourcesPtszParams

Bases: object

Empty-initialize an instance of cuGraphicsMapResources_ptsz_params.

See also

cuGraphicsMapResources_ptsz_params

count

int:

static from_buffer(buffer)

Create an CuGraphicsMapResourcesPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphicsMapResourcesPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graphics_map_resources_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphicsMapResourcesPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

resources

int:

class cupti.cupti.CuGraphicsUnmapResourcesPtszParams

Bases: object

Empty-initialize an instance of cuGraphicsUnmapResources_ptsz_params.

See also

cuGraphicsUnmapResources_ptsz_params

count

int:

static from_buffer(buffer)

Create an CuGraphicsUnmapResourcesPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphicsUnmapResourcesPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graphics_unmap_resources_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphicsUnmapResourcesPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

resources

int:

class cupti.cupti.CuGetProcAddressV2Params

Bases: object

Empty-initialize an instance of cuGetProcAddress_v2_params.

See also

cuGetProcAddress_v2_params

cuda_version

int:

flags_

int:

static from_buffer(buffer)

Create an CuGetProcAddressV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGetProcAddressV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_get_proc_address_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGetProcAddressV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pfn

int:

ptr

Get the pointer address to the data as Python int.

symbol

str:

symbol_status

int:

class cupti.cupti.CuCoredumpGetAttributeParams

Bases: object

Empty-initialize an instance of cuCoredumpGetAttribute_params.

See also

cuCoredumpGetAttribute_params

attrib

int:

static from_buffer(buffer)

Create an CuCoredumpGetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCoredumpGetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_coredump_get_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCoredumpGetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

size_

int:

value

int:

class cupti.cupti.CuCoredumpGetAttributeGlobalParams

Bases: object

Empty-initialize an instance of cuCoredumpGetAttributeGlobal_params.

See also

cuCoredumpGetAttributeGlobal_params

attrib

int:

static from_buffer(buffer)

Create an CuCoredumpGetAttributeGlobalParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCoredumpGetAttributeGlobalParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_coredump_get_attribute_global_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCoredumpGetAttributeGlobalParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

size_

int:

value

int:

class cupti.cupti.CuCoredumpSetAttributeParams

Bases: object

Empty-initialize an instance of cuCoredumpSetAttribute_params.

See also

cuCoredumpSetAttribute_params

attrib

int:

static from_buffer(buffer)

Create an CuCoredumpSetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCoredumpSetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_coredump_set_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCoredumpSetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

size_

int:

value

int:

class cupti.cupti.CuCoredumpSetAttributeGlobalParams

Bases: object

Empty-initialize an instance of cuCoredumpSetAttributeGlobal_params.

See also

cuCoredumpSetAttributeGlobal_params

attrib

int:

static from_buffer(buffer)

Create an CuCoredumpSetAttributeGlobalParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCoredumpSetAttributeGlobalParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_coredump_set_attribute_global_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCoredumpSetAttributeGlobalParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

size_

int:

value

int:

class cupti.cupti.CuGetExportTableParams

Bases: object

Empty-initialize an instance of cuGetExportTable_params.

See also

cuGetExportTable_params

static from_buffer(buffer)

Create an CuGetExportTableParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGetExportTableParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_get_export_table_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGetExportTableParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_export_table_id

int:

pp_export_table

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGreenCtxCreateParams

Bases: object

Empty-initialize an instance of cuGreenCtxCreate_params.

See also

cuGreenCtxCreate_params

desc

int:

dev

int:

flags_

int:

static from_buffer(buffer)

Create an CuGreenCtxCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGreenCtxCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_green_ctx_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGreenCtxCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ph_ctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGreenCtxDestroyParams

Bases: object

Empty-initialize an instance of cuGreenCtxDestroy_params.

See also

cuGreenCtxDestroy_params

static from_buffer(buffer)

Create an CuGreenCtxDestroyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGreenCtxDestroyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_green_ctx_destroy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGreenCtxDestroyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_ctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxFromGreenCtxParams

Bases: object

Empty-initialize an instance of cuCtxFromGreenCtx_params.

See also

cuCtxFromGreenCtx_params

static from_buffer(buffer)

Create an CuCtxFromGreenCtxParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxFromGreenCtxParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_from_green_ctx_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxFromGreenCtxParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_ctx

int:

p_context

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetDevResourceParams

Bases: object

Empty-initialize an instance of cuDeviceGetDevResource_params.

See also

cuDeviceGetDevResource_params

device_

int:

static from_buffer(buffer)

Create an CuDeviceGetDevResourceParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetDevResourceParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_dev_resource_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetDevResourceParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

resource

int:

type

int:

class cupti.cupti.CuCtxGetDevResourceParams

Bases: object

Empty-initialize an instance of cuCtxGetDevResource_params.

See also

cuCtxGetDevResource_params

static from_buffer(buffer)

Create an CuCtxGetDevResourceParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxGetDevResourceParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_get_dev_resource_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxGetDevResourceParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_ctx

int:

ptr

Get the pointer address to the data as Python int.

resource

int:

type

int:

class cupti.cupti.CuGreenCtxGetDevResourceParams

Bases: object

Empty-initialize an instance of cuGreenCtxGetDevResource_params.

See also

cuGreenCtxGetDevResource_params

static from_buffer(buffer)

Create an CuGreenCtxGetDevResourceParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGreenCtxGetDevResourceParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_green_ctx_get_dev_resource_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGreenCtxGetDevResourceParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_ctx

int:

ptr

Get the pointer address to the data as Python int.

resource

int:

type

int:

class cupti.cupti.CuDevSmResourceSplitByCountParams

Bases: object

Empty-initialize an instance of cuDevSmResourceSplitByCount_params.

See also

cuDevSmResourceSplitByCount_params

flags_

int:

static from_buffer(buffer)

Create an CuDevSmResourceSplitByCountParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDevSmResourceSplitByCountParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_dev_sm_resource_split_by_count_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDevSmResourceSplitByCountParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

input

int:

min_count

int:

nb_groups

int:

ptr

Get the pointer address to the data as Python int.

remainder

int:

result

int:

class cupti.cupti.CuDevResourceGenerateDescParams

Bases: object

Empty-initialize an instance of cuDevResourceGenerateDesc_params.

See also

cuDevResourceGenerateDesc_params

static from_buffer(buffer)

Create an CuDevResourceGenerateDescParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDevResourceGenerateDescParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_dev_resource_generate_desc_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDevResourceGenerateDescParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

nb_resources

int:

ph_desc

int:

ptr

Get the pointer address to the data as Python int.

resources

int:

class cupti.cupti.CuGreenCtxRecordEventParams

Bases: object

Empty-initialize an instance of cuGreenCtxRecordEvent_params.

See also

cuGreenCtxRecordEvent_params

static from_buffer(buffer)

Create an CuGreenCtxRecordEventParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGreenCtxRecordEventParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_green_ctx_record_event_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGreenCtxRecordEventParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_ctx

int:

h_event

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGreenCtxWaitEventParams

Bases: object

Empty-initialize an instance of cuGreenCtxWaitEvent_params.

See also

cuGreenCtxWaitEvent_params

static from_buffer(buffer)

Create an CuGreenCtxWaitEventParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGreenCtxWaitEventParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_green_ctx_wait_event_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGreenCtxWaitEventParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_ctx

int:

h_event

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetGreenCtxParams

Bases: object

Empty-initialize an instance of cuStreamGetGreenCtx_params.

See also

cuStreamGetGreenCtx_params

static from_buffer(buffer)

Create an CuStreamGetGreenCtxParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetGreenCtxParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_green_ctx_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetGreenCtxParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ph_ctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGreenCtxStreamCreateParams

Bases: object

Empty-initialize an instance of cuGreenCtxStreamCreate_params.

See also

cuGreenCtxStreamCreate_params

flags_

int:

static from_buffer(buffer)

Create an CuGreenCtxStreamCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGreenCtxStreamCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_green_ctx_stream_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGreenCtxStreamCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

green_ctx

int:

ph_stream

int:

priority

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemHostRegisterParams

Bases: object

Empty-initialize an instance of cuMemHostRegister_params.

See also

cuMemHostRegister_params

bytesize

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemHostRegisterParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemHostRegisterParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_host_register_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemHostRegisterParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphicsResourceSetMapFlagsParams

Bases: object

Empty-initialize an instance of cuGraphicsResourceSetMapFlags_params.

See also

cuGraphicsResourceSetMapFlags_params

flags_

int:

static from_buffer(buffer)

Create an CuGraphicsResourceSetMapFlagsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphicsResourceSetMapFlagsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graphics_resource_set_map_flags_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphicsResourceSetMapFlagsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

resource

int:

class cupti.cupti.CuLinkCreateParams

Bases: object

Empty-initialize an instance of cuLinkCreate_params.

See also

cuLinkCreate_params

static from_buffer(buffer)

Create an CuLinkCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLinkCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_link_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLinkCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_options

int:

option_values

int:

options

int:

ptr

Get the pointer address to the data as Python int.

state_out

int:

class cupti.cupti.CuLinkAddDataParams

Bases: object

Empty-initialize an instance of cuLinkAddData_params.

See also

cuLinkAddData_params

data_

int:

static from_buffer(buffer)

Create an CuLinkAddDataParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLinkAddDataParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_link_add_data_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLinkAddDataParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

str:

num_options

int:

option_values

int:

options

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

state

int:

type

int:

class cupti.cupti.CuLinkAddFileParams

Bases: object

Empty-initialize an instance of cuLinkAddFile_params.

See also

cuLinkAddFile_params

static from_buffer(buffer)

Create an CuLinkAddFileParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLinkAddFileParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_link_add_file_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLinkAddFileParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_options

int:

option_values

int:

options

int:

path

str:

ptr

Get the pointer address to the data as Python int.

state

int:

type

int:

class cupti.cupti.CuTexRefSetAddress2DV2Params

Bases: object

Empty-initialize an instance of cuTexRefSetAddress2D_v2_params.

See also

cuTexRefSetAddress2D_v2_params

desc

int:

dptr

int:

static from_buffer(buffer)

Create an CuTexRefSetAddress2DV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetAddress2DV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_address2d_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetAddress2DV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

pitch

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceTotalMemParams

Bases: object

Empty-initialize an instance of cuDeviceTotalMem_params.

See also

cuDeviceTotalMem_params

bytes

int:

dev

int:

static from_buffer(buffer)

Create an CuDeviceTotalMemParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceTotalMemParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_total_mem_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceTotalMemParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxCreateParams

Bases: object

Empty-initialize an instance of cuCtxCreate_params.

See also

cuCtxCreate_params

dev

int:

flags_

int:

static from_buffer(buffer)

Create an CuCtxCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuModuleGetGlobalParams

Bases: object

Empty-initialize an instance of cuModuleGetGlobal_params.

See also

cuModuleGetGlobal_params

bytes

int:

dptr

int:

static from_buffer(buffer)

Create an CuModuleGetGlobalParams instance with the memory from the given buffer.

static from_data(data)

Create an CuModuleGetGlobalParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_module_get_global_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuModuleGetGlobalParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hmod

int:

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemGetInfoParams

Bases: object

Empty-initialize an instance of cuMemGetInfo_params.

See also

cuMemGetInfo_params

free

int:

static from_buffer(buffer)

Create an CuMemGetInfoParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemGetInfoParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_get_info_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemGetInfoParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

total

int:

class cupti.cupti.CuMemAllocParams

Bases: object

Empty-initialize an instance of cuMemAlloc_params.

See also

cuMemAlloc_params

bytesize

int:

dptr

int:

static from_buffer(buffer)

Create an CuMemAllocParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemAllocParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_alloc_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemAllocParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemAllocPitchParams

Bases: object

Empty-initialize an instance of cuMemAllocPitch_params.

See also

cuMemAllocPitch_params

dptr

int:

element_size_bytes

int:

static from_buffer(buffer)

Create an CuMemAllocPitchParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemAllocPitchParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_alloc_pitch_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemAllocPitchParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

p_pitch

int:

ptr

Get the pointer address to the data as Python int.

width_in_bytes

int:

class cupti.cupti.CuMemFreeParams

Bases: object

Empty-initialize an instance of cuMemFree_params.

See also

cuMemFree_params

dptr

int:

static from_buffer(buffer)

Create an CuMemFreeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemFreeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_free_params_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an CuMemFreeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemGetAddressRangeParams

Bases: object

Empty-initialize an instance of cuMemGetAddressRange_params.

See also

cuMemGetAddressRange_params

dptr

int:

static from_buffer(buffer)

Create an CuMemGetAddressRangeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemGetAddressRangeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_get_address_range_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemGetAddressRangeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pbase

int:

psize

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemAllocHostParams

Bases: object

Empty-initialize an instance of cuMemAllocHost_params.

See also

cuMemAllocHost_params

bytesize

int:

static from_buffer(buffer)

Create an CuMemAllocHostParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemAllocHostParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_alloc_host_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemAllocHostParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pp

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemHostGetDevicePointerParams

Bases: object

Empty-initialize an instance of cuMemHostGetDevicePointer_params.

See also

cuMemHostGetDevicePointer_params

flags_

int:

static from_buffer(buffer)

Create an CuMemHostGetDevicePointerParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemHostGetDevicePointerParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_host_get_device_pointer_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemHostGetDevicePointerParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p

int:

pdptr

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpyHtoDParams

Bases: object

Empty-initialize an instance of cuMemcpyHtoD_params.

See also

cuMemcpyHtoD_params

byte_count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyHtoDParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyHtoDParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_hto_d_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyHtoDParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_host

int:

class cupti.cupti.CuMemcpyDtoHParams

Bases: object

Empty-initialize an instance of cuMemcpyDtoH_params.

See also

cuMemcpyDtoH_params

byte_count

int:

dst_host

int:

static from_buffer(buffer)

Create an CuMemcpyDtoHParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyDtoHParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_dto_h_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyDtoHParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuMemcpyDtoDParams

Bases: object

Empty-initialize an instance of cuMemcpyDtoD_params.

See also

cuMemcpyDtoD_params

byte_count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyDtoDParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyDtoDParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_dto_d_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyDtoDParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuMemcpyDtoAParams

Bases: object

Empty-initialize an instance of cuMemcpyDtoA_params.

See also

cuMemcpyDtoA_params

byte_count

int:

dst_array

int:

dst_offset

int:

static from_buffer(buffer)

Create an CuMemcpyDtoAParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyDtoAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_dto_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyDtoAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuMemcpyAtoDParams

Bases: object

Empty-initialize an instance of cuMemcpyAtoD_params.

See also

cuMemcpyAtoD_params

byte_count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyAtoDParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyAtoDParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_ato_d_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyAtoDParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_array

int:

src_offset

int:

class cupti.cupti.CuMemcpyHtoAParams

Bases: object

Empty-initialize an instance of cuMemcpyHtoA_params.

See also

cuMemcpyHtoA_params

byte_count

int:

dst_array

int:

dst_offset

int:

static from_buffer(buffer)

Create an CuMemcpyHtoAParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyHtoAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_hto_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyHtoAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_host

int:

class cupti.cupti.CuMemcpyAtoHParams

Bases: object

Empty-initialize an instance of cuMemcpyAtoH_params.

See also

cuMemcpyAtoH_params

byte_count

int:

dst_host

int:

static from_buffer(buffer)

Create an CuMemcpyAtoHParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyAtoHParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_ato_h_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyAtoHParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_array

int:

src_offset

int:

class cupti.cupti.CuMemcpyAtoAParams

Bases: object

Empty-initialize an instance of cuMemcpyAtoA_params.

See also

cuMemcpyAtoA_params

byte_count

int:

dst_array

int:

dst_offset

int:

static from_buffer(buffer)

Create an CuMemcpyAtoAParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyAtoAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_ato_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyAtoAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_array

int:

src_offset

int:

class cupti.cupti.CuMemcpyHtoAAsyncParams

Bases: object

Empty-initialize an instance of cuMemcpyHtoAAsync_params.

See also

cuMemcpyHtoAAsync_params

byte_count

int:

dst_array

int:

dst_offset

int:

static from_buffer(buffer)

Create an CuMemcpyHtoAAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyHtoAAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_hto_aasync_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyHtoAAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_host

int:

class cupti.cupti.CuMemcpyAtoHAsyncParams

Bases: object

Empty-initialize an instance of cuMemcpyAtoHAsync_params.

See also

cuMemcpyAtoHAsync_params

byte_count

int:

dst_host

int:

static from_buffer(buffer)

Create an CuMemcpyAtoHAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyAtoHAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_ato_hasync_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyAtoHAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_array

int:

src_offset

int:

class cupti.cupti.CuMemcpy2DParams

Bases: object

Empty-initialize an instance of cuMemcpy2D_params.

See also

cuMemcpy2D_params

static from_buffer(buffer)

Create an CuMemcpy2DParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy2DParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy2d_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy2DParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpy2DUnalignedParams

Bases: object

Empty-initialize an instance of cuMemcpy2DUnaligned_params.

See also

cuMemcpy2DUnaligned_params

static from_buffer(buffer)

Create an CuMemcpy2DUnalignedParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy2DUnalignedParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy2d_unaligned_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy2DUnalignedParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpy3DParams

Bases: object

Empty-initialize an instance of cuMemcpy3D_params.

See also

cuMemcpy3D_params

static from_buffer(buffer)

Create an CuMemcpy3DParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3d_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpyHtoDAsyncParams

Bases: object

Empty-initialize an instance of cuMemcpyHtoDAsync_params.

See also

cuMemcpyHtoDAsync_params

byte_count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyHtoDAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyHtoDAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_hto_dasync_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyHtoDAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_host

int:

class cupti.cupti.CuMemcpyDtoHAsyncParams

Bases: object

Empty-initialize an instance of cuMemcpyDtoHAsync_params.

See also

cuMemcpyDtoHAsync_params

byte_count

int:

dst_host

int:

static from_buffer(buffer)

Create an CuMemcpyDtoHAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyDtoHAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_dto_hasync_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyDtoHAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuMemcpyDtoDAsyncParams

Bases: object

Empty-initialize an instance of cuMemcpyDtoDAsync_params.

See also

cuMemcpyDtoDAsync_params

byte_count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyDtoDAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyDtoDAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_dto_dasync_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyDtoDAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuMemcpy2DAsyncParams

Bases: object

Empty-initialize an instance of cuMemcpy2DAsync_params.

See also

cuMemcpy2DAsync_params

static from_buffer(buffer)

Create an CuMemcpy2DAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy2DAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy2_dasync_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy2DAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpy3DAsyncParams

Bases: object

Empty-initialize an instance of cuMemcpy3DAsync_params.

See also

cuMemcpy3DAsync_params

static from_buffer(buffer)

Create an CuMemcpy3DAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3_dasync_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemsetD8Params

Bases: object

Empty-initialize an instance of cuMemsetD8_params.

See also

cuMemsetD8_params

dst_device

int:

static from_buffer(buffer)

Create an CuMemsetD8Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD8Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d8params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD8Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

n

int:

ptr

Get the pointer address to the data as Python int.

uc

int:

class cupti.cupti.CuMemsetD16Params

Bases: object

Empty-initialize an instance of cuMemsetD16_params.

See also

cuMemsetD16_params

dst_device

int:

static from_buffer(buffer)

Create an CuMemsetD16Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD16Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d16params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD16Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

n

int:

ptr

Get the pointer address to the data as Python int.

us

int:

class cupti.cupti.CuMemsetD32Params

Bases: object

Empty-initialize an instance of cuMemsetD32_params.

See also

cuMemsetD32_params

dst_device

int:

static from_buffer(buffer)

Create an CuMemsetD32Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD32Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d32params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD32Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

n

int:

ptr

Get the pointer address to the data as Python int.

ui

int:

class cupti.cupti.CuMemsetD2D8Params

Bases: object

Empty-initialize an instance of cuMemsetD2D8_params.

See also

cuMemsetD2D8_params

dst_device

int:

dst_pitch

int:

static from_buffer(buffer)

Create an CuMemsetD2D8Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD2D8Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d2d8params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD2D8Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

ptr

Get the pointer address to the data as Python int.

uc

int:

width

int:

class cupti.cupti.CuMemsetD2D16Params

Bases: object

Empty-initialize an instance of cuMemsetD2D16_params.

See also

cuMemsetD2D16_params

dst_device

int:

dst_pitch

int:

static from_buffer(buffer)

Create an CuMemsetD2D16Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD2D16Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d2d16params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD2D16Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

ptr

Get the pointer address to the data as Python int.

us

int:

width

int:

class cupti.cupti.CuMemsetD2D32Params

Bases: object

Empty-initialize an instance of cuMemsetD2D32_params.

See also

cuMemsetD2D32_params

dst_device

int:

dst_pitch

int:

static from_buffer(buffer)

Create an CuMemsetD2D32Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD2D32Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d2d32params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD2D32Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

ptr

Get the pointer address to the data as Python int.

ui

int:

width

int:

class cupti.cupti.CuArrayCreateParams

Bases: object

Empty-initialize an instance of cuArrayCreate_params.

See also

cuArrayCreate_params

static from_buffer(buffer)

Create an CuArrayCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuArrayCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_array_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuArrayCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_allocate_array

int:

p_handle

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuArrayGetDescriptorParams

Bases: object

Empty-initialize an instance of cuArrayGetDescriptor_params.

See also

cuArrayGetDescriptor_params

static from_buffer(buffer)

Create an CuArrayGetDescriptorParams instance with the memory from the given buffer.

static from_data(data)

Create an CuArrayGetDescriptorParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_array_get_descriptor_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuArrayGetDescriptorParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_array

int:

p_array_descriptor

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuArray3DCreateParams

Bases: object

Empty-initialize an instance of cuArray3DCreate_params.

See also

cuArray3DCreate_params

static from_buffer(buffer)

Create an CuArray3DCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuArray3DCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_array3d_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuArray3DCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_allocate_array

int:

p_handle

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuArray3DGetDescriptorParams

Bases: object

Empty-initialize an instance of cuArray3DGetDescriptor_params.

See also

cuArray3DGetDescriptor_params

static from_buffer(buffer)

Create an CuArray3DGetDescriptorParams instance with the memory from the given buffer.

static from_data(data)

Create an CuArray3DGetDescriptorParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_array3d_get_descriptor_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuArray3DGetDescriptorParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_array

int:

p_array_descriptor

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefSetAddressParams

Bases: object

Empty-initialize an instance of cuTexRefSetAddress_params.

See also

cuTexRefSetAddress_params

byte_offset

int:

bytes

int:

dptr

int:

static from_buffer(buffer)

Create an CuTexRefSetAddressParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetAddressParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_address_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetAddressParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefSetAddress2DParams

Bases: object

Empty-initialize an instance of cuTexRefSetAddress2D_params.

See also

cuTexRefSetAddress2D_params

desc

int:

dptr

int:

static from_buffer(buffer)

Create an CuTexRefSetAddress2DParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefSetAddress2DParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_set_address2d_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefSetAddress2DParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

pitch

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuTexRefGetAddressParams

Bases: object

Empty-initialize an instance of cuTexRefGetAddress_params.

See also

cuTexRefGetAddress_params

static from_buffer(buffer)

Create an CuTexRefGetAddressParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTexRefGetAddressParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tex_ref_get_address_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTexRefGetAddressParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_tex_ref

int:

pdptr

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphicsResourceGetMappedPointerParams

Bases: object

Empty-initialize an instance of cuGraphicsResourceGetMappedPointer_params.

See also

cuGraphicsResourceGetMappedPointer_params

static from_buffer(buffer)

Create an CuGraphicsResourceGetMappedPointerParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphicsResourceGetMappedPointerParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graphics_resource_get_mapped_pointer_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphicsResourceGetMappedPointerParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_dev_ptr

int:

p_size

int:

ptr

Get the pointer address to the data as Python int.

resource

int:

class cupti.cupti.CuCtxDestroyParams

Bases: object

Empty-initialize an instance of cuCtxDestroy_params.

See also

cuCtxDestroy_params

ctx

int:

static from_buffer(buffer)

Create an CuCtxDestroyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxDestroyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_destroy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxDestroyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxPopCurrentParams

Bases: object

Empty-initialize an instance of cuCtxPopCurrent_params.

See also

cuCtxPopCurrent_params

static from_buffer(buffer)

Create an CuCtxPopCurrentParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxPopCurrentParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_pop_current_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxPopCurrentParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxPushCurrentParams

Bases: object

Empty-initialize an instance of cuCtxPushCurrent_params.

See also

cuCtxPushCurrent_params

ctx

int:

static from_buffer(buffer)

Create an CuCtxPushCurrentParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxPushCurrentParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_push_current_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxPushCurrentParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamDestroyParams

Bases: object

Empty-initialize an instance of cuStreamDestroy_params.

See also

cuStreamDestroy_params

static from_buffer(buffer)

Create an CuStreamDestroyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamDestroyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_destroy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamDestroyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuEventDestroyParams

Bases: object

Empty-initialize an instance of cuEventDestroy_params.

See also

cuEventDestroy_params

static from_buffer(buffer)

Create an CuEventDestroyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuEventDestroyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_event_destroy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuEventDestroyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_event

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDevicePrimaryCtxReleaseParams

Bases: object

Empty-initialize an instance of cuDevicePrimaryCtxRelease_params.

See also

cuDevicePrimaryCtxRelease_params

dev

int:

static from_buffer(buffer)

Create an CuDevicePrimaryCtxReleaseParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDevicePrimaryCtxReleaseParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_primary_ctx_release_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDevicePrimaryCtxReleaseParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDevicePrimaryCtxResetParams

Bases: object

Empty-initialize an instance of cuDevicePrimaryCtxReset_params.

See also

cuDevicePrimaryCtxReset_params

dev

int:

static from_buffer(buffer)

Create an CuDevicePrimaryCtxResetParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDevicePrimaryCtxResetParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_primary_ctx_reset_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDevicePrimaryCtxResetParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDevicePrimaryCtxSetFlagsParams

Bases: object

Empty-initialize an instance of cuDevicePrimaryCtxSetFlags_params.

See also

cuDevicePrimaryCtxSetFlags_params

dev

int:

flags_

int:

static from_buffer(buffer)

Create an CuDevicePrimaryCtxSetFlagsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDevicePrimaryCtxSetFlagsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_primary_ctx_set_flags_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDevicePrimaryCtxSetFlagsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpyHtoDV2Params

Bases: object

Empty-initialize an instance of cuMemcpyHtoD_v2_params.

See also

cuMemcpyHtoD_v2_params

byte_count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyHtoDV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyHtoDV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_hto_d_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyHtoDV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_host

int:

class cupti.cupti.CuMemcpyDtoHV2Params

Bases: object

Empty-initialize an instance of cuMemcpyDtoH_v2_params.

See also

cuMemcpyDtoH_v2_params

byte_count

int:

dst_host

int:

static from_buffer(buffer)

Create an CuMemcpyDtoHV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyDtoHV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_dto_h_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyDtoHV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuMemcpyDtoDV2Params

Bases: object

Empty-initialize an instance of cuMemcpyDtoD_v2_params.

See also

cuMemcpyDtoD_v2_params

byte_count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyDtoDV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyDtoDV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_dto_d_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyDtoDV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuMemcpyDtoAV2Params

Bases: object

Empty-initialize an instance of cuMemcpyDtoA_v2_params.

See also

cuMemcpyDtoA_v2_params

byte_count

int:

dst_array

int:

dst_offset

int:

static from_buffer(buffer)

Create an CuMemcpyDtoAV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyDtoAV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_dto_a_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyDtoAV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuMemcpyAtoDV2Params

Bases: object

Empty-initialize an instance of cuMemcpyAtoD_v2_params.

See also

cuMemcpyAtoD_v2_params

byte_count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyAtoDV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyAtoDV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_ato_d_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyAtoDV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_array

int:

src_offset

int:

class cupti.cupti.CuMemcpyHtoAV2Params

Bases: object

Empty-initialize an instance of cuMemcpyHtoA_v2_params.

See also

cuMemcpyHtoA_v2_params

byte_count

int:

dst_array

int:

dst_offset

int:

static from_buffer(buffer)

Create an CuMemcpyHtoAV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyHtoAV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_hto_a_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyHtoAV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_host

int:

class cupti.cupti.CuMemcpyAtoHV2Params

Bases: object

Empty-initialize an instance of cuMemcpyAtoH_v2_params.

See also

cuMemcpyAtoH_v2_params

byte_count

int:

dst_host

int:

static from_buffer(buffer)

Create an CuMemcpyAtoHV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyAtoHV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_ato_h_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyAtoHV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_array

int:

src_offset

int:

class cupti.cupti.CuMemcpyAtoAV2Params

Bases: object

Empty-initialize an instance of cuMemcpyAtoA_v2_params.

See also

cuMemcpyAtoA_v2_params

byte_count

int:

dst_array

int:

dst_offset

int:

static from_buffer(buffer)

Create an CuMemcpyAtoAV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyAtoAV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_ato_a_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyAtoAV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_array

int:

src_offset

int:

class cupti.cupti.CuMemcpyHtoAAsyncV2Params

Bases: object

Empty-initialize an instance of cuMemcpyHtoAAsync_v2_params.

See also

cuMemcpyHtoAAsync_v2_params

byte_count

int:

dst_array

int:

dst_offset

int:

static from_buffer(buffer)

Create an CuMemcpyHtoAAsyncV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyHtoAAsyncV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_hto_aasync_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyHtoAAsyncV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_host

int:

class cupti.cupti.CuMemcpyAtoHAsyncV2Params

Bases: object

Empty-initialize an instance of cuMemcpyAtoHAsync_v2_params.

See also

cuMemcpyAtoHAsync_v2_params

byte_count

int:

dst_host

int:

static from_buffer(buffer)

Create an CuMemcpyAtoHAsyncV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyAtoHAsyncV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_ato_hasync_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyAtoHAsyncV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_array

int:

src_offset

int:

class cupti.cupti.CuMemcpy2DV2Params

Bases: object

Empty-initialize an instance of cuMemcpy2D_v2_params.

See also

cuMemcpy2D_v2_params

static from_buffer(buffer)

Create an CuMemcpy2DV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy2DV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy2d_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy2DV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpy2DUnalignedV2Params

Bases: object

Empty-initialize an instance of cuMemcpy2DUnaligned_v2_params.

See also

cuMemcpy2DUnaligned_v2_params

static from_buffer(buffer)

Create an CuMemcpy2DUnalignedV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy2DUnalignedV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy2d_unaligned_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy2DUnalignedV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpy3DV2Params

Bases: object

Empty-initialize an instance of cuMemcpy3D_v2_params.

See also

cuMemcpy3D_v2_params

static from_buffer(buffer)

Create an CuMemcpy3DV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3d_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpyHtoDAsyncV2Params

Bases: object

Empty-initialize an instance of cuMemcpyHtoDAsync_v2_params.

See also

cuMemcpyHtoDAsync_v2_params

byte_count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyHtoDAsyncV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyHtoDAsyncV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_hto_dasync_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyHtoDAsyncV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_host

int:

class cupti.cupti.CuMemcpyDtoHAsyncV2Params

Bases: object

Empty-initialize an instance of cuMemcpyDtoHAsync_v2_params.

See also

cuMemcpyDtoHAsync_v2_params

byte_count

int:

dst_host

int:

static from_buffer(buffer)

Create an CuMemcpyDtoHAsyncV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyDtoHAsyncV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_dto_hasync_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyDtoHAsyncV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuMemcpyDtoDAsyncV2Params

Bases: object

Empty-initialize an instance of cuMemcpyDtoDAsync_v2_params.

See also

cuMemcpyDtoDAsync_v2_params

byte_count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyDtoDAsyncV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyDtoDAsyncV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_dto_dasync_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyDtoDAsyncV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuMemcpy2DAsyncV2Params

Bases: object

Empty-initialize an instance of cuMemcpy2DAsync_v2_params.

See also

cuMemcpy2DAsync_v2_params

static from_buffer(buffer)

Create an CuMemcpy2DAsyncV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy2DAsyncV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy2_dasync_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy2DAsyncV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpy3DAsyncV2Params

Bases: object

Empty-initialize an instance of cuMemcpy3DAsync_v2_params.

See also

cuMemcpy3DAsync_v2_params

static from_buffer(buffer)

Create an CuMemcpy3DAsyncV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DAsyncV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3_dasync_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DAsyncV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemsetD8V2Params

Bases: object

Empty-initialize an instance of cuMemsetD8_v2_params.

See also

cuMemsetD8_v2_params

dst_device

int:

static from_buffer(buffer)

Create an CuMemsetD8V2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD8V2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d8v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD8V2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

n

int:

ptr

Get the pointer address to the data as Python int.

uc

int:

class cupti.cupti.CuMemsetD16V2Params

Bases: object

Empty-initialize an instance of cuMemsetD16_v2_params.

See also

cuMemsetD16_v2_params

dst_device

int:

static from_buffer(buffer)

Create an CuMemsetD16V2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD16V2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d16v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD16V2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

n

int:

ptr

Get the pointer address to the data as Python int.

us

int:

class cupti.cupti.CuMemsetD32V2Params

Bases: object

Empty-initialize an instance of cuMemsetD32_v2_params.

See also

cuMemsetD32_v2_params

dst_device

int:

static from_buffer(buffer)

Create an CuMemsetD32V2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD32V2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d32v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD32V2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

n

int:

ptr

Get the pointer address to the data as Python int.

ui

int:

class cupti.cupti.CuMemsetD2D8V2Params

Bases: object

Empty-initialize an instance of cuMemsetD2D8_v2_params.

See also

cuMemsetD2D8_v2_params

dst_device

int:

dst_pitch

int:

static from_buffer(buffer)

Create an CuMemsetD2D8V2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD2D8V2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d2d8v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD2D8V2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

ptr

Get the pointer address to the data as Python int.

uc

int:

width

int:

class cupti.cupti.CuMemsetD2D16V2Params

Bases: object

Empty-initialize an instance of cuMemsetD2D16_v2_params.

See also

cuMemsetD2D16_v2_params

dst_device

int:

dst_pitch

int:

static from_buffer(buffer)

Create an CuMemsetD2D16V2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD2D16V2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d2d16v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD2D16V2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

ptr

Get the pointer address to the data as Python int.

us

int:

width

int:

class cupti.cupti.CuMemsetD2D32V2Params

Bases: object

Empty-initialize an instance of cuMemsetD2D32_v2_params.

See also

cuMemsetD2D32_v2_params

dst_device

int:

dst_pitch

int:

static from_buffer(buffer)

Create an CuMemsetD2D32V2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD2D32V2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d2d32v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD2D32V2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

height

int:

ptr

Get the pointer address to the data as Python int.

ui

int:

width

int:

class cupti.cupti.CuMemcpyParams

Bases: object

Empty-initialize an instance of cuMemcpy_params.

See also

cuMemcpy_params

byte_count

int:

dst

int:

static from_buffer(buffer)

Create an CuMemcpyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_params_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an CuMemcpyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src

int:

class cupti.cupti.CuMemcpyAsyncParams

Bases: object

Empty-initialize an instance of cuMemcpyAsync_params.

See also

cuMemcpyAsync_params

byte_count

int:

dst

int:

static from_buffer(buffer)

Create an CuMemcpyAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src

int:

class cupti.cupti.CuMemcpyPeerParams

Bases: object

Empty-initialize an instance of cuMemcpyPeer_params.

See also

cuMemcpyPeer_params

byte_count

int:

dst_context

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyPeerParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyPeerParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_peer_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyPeerParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_context

int:

src_device

int:

class cupti.cupti.CuMemcpyPeerAsyncParams

Bases: object

Empty-initialize an instance of cuMemcpyPeerAsync_params.

See also

cuMemcpyPeerAsync_params

byte_count

int:

dst_context

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemcpyPeerAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyPeerAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_peer_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyPeerAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

src_context

int:

src_device

int:

class cupti.cupti.CuMemcpy3DPeerParams

Bases: object

Empty-initialize an instance of cuMemcpy3DPeer_params.

See also

cuMemcpy3DPeer_params

static from_buffer(buffer)

Create an CuMemcpy3DPeerParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DPeerParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3d_peer_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DPeerParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpy3DPeerAsyncParams

Bases: object

Empty-initialize an instance of cuMemcpy3DPeerAsync_params.

See also

cuMemcpy3DPeerAsync_params

static from_buffer(buffer)

Create an CuMemcpy3DPeerAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DPeerAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3d_peer_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DPeerAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

p_copy

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpyBatchAsyncParams

Bases: object

Empty-initialize an instance of cuMemcpyBatchAsync_params.

See also

cuMemcpyBatchAsync_params

attrs

int:

attrs_idxs

int:

count

int:

dsts

int:

fail_idx

int:

static from_buffer(buffer)

Create an CuMemcpyBatchAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyBatchAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_batch_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyBatchAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_attrs

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

srcs

int:

class cupti.cupti.CuMemcpy3DBatchAsyncParams

Bases: object

Empty-initialize an instance of cuMemcpy3DBatchAsync_params.

See also

cuMemcpy3DBatchAsync_params

fail_idx

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemcpy3DBatchAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DBatchAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3d_batch_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DBatchAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_ops

int:

op_list

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemsetD8AsyncParams

Bases: object

Empty-initialize an instance of cuMemsetD8Async_params.

See also

cuMemsetD8Async_params

dst_device

int:

static from_buffer(buffer)

Create an CuMemsetD8AsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD8AsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d8_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD8AsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

n

int:

ptr

Get the pointer address to the data as Python int.

uc

int:

class cupti.cupti.CuMemsetD16AsyncParams

Bases: object

Empty-initialize an instance of cuMemsetD16Async_params.

See also

cuMemsetD16Async_params

dst_device

int:

static from_buffer(buffer)

Create an CuMemsetD16AsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD16AsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d16_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD16AsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

n

int:

ptr

Get the pointer address to the data as Python int.

us

int:

class cupti.cupti.CuMemsetD32AsyncParams

Bases: object

Empty-initialize an instance of cuMemsetD32Async_params.

See also

cuMemsetD32Async_params

dst_device

int:

static from_buffer(buffer)

Create an CuMemsetD32AsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD32AsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d32_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD32AsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

n

int:

ptr

Get the pointer address to the data as Python int.

ui

int:

class cupti.cupti.CuMemsetD2D8AsyncParams

Bases: object

Empty-initialize an instance of cuMemsetD2D8Async_params.

See also

cuMemsetD2D8Async_params

dst_device

int:

dst_pitch

int:

static from_buffer(buffer)

Create an CuMemsetD2D8AsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD2D8AsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d2d8_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD2D8AsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

height

int:

ptr

Get the pointer address to the data as Python int.

uc

int:

width

int:

class cupti.cupti.CuMemsetD2D16AsyncParams

Bases: object

Empty-initialize an instance of cuMemsetD2D16Async_params.

See also

cuMemsetD2D16Async_params

dst_device

int:

dst_pitch

int:

static from_buffer(buffer)

Create an CuMemsetD2D16AsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD2D16AsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d2d16_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD2D16AsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

height

int:

ptr

Get the pointer address to the data as Python int.

us

int:

width

int:

class cupti.cupti.CuMemsetD2D32AsyncParams

Bases: object

Empty-initialize an instance of cuMemsetD2D32Async_params.

See also

cuMemsetD2D32Async_params

dst_device

int:

dst_pitch

int:

static from_buffer(buffer)

Create an CuMemsetD2D32AsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemsetD2D32AsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memset_d2d32_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemsetD2D32AsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

height

int:

ptr

Get the pointer address to the data as Python int.

ui

int:

width

int:

class cupti.cupti.CuStreamGetPriorityParams

Bases: object

Empty-initialize an instance of cuStreamGetPriority_params.

See also

cuStreamGetPriority_params

static from_buffer(buffer)

Create an CuStreamGetPriorityParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetPriorityParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_priority_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetPriorityParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

priority

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetIdParams

Bases: object

Empty-initialize an instance of cuStreamGetId_params.

See also

cuStreamGetId_params

static from_buffer(buffer)

Create an CuStreamGetIdParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetIdParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_id_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetIdParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

stream_id

int:

class cupti.cupti.CuStreamGetFlagsParams

Bases: object

Empty-initialize an instance of cuStreamGetFlags_params.

See also

cuStreamGetFlags_params

flags_

int:

static from_buffer(buffer)

Create an CuStreamGetFlagsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetFlagsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_flags_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetFlagsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetDeviceParams

Bases: object

Empty-initialize an instance of cuStreamGetDevice_params.

See also

cuStreamGetDevice_params

device_

int:

static from_buffer(buffer)

Create an CuStreamGetDeviceParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetDeviceParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_device_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetDeviceParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetCtxParams

Bases: object

Empty-initialize an instance of cuStreamGetCtx_params.

See also

cuStreamGetCtx_params

static from_buffer(buffer)

Create an CuStreamGetCtxParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetCtxParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_ctx_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetCtxParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

pctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetCtxV2Params

Bases: object

Empty-initialize an instance of cuStreamGetCtx_v2_params.

See also

cuStreamGetCtx_v2_params

static from_buffer(buffer)

Create an CuStreamGetCtxV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetCtxV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_ctx_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetCtxV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

p_ctx

int:

p_green_ctx

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamWaitEventParams

Bases: object

Empty-initialize an instance of cuStreamWaitEvent_params.

See also

cuStreamWaitEvent_params

flags_

int:

static from_buffer(buffer)

Create an CuStreamWaitEventParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWaitEventParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_wait_event_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWaitEventParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_event

int:

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamAddCallbackParams

Bases: object

Empty-initialize an instance of cuStreamAddCallback_params.

See also

cuStreamAddCallback_params

callback

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamAddCallbackParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamAddCallbackParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_add_callback_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamAddCallbackParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

user_data

int:

class cupti.cupti.CuStreamAttachMemAsyncParams

Bases: object

Empty-initialize an instance of cuStreamAttachMemAsync_params.

See also

cuStreamAttachMemAsync_params

dptr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamAttachMemAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamAttachMemAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_attach_mem_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamAttachMemAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

length

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamQueryParams

Bases: object

Empty-initialize an instance of cuStreamQuery_params.

See also

cuStreamQuery_params

static from_buffer(buffer)

Create an CuStreamQueryParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamQueryParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_query_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamQueryParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamSynchronizeParams

Bases: object

Empty-initialize an instance of cuStreamSynchronize_params.

See also

cuStreamSynchronize_params

static from_buffer(buffer)

Create an CuStreamSynchronizeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamSynchronizeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_synchronize_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamSynchronizeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuEventRecordParams

Bases: object

Empty-initialize an instance of cuEventRecord_params.

See also

cuEventRecord_params

static from_buffer(buffer)

Create an CuEventRecordParams instance with the memory from the given buffer.

static from_data(data)

Create an CuEventRecordParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_event_record_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuEventRecordParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_event

int:

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuEventRecordWithFlagsParams

Bases: object

Empty-initialize an instance of cuEventRecordWithFlags_params.

See also

cuEventRecordWithFlags_params

flags_

int:

static from_buffer(buffer)

Create an CuEventRecordWithFlagsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuEventRecordWithFlagsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_event_record_with_flags_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuEventRecordWithFlagsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_event

int:

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLaunchKernelParams

Bases: object

Empty-initialize an instance of cuLaunchKernel_params.

See also

cuLaunchKernel_params

block_dim_x

int:

block_dim_y

int:

block_dim_z

int:

extra

int:

f

int:

static from_buffer(buffer)

Create an CuLaunchKernelParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLaunchKernelParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_launch_kernel_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLaunchKernelParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

grid_dim_x

int:

grid_dim_y

int:

grid_dim_z

int:

h_stream

int:

kernel_params

int:

ptr

Get the pointer address to the data as Python int.

shared_mem_bytes

int:

class cupti.cupti.CuLaunchKernelExParams

Bases: object

Empty-initialize an instance of cuLaunchKernelEx_params.

See also

cuLaunchKernelEx_params

config

int:

extra

int:

f

int:

static from_buffer(buffer)

Create an CuLaunchKernelExParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLaunchKernelExParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_launch_kernel_ex_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLaunchKernelExParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kernel_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLaunchHostFuncParams

Bases: object

Empty-initialize an instance of cuLaunchHostFunc_params.

See also

cuLaunchHostFunc_params

fn

int:

static from_buffer(buffer)

Create an CuLaunchHostFuncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLaunchHostFuncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_launch_host_func_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLaunchHostFuncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

user_data

int:

class cupti.cupti.CuGraphicsMapResourcesParams

Bases: object

Empty-initialize an instance of cuGraphicsMapResources_params.

See also

cuGraphicsMapResources_params

count

int:

static from_buffer(buffer)

Create an CuGraphicsMapResourcesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphicsMapResourcesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graphics_map_resources_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphicsMapResourcesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

resources

int:

class cupti.cupti.CuGraphicsUnmapResourcesParams

Bases: object

Empty-initialize an instance of cuGraphicsUnmapResources_params.

See also

cuGraphicsUnmapResources_params

count

int:

static from_buffer(buffer)

Create an CuGraphicsUnmapResourcesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphicsUnmapResourcesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graphics_unmap_resources_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphicsUnmapResourcesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

resources

int:

class cupti.cupti.CuStreamWriteValue32Params

Bases: object

Empty-initialize an instance of cuStreamWriteValue32_params.

See also

cuStreamWriteValue32_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWriteValue32Params instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWriteValue32Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_write_value32params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWriteValue32Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamWaitValue32Params

Bases: object

Empty-initialize an instance of cuStreamWaitValue32_params.

See also

cuStreamWaitValue32_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWaitValue32Params instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWaitValue32Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_wait_value32params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWaitValue32Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamWriteValue64Params

Bases: object

Empty-initialize an instance of cuStreamWriteValue64_params.

See also

cuStreamWriteValue64_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWriteValue64Params instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWriteValue64Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_write_value64params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWriteValue64Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamWaitValue64Params

Bases: object

Empty-initialize an instance of cuStreamWaitValue64_params.

See also

cuStreamWaitValue64_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWaitValue64Params instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWaitValue64Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_wait_value64params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWaitValue64Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamBatchMemOpParams

Bases: object

Empty-initialize an instance of cuStreamBatchMemOp_params.

See also

cuStreamBatchMemOp_params

count

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamBatchMemOpParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamBatchMemOpParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_batch_mem_op_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamBatchMemOpParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

param_array

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CuStreamWriteValue32PtszParams

Bases: object

Empty-initialize an instance of cuStreamWriteValue32_ptsz_params.

See also

cuStreamWriteValue32_ptsz_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWriteValue32PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWriteValue32PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_write_value32ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWriteValue32PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamWaitValue32PtszParams

Bases: object

Empty-initialize an instance of cuStreamWaitValue32_ptsz_params.

See also

cuStreamWaitValue32_ptsz_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWaitValue32PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWaitValue32PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_wait_value32ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWaitValue32PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamWriteValue64PtszParams

Bases: object

Empty-initialize an instance of cuStreamWriteValue64_ptsz_params.

See also

cuStreamWriteValue64_ptsz_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWriteValue64PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWriteValue64PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_write_value64ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWriteValue64PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamWaitValue64PtszParams

Bases: object

Empty-initialize an instance of cuStreamWaitValue64_ptsz_params.

See also

cuStreamWaitValue64_ptsz_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWaitValue64PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWaitValue64PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_wait_value64ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWaitValue64PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamBatchMemOpPtszParams

Bases: object

Empty-initialize an instance of cuStreamBatchMemOp_ptsz_params.

See also

cuStreamBatchMemOp_ptsz_params

count

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamBatchMemOpPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamBatchMemOpPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_batch_mem_op_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamBatchMemOpPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

param_array

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CuStreamWriteValue32V2Params

Bases: object

Empty-initialize an instance of cuStreamWriteValue32_v2_params.

See also

cuStreamWriteValue32_v2_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWriteValue32V2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWriteValue32V2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_write_value32v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWriteValue32V2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamWaitValue32V2Params

Bases: object

Empty-initialize an instance of cuStreamWaitValue32_v2_params.

See also

cuStreamWaitValue32_v2_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWaitValue32V2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWaitValue32V2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_wait_value32v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWaitValue32V2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamWriteValue64V2Params

Bases: object

Empty-initialize an instance of cuStreamWriteValue64_v2_params.

See also

cuStreamWriteValue64_v2_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWriteValue64V2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWriteValue64V2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_write_value64v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWriteValue64V2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamWaitValue64V2Params

Bases: object

Empty-initialize an instance of cuStreamWaitValue64_v2_params.

See also

cuStreamWaitValue64_v2_params

addr

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamWaitValue64V2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamWaitValue64V2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_wait_value64v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamWaitValue64V2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

stream

int:

value

int:

class cupti.cupti.CuStreamBatchMemOpV2Params

Bases: object

Empty-initialize an instance of cuStreamBatchMemOp_v2_params.

See also

cuStreamBatchMemOp_v2_params

count

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamBatchMemOpV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamBatchMemOpV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_batch_mem_op_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamBatchMemOpV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

param_array

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CuMemPrefetchAsyncParams

Bases: object

Empty-initialize an instance of cuMemPrefetchAsync_params.

See also

cuMemPrefetchAsync_params

count

int:

dev_ptr

int:

dst_device

int:

static from_buffer(buffer)

Create an CuMemPrefetchAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPrefetchAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_prefetch_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPrefetchAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLaunchCooperativeKernelParams

Bases: object

Empty-initialize an instance of cuLaunchCooperativeKernel_params.

See also

cuLaunchCooperativeKernel_params

block_dim_x

int:

block_dim_y

int:

block_dim_z

int:

f

int:

static from_buffer(buffer)

Create an CuLaunchCooperativeKernelParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLaunchCooperativeKernelParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_launch_cooperative_kernel_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLaunchCooperativeKernelParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

grid_dim_x

int:

grid_dim_y

int:

grid_dim_z

int:

h_stream

int:

kernel_params

int:

ptr

Get the pointer address to the data as Python int.

shared_mem_bytes

int:

class cupti.cupti.CuSignalExternalSemaphoresAsyncParams

Bases: object

Empty-initialize an instance of cuSignalExternalSemaphoresAsync_params.

See also

cuSignalExternalSemaphoresAsync_params

ext_sem_array

int:

static from_buffer(buffer)

Create an CuSignalExternalSemaphoresAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuSignalExternalSemaphoresAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_signal_external_semaphores_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuSignalExternalSemaphoresAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_ext_sems

int:

params_array

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CuWaitExternalSemaphoresAsyncParams

Bases: object

Empty-initialize an instance of cuWaitExternalSemaphoresAsync_params.

See also

cuWaitExternalSemaphoresAsync_params

ext_sem_array

int:

static from_buffer(buffer)

Create an CuWaitExternalSemaphoresAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuWaitExternalSemaphoresAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_wait_external_semaphores_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuWaitExternalSemaphoresAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_ext_sems

int:

params_array

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CuStreamBeginCaptureParams

Bases: object

Empty-initialize an instance of cuStreamBeginCapture_params.

See also

cuStreamBeginCapture_params

static from_buffer(buffer)

Create an CuStreamBeginCaptureParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamBeginCaptureParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_begin_capture_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamBeginCaptureParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamBeginCapturePtszParams

Bases: object

Empty-initialize an instance of cuStreamBeginCapture_ptsz_params.

See also

cuStreamBeginCapture_ptsz_params

static from_buffer(buffer)

Create an CuStreamBeginCapturePtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamBeginCapturePtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_begin_capture_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamBeginCapturePtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamBeginCaptureV2Params

Bases: object

Empty-initialize an instance of cuStreamBeginCapture_v2_params.

See also

cuStreamBeginCapture_v2_params

static from_buffer(buffer)

Create an CuStreamBeginCaptureV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamBeginCaptureV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_begin_capture_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamBeginCaptureV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

mode

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamBeginCaptureToGraphParams

Bases: object

Empty-initialize an instance of cuStreamBeginCaptureToGraph_params.

See also

cuStreamBeginCaptureToGraph_params

dependencies

int:

dependency_data

int:

static from_buffer(buffer)

Create an CuStreamBeginCaptureToGraphParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamBeginCaptureToGraphParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_begin_capture_to_graph_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamBeginCaptureToGraphParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

h_stream

int:

mode

int:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamEndCaptureParams

Bases: object

Empty-initialize an instance of cuStreamEndCapture_params.

See also

cuStreamEndCapture_params

static from_buffer(buffer)

Create an CuStreamEndCaptureParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamEndCaptureParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_end_capture_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamEndCaptureParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ph_graph

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamIsCapturingParams

Bases: object

Empty-initialize an instance of cuStreamIsCapturing_params.

See also

cuStreamIsCapturing_params

capture_status

int:

static from_buffer(buffer)

Create an CuStreamIsCapturingParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamIsCapturingParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_is_capturing_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamIsCapturingParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetCaptureInfoParams

Bases: object

Empty-initialize an instance of cuStreamGetCaptureInfo_params.

See also

cuStreamGetCaptureInfo_params

capture_status_out

int:

static from_buffer(buffer)

Create an CuStreamGetCaptureInfoParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetCaptureInfoParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_capture_info_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetCaptureInfoParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

id_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetCaptureInfoPtszParams

Bases: object

Empty-initialize an instance of cuStreamGetCaptureInfo_ptsz_params.

See also

cuStreamGetCaptureInfo_ptsz_params

capture_status_out

int:

static from_buffer(buffer)

Create an CuStreamGetCaptureInfoPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetCaptureInfoPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_capture_info_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetCaptureInfoPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

id_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetCaptureInfoV2Params

Bases: object

Empty-initialize an instance of cuStreamGetCaptureInfo_v2_params.

See also

cuStreamGetCaptureInfo_v2_params

capture_status_out

int:

dependencies_out

int:

static from_buffer(buffer)

Create an CuStreamGetCaptureInfoV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetCaptureInfoV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_capture_info_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetCaptureInfoV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_out

int:

h_stream

int:

id_out

int:

num_dependencies_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetCaptureInfoV3Params

Bases: object

Empty-initialize an instance of cuStreamGetCaptureInfo_v3_params.

See also

cuStreamGetCaptureInfo_v3_params

capture_status_out

int:

dependencies_out

int:

edge_data_out

int:

static from_buffer(buffer)

Create an CuStreamGetCaptureInfoV3Params instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetCaptureInfoV3Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_capture_info_v3params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetCaptureInfoV3Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_out

int:

h_stream

int:

id_out

int:

num_dependencies_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddKernelNodeParams

Bases: object

Empty-initialize an instance of cuGraphAddKernelNode_params.

See also

cuGraphAddKernelNode_params

dependencies

int:

static from_buffer(buffer)

Create an CuGraphAddKernelNodeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddKernelNodeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_kernel_node_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddKernelNodeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

node_params

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphKernelNodeGetParamsParams

Bases: object

Empty-initialize an instance of cuGraphKernelNodeGetParams_params.

See also

cuGraphKernelNodeGetParams_params

static from_buffer(buffer)

Create an CuGraphKernelNodeGetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphKernelNodeGetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_kernel_node_get_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphKernelNodeGetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphKernelNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphKernelNodeSetParams_params.

See also

cuGraphKernelNodeSetParams_params

static from_buffer(buffer)

Create an CuGraphKernelNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphKernelNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_kernel_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphKernelNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecKernelNodeSetParamsParams

Bases: object

Empty-initialize an instance of cuGraphExecKernelNodeSetParams_params.

See also

cuGraphExecKernelNodeSetParams_params

static from_buffer(buffer)

Create an CuGraphExecKernelNodeSetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecKernelNodeSetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_kernel_node_set_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecKernelNodeSetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph_exec

int:

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphInstantiateWithParamsParams

Bases: object

Empty-initialize an instance of cuGraphInstantiateWithParams_params.

See also

cuGraphInstantiateWithParams_params

static from_buffer(buffer)

Create an CuGraphInstantiateWithParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphInstantiateWithParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_instantiate_with_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphInstantiateWithParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

instantiate_params

int:

ph_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecUpdateParams

Bases: object

Empty-initialize an instance of cuGraphExecUpdate_params.

See also

cuGraphExecUpdate_params

static from_buffer(buffer)

Create an CuGraphExecUpdateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecUpdateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_update_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecUpdateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_error_node_out

int:

h_graph

int:

h_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

update_result_out

int:

class cupti.cupti.CuGraphUploadParams

Bases: object

Empty-initialize an instance of cuGraphUpload_params.

See also

cuGraphUpload_params

static from_buffer(buffer)

Create an CuGraphUploadParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphUploadParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_upload_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphUploadParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphLaunchParams

Bases: object

Empty-initialize an instance of cuGraphLaunch_params.

See also

cuGraphLaunch_params

static from_buffer(buffer)

Create an CuGraphLaunchParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphLaunchParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_launch_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphLaunchParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamCopyAttributesParams

Bases: object

Empty-initialize an instance of cuStreamCopyAttributes_params.

See also

cuStreamCopyAttributes_params

dst_stream

int:

static from_buffer(buffer)

Create an CuStreamCopyAttributesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamCopyAttributesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_copy_attributes_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamCopyAttributesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

src_stream

int:

class cupti.cupti.CuStreamGetAttributeParams

Bases: object

Empty-initialize an instance of cuStreamGetAttribute_params.

See also

cuStreamGetAttribute_params

attr

int:

static from_buffer(buffer)

Create an CuStreamGetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

value

int:

class cupti.cupti.CuStreamSetAttributeParams

Bases: object

Empty-initialize an instance of cuStreamSetAttribute_params.

See also

cuStreamSetAttribute_params

attr

int:

static from_buffer(buffer)

Create an CuStreamSetAttributeParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamSetAttributeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_set_attribute_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamSetAttributeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

param

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphInstantiateParams

Bases: object

Empty-initialize an instance of cuGraphInstantiate_params.

See also

cuGraphInstantiate_params

buffer_size

int:

static from_buffer(buffer)

Create an CuGraphInstantiateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphInstantiateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_instantiate_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphInstantiateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

log_buffer

str:

ph_error_node

int:

ph_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphInstantiateV2Params

Bases: object

Empty-initialize an instance of cuGraphInstantiate_v2_params.

See also

cuGraphInstantiate_v2_params

buffer_size

int:

static from_buffer(buffer)

Create an CuGraphInstantiateV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphInstantiateV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_instantiate_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphInstantiateV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

log_buffer

str:

ph_error_node

int:

ph_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemMapArrayAsyncParams

Bases: object

Empty-initialize an instance of cuMemMapArrayAsync_params.

See also

cuMemMapArrayAsync_params

count

int:

static from_buffer(buffer)

Create an CuMemMapArrayAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemMapArrayAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_map_array_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemMapArrayAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

map_info_list

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemFreeAsyncParams

Bases: object

Empty-initialize an instance of cuMemFreeAsync_params.

See also

cuMemFreeAsync_params

dptr

int:

static from_buffer(buffer)

Create an CuMemFreeAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemFreeAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_free_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemFreeAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemAllocAsyncParams

Bases: object

Empty-initialize an instance of cuMemAllocAsync_params.

See also

cuMemAllocAsync_params

bytesize

int:

dptr

int:

static from_buffer(buffer)

Create an CuMemAllocAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemAllocAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_alloc_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemAllocAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemAllocFromPoolAsyncParams

Bases: object

Empty-initialize an instance of cuMemAllocFromPoolAsync_params.

See also

cuMemAllocFromPoolAsync_params

bytesize

int:

dptr

int:

static from_buffer(buffer)

Create an CuMemAllocFromPoolAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemAllocFromPoolAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_alloc_from_pool_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemAllocFromPoolAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

pool

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamUpdateCaptureDependenciesParams

Bases: object

Empty-initialize an instance of cuStreamUpdateCaptureDependencies_params.

See also

cuStreamUpdateCaptureDependencies_params

dependencies

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamUpdateCaptureDependenciesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamUpdateCaptureDependenciesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_update_capture_dependencies_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamUpdateCaptureDependenciesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamUpdateCaptureDependenciesV2Params

Bases: object

Empty-initialize an instance of cuStreamUpdateCaptureDependencies_v2_params.

See also

cuStreamUpdateCaptureDependencies_v2_params

dependencies

int:

dependency_data

int:

flags_

int:

static from_buffer(buffer)

Create an CuStreamUpdateCaptureDependenciesV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamUpdateCaptureDependenciesV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_update_capture_dependencies_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamUpdateCaptureDependenciesV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_dependencies

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemBatchDecompressAsyncParams

Bases: object

Empty-initialize an instance of cuMemBatchDecompressAsync_params.

See also

cuMemBatchDecompressAsync_params

count

int:

error_index

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemBatchDecompressAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemBatchDecompressAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_batch_decompress_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemBatchDecompressAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

params_array

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.CuGetProcAddressParams

Bases: object

Empty-initialize an instance of cuGetProcAddress_params.

See also

cuGetProcAddress_params

cuda_version

int:

flags_

int:

static from_buffer(buffer)

Create an CuGetProcAddressParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGetProcAddressParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_get_proc_address_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGetProcAddressParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pfn

int:

ptr

Get the pointer address to the data as Python int.

symbol

str:

class cupti.cupti.CuCheckpointProcessGetRestoreThreadIdParams

Bases: object

Empty-initialize an instance of cuCheckpointProcessGetRestoreThreadId_params.

See also

cuCheckpointProcessGetRestoreThreadId_params

static from_buffer(buffer)

Create an CuCheckpointProcessGetRestoreThreadIdParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCheckpointProcessGetRestoreThreadIdParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_checkpoint_process_get_restore_thread_id_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCheckpointProcessGetRestoreThreadIdParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pid

int:

ptr

Get the pointer address to the data as Python int.

tid

int:

class cupti.cupti.CuCheckpointProcessGetStateParams

Bases: object

Empty-initialize an instance of cuCheckpointProcessGetState_params.

See also

cuCheckpointProcessGetState_params

static from_buffer(buffer)

Create an CuCheckpointProcessGetStateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCheckpointProcessGetStateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_checkpoint_process_get_state_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCheckpointProcessGetStateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pid

int:

ptr

Get the pointer address to the data as Python int.

state

int:

class cupti.cupti.CuCheckpointProcessLockParams

Bases: object

Empty-initialize an instance of cuCheckpointProcessLock_params.

See also

cuCheckpointProcessLock_params

args

int:

static from_buffer(buffer)

Create an CuCheckpointProcessLockParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCheckpointProcessLockParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_checkpoint_process_lock_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCheckpointProcessLockParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pid

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCheckpointProcessCheckpointParams

Bases: object

Empty-initialize an instance of cuCheckpointProcessCheckpoint_params.

See also

cuCheckpointProcessCheckpoint_params

args

int:

static from_buffer(buffer)

Create an CuCheckpointProcessCheckpointParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCheckpointProcessCheckpointParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_checkpoint_process_checkpoint_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCheckpointProcessCheckpointParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pid

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCheckpointProcessRestoreParams

Bases: object

Empty-initialize an instance of cuCheckpointProcessRestore_params.

See also

cuCheckpointProcessRestore_params

args

int:

static from_buffer(buffer)

Create an CuCheckpointProcessRestoreParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCheckpointProcessRestoreParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_checkpoint_process_restore_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCheckpointProcessRestoreParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pid

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCheckpointProcessUnlockParams

Bases: object

Empty-initialize an instance of cuCheckpointProcessUnlock_params.

See also

cuCheckpointProcessUnlock_params

args

int:

static from_buffer(buffer)

Create an CuCheckpointProcessUnlockParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCheckpointProcessUnlockParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_checkpoint_process_unlock_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCheckpointProcessUnlockParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pid

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetHostAtomicCapabilitiesParams

Bases: object

Empty-initialize an instance of cuDeviceGetHostAtomicCapabilities_params.

See also

cuDeviceGetHostAtomicCapabilities_params

capabilities

int:

count

int:

dev

int:

static from_buffer(buffer)

Create an CuDeviceGetHostAtomicCapabilitiesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetHostAtomicCapabilitiesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_host_atomic_capabilities_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetHostAtomicCapabilitiesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

operations

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxGetDeviceV2Params

Bases: object

Empty-initialize an instance of cuCtxGetDevice_v2_params.

See also

cuCtxGetDevice_v2_params

ctx

int:

device_

int:

static from_buffer(buffer)

Create an CuCtxGetDeviceV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxGetDeviceV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_get_device_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxGetDeviceV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCtxSynchronizeV2Params

Bases: object

Empty-initialize an instance of cuCtxSynchronize_v2_params.

See also

cuCtxSynchronize_v2_params

ctx

int:

static from_buffer(buffer)

Create an CuCtxSynchronizeV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuCtxSynchronizeV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ctx_synchronize_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCtxSynchronizeV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpyBatchAsyncV2PtszParams

Bases: object

Empty-initialize an instance of cuMemcpyBatchAsync_v2_ptsz_params.

See also

cuMemcpyBatchAsync_v2_ptsz_params

attrs

int:

attrs_idxs

int:

count

int:

dsts

int:

static from_buffer(buffer)

Create an CuMemcpyBatchAsyncV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyBatchAsyncV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_batch_async_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyBatchAsyncV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_attrs

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

srcs

int:

class cupti.cupti.CuMemcpy3DBatchAsyncV2PtszParams

Bases: object

Empty-initialize an instance of cuMemcpy3DBatchAsync_v2_ptsz_params.

See also

cuMemcpy3DBatchAsync_v2_ptsz_params

flags_

int:

static from_buffer(buffer)

Create an CuMemcpy3DBatchAsyncV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DBatchAsyncV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3d_batch_async_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DBatchAsyncV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_ops

int:

op_list

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemGetDefaultMemPoolParams

Bases: object

Empty-initialize an instance of cuMemGetDefaultMemPool_params.

See also

cuMemGetDefaultMemPool_params

static from_buffer(buffer)

Create an CuMemGetDefaultMemPoolParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemGetDefaultMemPoolParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_get_default_mem_pool_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemGetDefaultMemPoolParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

location

int:

pool_out

int:

ptr

Get the pointer address to the data as Python int.

type

int:

class cupti.cupti.CuMemGetMemPoolParams

Bases: object

Empty-initialize an instance of cuMemGetMemPool_params.

See also

cuMemGetMemPool_params

static from_buffer(buffer)

Create an CuMemGetMemPoolParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemGetMemPoolParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_get_mem_pool_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemGetMemPoolParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

location

int:

pool

int:

ptr

Get the pointer address to the data as Python int.

type

int:

class cupti.cupti.CuMemSetMemPoolParams

Bases: object

Empty-initialize an instance of cuMemSetMemPool_params.

See also

cuMemSetMemPool_params

static from_buffer(buffer)

Create an CuMemSetMemPoolParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemSetMemPoolParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_set_mem_pool_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemSetMemPoolParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

location

int:

pool

int:

ptr

Get the pointer address to the data as Python int.

type

int:

class cupti.cupti.CuMemPrefetchBatchAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemPrefetchBatchAsync_ptsz_params.

See also

cuMemPrefetchBatchAsync_ptsz_params

count

int:

dptrs

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemPrefetchBatchAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPrefetchBatchAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_prefetch_batch_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPrefetchBatchAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_prefetch_locs

int:

prefetch_loc_idxs

int:

prefetch_locs

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

class cupti.cupti.CuMemDiscardBatchAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemDiscardBatchAsync_ptsz_params.

See also

cuMemDiscardBatchAsync_ptsz_params

count

int:

dptrs

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemDiscardBatchAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemDiscardBatchAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_discard_batch_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemDiscardBatchAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

class cupti.cupti.CuMemDiscardAndPrefetchBatchAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemDiscardAndPrefetchBatchAsync_ptsz_params.

See also

cuMemDiscardAndPrefetchBatchAsync_ptsz_params

count

int:

dptrs

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemDiscardAndPrefetchBatchAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemDiscardAndPrefetchBatchAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_discard_and_prefetch_batch_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemDiscardAndPrefetchBatchAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_prefetch_locs

int:

prefetch_loc_idxs

int:

prefetch_locs

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

class cupti.cupti.CuTensorMapEncodeIm2colWideParams

Bases: object

Empty-initialize an instance of cuTensorMapEncodeIm2colWide_params.

See also

cuTensorMapEncodeIm2colWide_params

channels_per_pixel

int:

element_strides

int:

static from_buffer(buffer)

Create an CuTensorMapEncodeIm2colWideParams instance with the memory from the given buffer.

static from_data(data)

Create an CuTensorMapEncodeIm2colWideParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_tensor_map_encode_im2col_wide_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuTensorMapEncodeIm2colWideParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

global_address

int:

global_dim

int:

global_strides

int:

interleave

int:

l2promotion

int:

mode

int:

oob_fill

int:

pixel_box_lower_corner_width

int:

pixel_box_upper_corner_width

int:

pixels_per_column

int:

ptr

Get the pointer address to the data as Python int.

swizzle

int:

tensor_data_type

int:

tensor_map

int:

tensor_rank

int:

class cupti.cupti.CuDeviceGetP2PAtomicCapabilitiesParams

Bases: object

Empty-initialize an instance of cuDeviceGetP2PAtomicCapabilities_params.

See also

cuDeviceGetP2PAtomicCapabilities_params

capabilities

int:

count

int:

dst_device

int:

static from_buffer(buffer)

Create an CuDeviceGetP2PAtomicCapabilitiesParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetP2PAtomicCapabilitiesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_p2p_atomic_capabilities_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetP2PAtomicCapabilitiesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

operations

int:

ptr

Get the pointer address to the data as Python int.

src_device

int:

class cupti.cupti.CuGreenCtxGetIdParams

Bases: object

Empty-initialize an instance of cuGreenCtxGetId_params.

See also

cuGreenCtxGetId_params

static from_buffer(buffer)

Create an CuGreenCtxGetIdParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGreenCtxGetIdParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_green_ctx_get_id_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGreenCtxGetIdParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

green_ctx

int:

green_ctx_id

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLogsRegisterCallbackParams

Bases: object

Empty-initialize an instance of cuLogsRegisterCallback_params.

See also

cuLogsRegisterCallback_params

callback_func

int:

callback_out

int:

static from_buffer(buffer)

Create an CuLogsRegisterCallbackParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogsRegisterCallbackParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logs_register_callback_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogsRegisterCallbackParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

user_data

int:

class cupti.cupti.CuLogsUnregisterCallbackParams

Bases: object

Empty-initialize an instance of cuLogsUnregisterCallback_params.

See also

cuLogsUnregisterCallback_params

callback

int:

static from_buffer(buffer)

Create an CuLogsUnregisterCallbackParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogsUnregisterCallbackParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logs_unregister_callback_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogsUnregisterCallbackParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLogsCurrentParams

Bases: object

Empty-initialize an instance of cuLogsCurrent_params.

See also

cuLogsCurrent_params

flags_

int:

static from_buffer(buffer)

Create an CuLogsCurrentParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogsCurrentParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logs_current_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogsCurrentParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

iterator_out

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLogsDumpToFileParams

Bases: object

Empty-initialize an instance of cuLogsDumpToFile_params.

See also

cuLogsDumpToFile_params

flags_

int:

static from_buffer(buffer)

Create an CuLogsDumpToFileParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogsDumpToFileParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logs_dump_to_file_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogsDumpToFileParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

iterator

int:

path_to_file

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLogsDumpToMemoryParams

Bases: object

Empty-initialize an instance of cuLogsDumpToMemory_params.

See also

cuLogsDumpToMemory_params

buffer

str:

flags_

int:

static from_buffer(buffer)

Create an CuLogsDumpToMemoryParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogsDumpToMemoryParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logs_dump_to_memory_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogsDumpToMemoryParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

iterator

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CuMemcpyBatchAsyncV2Params

Bases: object

Empty-initialize an instance of cuMemcpyBatchAsync_v2_params.

See also

cuMemcpyBatchAsync_v2_params

attrs

int:

attrs_idxs

int:

count

int:

dsts

int:

static from_buffer(buffer)

Create an CuMemcpyBatchAsyncV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyBatchAsyncV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_batch_async_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyBatchAsyncV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_attrs

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

srcs

int:

class cupti.cupti.CuMemcpy3DBatchAsyncV2Params

Bases: object

Empty-initialize an instance of cuMemcpy3DBatchAsync_v2_params.

See also

cuMemcpy3DBatchAsync_v2_params

flags_

int:

static from_buffer(buffer)

Create an CuMemcpy3DBatchAsyncV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DBatchAsyncV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3d_batch_async_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DBatchAsyncV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_ops

int:

op_list

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemPrefetchBatchAsyncParams

Bases: object

Empty-initialize an instance of cuMemPrefetchBatchAsync_params.

See also

cuMemPrefetchBatchAsync_params

count

int:

dptrs

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemPrefetchBatchAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPrefetchBatchAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_prefetch_batch_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPrefetchBatchAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_prefetch_locs

int:

prefetch_loc_idxs

int:

prefetch_locs

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

class cupti.cupti.CuMemDiscardBatchAsyncParams

Bases: object

Empty-initialize an instance of cuMemDiscardBatchAsync_params.

See also

cuMemDiscardBatchAsync_params

count

int:

dptrs

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemDiscardBatchAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemDiscardBatchAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_discard_batch_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemDiscardBatchAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

class cupti.cupti.CuMemDiscardAndPrefetchBatchAsyncParams

Bases: object

Empty-initialize an instance of cuMemDiscardAndPrefetchBatchAsync_params.

See also

cuMemDiscardAndPrefetchBatchAsync_params

count

int:

dptrs

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemDiscardAndPrefetchBatchAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemDiscardAndPrefetchBatchAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_discard_and_prefetch_batch_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemDiscardAndPrefetchBatchAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

num_prefetch_locs

int:

prefetch_loc_idxs

int:

prefetch_locs

int:

ptr

Get the pointer address to the data as Python int.

sizes

int:

class cupti.cupti.CuKernelGetParamCountParams

Bases: object

Empty-initialize an instance of cuKernelGetParamCount_params.

See also

cuKernelGetParamCount_params

static from_buffer(buffer)

Create an CuKernelGetParamCountParams instance with the memory from the given buffer.

static from_data(data)

Create an CuKernelGetParamCountParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_kernel_get_param_count_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuKernelGetParamCountParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

kernel

int:

param_count

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemcpyWithAttributesAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemcpyWithAttributesAsync_ptsz_params.

See also

cuMemcpyWithAttributesAsync_ptsz_params

attr

int:

dst

int:

static from_buffer(buffer)

Create an CuMemcpyWithAttributesAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyWithAttributesAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_with_attributes_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyWithAttributesAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

src

int:

class cupti.cupti.CuMemcpy3DWithAttributesAsyncPtszParams

Bases: object

Empty-initialize an instance of cuMemcpy3DWithAttributesAsync_ptsz_params.

See also

cuMemcpy3DWithAttributesAsync_ptsz_params

flags_

int:

static from_buffer(buffer)

Create an CuMemcpy3DWithAttributesAsyncPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DWithAttributesAsyncPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3d_with_attributes_async_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DWithAttributesAsyncPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

op

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMulticastBindMemV2Params

Bases: object

Empty-initialize an instance of cuMulticastBindMem_v2_params.

See also

cuMulticastBindMem_v2_params

dev

int:

flags_

int:

static from_buffer(buffer)

Create an CuMulticastBindMemV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMulticastBindMemV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_multicast_bind_mem_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMulticastBindMemV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mc_handle

int:

mc_offset

int:

mem_handle

int:

mem_offset

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CuMulticastBindAddrV2Params

Bases: object

Empty-initialize an instance of cuMulticastBindAddr_v2_params.

See also

cuMulticastBindAddr_v2_params

dev

int:

flags_

int:

static from_buffer(buffer)

Create an CuMulticastBindAddrV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMulticastBindAddrV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_multicast_bind_addr_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMulticastBindAddrV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

mc_handle

int:

mc_offset

int:

memptr

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CuStreamBeginCaptureToCigPtszParams

Bases: object

Empty-initialize an instance of cuStreamBeginCaptureToCig_ptsz_params.

See also

cuStreamBeginCaptureToCig_ptsz_params

static from_buffer(buffer)

Create an CuStreamBeginCaptureToCigPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamBeginCaptureToCigPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_begin_capture_to_cig_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamBeginCaptureToCigPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

stream_cig_capture_params

int:

class cupti.cupti.CuStreamEndCaptureToCigPtszParams

Bases: object

Empty-initialize an instance of cuStreamEndCaptureToCig_ptsz_params.

See also

cuStreamEndCaptureToCig_ptsz_params

static from_buffer(buffer)

Create an CuStreamEndCaptureToCigPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamEndCaptureToCigPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_end_capture_to_cig_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamEndCaptureToCigPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuFuncGetParamCountParams

Bases: object

Empty-initialize an instance of cuFuncGetParamCount_params.

See also

cuFuncGetParamCount_params

static from_buffer(buffer)

Create an CuFuncGetParamCountParams instance with the memory from the given buffer.

static from_data(data)

Create an CuFuncGetParamCountParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_func_get_param_count_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuFuncGetParamCountParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

func

int:

param_count

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLaunchHostFuncV2PtszParams

Bases: object

Empty-initialize an instance of cuLaunchHostFunc_v2_ptsz_params.

See also

cuLaunchHostFunc_v2_ptsz_params

fn

int:

static from_buffer(buffer)

Create an CuLaunchHostFuncV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLaunchHostFuncV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_launch_host_func_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLaunchHostFuncV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

sync_mode

int:

user_data

int:

class cupti.cupti.CuGraphNodeGetContainingGraphParams

Bases: object

Empty-initialize an instance of cuGraphNodeGetContainingGraph_params.

See also

cuGraphNodeGetContainingGraph_params

static from_buffer(buffer)

Create an CuGraphNodeGetContainingGraphParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphNodeGetContainingGraphParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_node_get_containing_graph_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphNodeGetContainingGraphParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ph_graph

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphNodeGetLocalIdParams

Bases: object

Empty-initialize an instance of cuGraphNodeGetLocalId_params.

See also

cuGraphNodeGetLocalId_params

static from_buffer(buffer)

Create an CuGraphNodeGetLocalIdParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphNodeGetLocalIdParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_node_get_local_id_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphNodeGetLocalIdParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_id

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphNodeGetToolsIdParams

Bases: object

Empty-initialize an instance of cuGraphNodeGetToolsId_params.

See also

cuGraphNodeGetToolsId_params

static from_buffer(buffer)

Create an CuGraphNodeGetToolsIdParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphNodeGetToolsIdParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_node_get_tools_id_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphNodeGetToolsIdParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

ptr

Get the pointer address to the data as Python int.

tools_node_id

int:

class cupti.cupti.CuGraphGetIdParams

Bases: object

Empty-initialize an instance of cuGraphGetId_params.

See also

cuGraphGetId_params

static from_buffer(buffer)

Create an CuGraphGetIdParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphGetIdParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_get_id_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphGetIdParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

int:

h_graph

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphExecGetIdParams

Bases: object

Empty-initialize an instance of cuGraphExecGetId_params.

See also

cuGraphExecGetId_params

static from_buffer(buffer)

Create an CuGraphExecGetIdParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphExecGetIdParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_exec_get_id_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphExecGetIdParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

graph_id

int:

h_graph_exec

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphNodeGetParamsParams

Bases: object

Empty-initialize an instance of cuGraphNodeGetParams_params.

See also

cuGraphNodeGetParams_params

static from_buffer(buffer)

Create an CuGraphNodeGetParamsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphNodeGetParamsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_node_get_params_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphNodeGetParamsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCoredumpRegisterStartCallbackParams

Bases: object

Empty-initialize an instance of cuCoredumpRegisterStartCallback_params.

See also

cuCoredumpRegisterStartCallback_params

callback

int:

callback_out

int:

static from_buffer(buffer)

Create an CuCoredumpRegisterStartCallbackParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCoredumpRegisterStartCallbackParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_coredump_register_start_callback_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCoredumpRegisterStartCallbackParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

user_data

int:

class cupti.cupti.CuCoredumpRegisterCompleteCallbackParams

Bases: object

Empty-initialize an instance of cuCoredumpRegisterCompleteCallback_params.

See also

cuCoredumpRegisterCompleteCallback_params

callback

int:

callback_out

int:

static from_buffer(buffer)

Create an CuCoredumpRegisterCompleteCallbackParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCoredumpRegisterCompleteCallbackParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_coredump_register_complete_callback_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCoredumpRegisterCompleteCallbackParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

user_data

int:

class cupti.cupti.CuCoredumpDeregisterStartCallbackParams

Bases: object

Empty-initialize an instance of cuCoredumpDeregisterStartCallback_params.

See also

cuCoredumpDeregisterStartCallback_params

callback

int:

static from_buffer(buffer)

Create an CuCoredumpDeregisterStartCallbackParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCoredumpDeregisterStartCallbackParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_coredump_deregister_start_callback_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCoredumpDeregisterStartCallbackParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuCoredumpDeregisterCompleteCallbackParams

Bases: object

Empty-initialize an instance of cuCoredumpDeregisterCompleteCallback_params.

See also

cuCoredumpDeregisterCompleteCallback_params

callback

int:

static from_buffer(buffer)

Create an CuCoredumpDeregisterCompleteCallbackParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCoredumpDeregisterCompleteCallbackParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_coredump_deregister_complete_callback_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCoredumpDeregisterCompleteCallbackParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDevSmResourceSplitParams

Bases: object

Empty-initialize an instance of cuDevSmResourceSplit_params.

See also

cuDevSmResourceSplit_params

flags_

int:

static from_buffer(buffer)

Create an CuDevSmResourceSplitParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDevSmResourceSplitParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_dev_sm_resource_split_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDevSmResourceSplitParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

group_params

int:

input

int:

nb_groups

int:

ptr

Get the pointer address to the data as Python int.

remainder

int:

result

int:

class cupti.cupti.CuStreamGetDevResourcePtszParams

Bases: object

Empty-initialize an instance of cuStreamGetDevResource_ptsz_params.

See also

cuStreamGetDevResource_ptsz_params

static from_buffer(buffer)

Create an CuStreamGetDevResourcePtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetDevResourcePtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_dev_resource_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetDevResourcePtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

resource

int:

type

int:

class cupti.cupti.CuMemcpyWithAttributesAsyncParams

Bases: object

Empty-initialize an instance of cuMemcpyWithAttributesAsync_params.

See also

cuMemcpyWithAttributesAsync_params

attr

int:

dst

int:

static from_buffer(buffer)

Create an CuMemcpyWithAttributesAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpyWithAttributesAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy_with_attributes_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpyWithAttributesAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

src

int:

class cupti.cupti.CuMemcpy3DWithAttributesAsyncParams

Bases: object

Empty-initialize an instance of cuMemcpy3DWithAttributesAsync_params.

See also

cuMemcpy3DWithAttributesAsync_params

flags_

int:

static from_buffer(buffer)

Create an CuMemcpy3DWithAttributesAsyncParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemcpy3DWithAttributesAsyncParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_memcpy3d_with_attributes_async_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemcpy3DWithAttributesAsyncParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

op

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamGetDevResourceParams

Bases: object

Empty-initialize an instance of cuStreamGetDevResource_params.

See also

cuStreamGetDevResource_params

static from_buffer(buffer)

Create an CuStreamGetDevResourceParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamGetDevResourceParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_get_dev_resource_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamGetDevResourceParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

resource

int:

type

int:

class cupti.cupti.CuLaunchHostFuncV2Params

Bases: object

Empty-initialize an instance of cuLaunchHostFunc_v2_params.

See also

cuLaunchHostFunc_v2_params

fn

int:

static from_buffer(buffer)

Create an CuLaunchHostFuncV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuLaunchHostFuncV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_launch_host_func_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLaunchHostFuncV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

sync_mode

int:

user_data

int:

class cupti.cupti.CuStreamBeginCaptureToCigParams

Bases: object

Empty-initialize an instance of cuStreamBeginCaptureToCig_params.

See also

cuStreamBeginCaptureToCig_params

static from_buffer(buffer)

Create an CuStreamBeginCaptureToCigParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamBeginCaptureToCigParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_begin_capture_to_cig_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamBeginCaptureToCigParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

stream_cig_capture_params

int:

class cupti.cupti.CuStreamEndCaptureToCigParams

Bases: object

Empty-initialize an instance of cuStreamEndCaptureToCig_params.

See also

cuStreamEndCaptureToCig_params

static from_buffer(buffer)

Create an CuStreamEndCaptureToCigParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamEndCaptureToCigParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_end_capture_to_cig_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamEndCaptureToCigParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuStreamBeginRecaptureToGraphPtszParams

Bases: object

Empty-initialize an instance of cuStreamBeginRecaptureToGraph_ptsz_params.

See also

cuStreamBeginRecaptureToGraph_ptsz_params

callback_func

int:

static from_buffer(buffer)

Create an CuStreamBeginRecaptureToGraphPtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamBeginRecaptureToGraphPtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_begin_recapture_to_graph_ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamBeginRecaptureToGraphPtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

h_stream

int:

mode

int:

ptr

Get the pointer address to the data as Python int.

user_data

int:

class cupti.cupti.CuStreamBeginRecaptureToGraphParams

Bases: object

Empty-initialize an instance of cuStreamBeginRecaptureToGraph_params.

See also

cuStreamBeginRecaptureToGraph_params

callback_func

int:

static from_buffer(buffer)

Create an CuStreamBeginRecaptureToGraphParams instance with the memory from the given buffer.

static from_data(data)

Create an CuStreamBeginRecaptureToGraphParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_stream_begin_recapture_to_graph_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuStreamBeginRecaptureToGraphParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

h_stream

int:

mode

int:

ptr

Get the pointer address to the data as Python int.

user_data

int:

class cupti.cupti.CuLogicalEndpointIdReserveParams

Bases: object

Empty-initialize an instance of cuLogicalEndpointIdReserve_params.

See also

cuLogicalEndpointIdReserve_params

base_le_id

int:

count

int:

static from_buffer(buffer)

Create an CuLogicalEndpointIdReserveParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogicalEndpointIdReserveParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logical_endpoint_id_reserve_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogicalEndpointIdReserveParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLogicalEndpointIdReleaseParams

Bases: object

Empty-initialize an instance of cuLogicalEndpointIdRelease_params.

See also

cuLogicalEndpointIdRelease_params

base_le_id

int:

count

int:

static from_buffer(buffer)

Create an CuLogicalEndpointIdReleaseParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogicalEndpointIdReleaseParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logical_endpoint_id_release_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogicalEndpointIdReleaseParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLogicalEndpointCreateParams

Bases: object

Empty-initialize an instance of cuLogicalEndpointCreate_params.

See also

cuLogicalEndpointCreate_params

static from_buffer(buffer)

Create an CuLogicalEndpointCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogicalEndpointCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logical_endpoint_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogicalEndpointCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

le_id

int:

prop

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLogicalEndpointAddDeviceParams

Bases: object

Empty-initialize an instance of cuLogicalEndpointAddDevice_params.

See also

cuLogicalEndpointAddDevice_params

dev

int:

static from_buffer(buffer)

Create an CuLogicalEndpointAddDeviceParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogicalEndpointAddDeviceParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logical_endpoint_add_device_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogicalEndpointAddDeviceParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

le_id

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLogicalEndpointDestroyParams

Bases: object

Empty-initialize an instance of cuLogicalEndpointDestroy_params.

See also

cuLogicalEndpointDestroy_params

static from_buffer(buffer)

Create an CuLogicalEndpointDestroyParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogicalEndpointDestroyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logical_endpoint_destroy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogicalEndpointDestroyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

le_id

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLogicalEndpointBindAddrParams

Bases: object

Empty-initialize an instance of cuLogicalEndpointBindAddr_params.

See also

cuLogicalEndpointBindAddr_params

dev

int:

flags_

int:

static from_buffer(buffer)

Create an CuLogicalEndpointBindAddrParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogicalEndpointBindAddrParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logical_endpoint_bind_addr_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogicalEndpointBindAddrParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

le_id

int:

offset

int:

ptr

Get the pointer address to the data as Python int.

ptr_

int:

size_

int:

class cupti.cupti.CuLogicalEndpointBindMemParams

Bases: object

Empty-initialize an instance of cuLogicalEndpointBindMem_params.

See also

cuLogicalEndpointBindMem_params

dev

int:

flags_

int:

static from_buffer(buffer)

Create an CuLogicalEndpointBindMemParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogicalEndpointBindMemParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logical_endpoint_bind_mem_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogicalEndpointBindMemParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

le_id

int:

mem_handle

int:

mem_offset

int:

offset

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CuLogicalEndpointUnbindParams

Bases: object

Empty-initialize an instance of cuLogicalEndpointUnbind_params.

See also

cuLogicalEndpointUnbind_params

dev

int:

static from_buffer(buffer)

Create an CuLogicalEndpointUnbindParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogicalEndpointUnbindParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logical_endpoint_unbind_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogicalEndpointUnbindParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

le_id

int:

offset

int:

ptr

Get the pointer address to the data as Python int.

size_

int:

class cupti.cupti.CuLogicalEndpointExportParams

Bases: object

Empty-initialize an instance of cuLogicalEndpointExport_params.

See also

cuLogicalEndpointExport_params

static from_buffer(buffer)

Create an CuLogicalEndpointExportParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogicalEndpointExportParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logical_endpoint_export_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogicalEndpointExportParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

int:

handle_type

int:

le_id

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLogicalEndpointImportParams

Bases: object

Empty-initialize an instance of cuLogicalEndpointImport_params.

See also

cuLogicalEndpointImport_params

static from_buffer(buffer)

Create an CuLogicalEndpointImportParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogicalEndpointImportParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logical_endpoint_import_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogicalEndpointImportParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

int:

handle_type

int:

le_id

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLogicalEndpointGetLimitsParams

Bases: object

Empty-initialize an instance of cuLogicalEndpointGetLimits_params.

See also

cuLogicalEndpointGetLimits_params

bind_alignment

int:

static from_buffer(buffer)

Create an CuLogicalEndpointGetLimitsParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogicalEndpointGetLimitsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logical_endpoint_get_limits_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogicalEndpointGetLimitsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

max_size

int:

prop

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuLogicalEndpointQueryParams

Bases: object

Empty-initialize an instance of cuLogicalEndpointQuery_params.

See also

cuLogicalEndpointQuery_params

count

int:

static from_buffer(buffer)

Create an CuLogicalEndpointQueryParams instance with the memory from the given buffer.

static from_data(data)

Create an CuLogicalEndpointQueryParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_logical_endpoint_query_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuLogicalEndpointQueryParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

le_id

int:

ptr

Get the pointer address to the data as Python int.

query_status

int:

class cupti.cupti.CuMemGetLocationInfoParams

Bases: object

Empty-initialize an instance of cuMemGetLocationInfo_params.

See also

cuMemGetLocationInfo_params

static from_buffer(buffer)

Create an CuMemGetLocationInfoParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemGetLocationInfoParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_get_location_info_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemGetLocationInfoParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

location_out

int:

ptr

Get the pointer address to the data as Python int.

ptr_

int:

sampling_granularity

int:

size_

int:

summary_granularity

int:

class cupti.cupti.CuCheckpointOperationCompleteParams

Bases: object

Empty-initialize an instance of cuCheckpointOperationComplete_params.

See also

cuCheckpointOperationComplete_params

static from_buffer(buffer)

Create an CuCheckpointOperationCompleteParams instance with the memory from the given buffer.

static from_data(data)

Create an CuCheckpointOperationCompleteParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_checkpoint_operation_complete_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuCheckpointOperationCompleteParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphAddNodeV3Params

Bases: object

Empty-initialize an instance of cuGraphAddNode_v3_params.

See also

cuGraphAddNode_v3_params

dependencies

int:

dependency_data

int:

static from_buffer(buffer)

Create an CuGraphAddNodeV3Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphAddNodeV3Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_add_node_v3params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphAddNodeV3Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_graph

int:

node_params

int:

num_dependencies

int:

ph_graph_node

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuGraphNodeSetParamsV2Params

Bases: object

Empty-initialize an instance of cuGraphNodeSetParams_v2_params.

See also

cuGraphNodeSetParams_v2_params

static from_buffer(buffer)

Create an CuGraphNodeSetParamsV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuGraphNodeSetParamsV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_graph_node_set_params_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuGraphNodeSetParamsV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_node

int:

node_params

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetFabricClusterUuidParams

Bases: object

Empty-initialize an instance of cuDeviceGetFabricClusterUuid_params.

See also

cuDeviceGetFabricClusterUuid_params

dev

int:

static from_buffer(buffer)

Create an CuDeviceGetFabricClusterUuidParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetFabricClusterUuidParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_fabric_cluster_uuid_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetFabricClusterUuidParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

uuid

int:

class cupti.cupti.CuDeviceGetCliqueCountParams

Bases: object

Empty-initialize an instance of cuDeviceGetCliqueCount_params.

See also

cuDeviceGetCliqueCount_params

count

int:

dev

int:

static from_buffer(buffer)

Create an CuDeviceGetCliqueCountParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetCliqueCountParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_clique_count_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetCliqueCountParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuDeviceGetCliqueInfoParams

Bases: object

Empty-initialize an instance of cuDeviceGetCliqueInfo_params.

See also

cuDeviceGetCliqueInfo_params

clique_info

int:

count

int:

dev

int:

static from_buffer(buffer)

Create an CuDeviceGetCliqueInfoParams instance with the memory from the given buffer.

static from_data(data)

Create an CuDeviceGetCliqueInfoParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_device_get_clique_info_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuDeviceGetCliqueInfoParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuIpcOpenEventHandleParams

Bases: object

Empty-initialize an instance of cuIpcOpenEventHandle_params.

See also

cuIpcOpenEventHandle_params

static from_buffer(buffer)

Create an CuIpcOpenEventHandleParams instance with the memory from the given buffer.

static from_data(data)

Create an CuIpcOpenEventHandleParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ipc_open_event_handle_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuIpcOpenEventHandleParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ph_event

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuIpcOpenMemHandleParams

Bases: object

Empty-initialize an instance of cuIpcOpenMemHandle_params.

See also

cuIpcOpenMemHandle_params

flags_

int:

static from_buffer(buffer)

Create an CuIpcOpenMemHandleParams instance with the memory from the given buffer.

static from_data(data)

Create an CuIpcOpenMemHandleParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ipc_open_mem_handle_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuIpcOpenMemHandleParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pdptr

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuIpcOpenMemHandleV2Params

Bases: object

Empty-initialize an instance of cuIpcOpenMemHandle_v2_params.

See also

cuIpcOpenMemHandle_v2_params

flags_

int:

static from_buffer(buffer)

Create an CuIpcOpenMemHandleV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuIpcOpenMemHandleV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_ipc_open_mem_handle_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuIpcOpenMemHandleV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

pdptr

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemAdviseV2Params

Bases: object

Empty-initialize an instance of cuMemAdvise_v2_params.

See also

cuMemAdvise_v2_params

advice

int:

count

int:

dev_ptr

int:

static from_buffer(buffer)

Create an CuMemAdviseV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemAdviseV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_advise_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemAdviseV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemPrefetchAsyncV2Params

Bases: object

Empty-initialize an instance of cuMemPrefetchAsync_v2_params.

See also

cuMemPrefetchAsync_v2_params

count

int:

dev_ptr

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemPrefetchAsyncV2Params instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPrefetchAsyncV2Params instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_prefetch_async_v2params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPrefetchAsyncV2Params instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.CuMemPrefetchAsyncV2PtszParams

Bases: object

Empty-initialize an instance of cuMemPrefetchAsync_v2_ptsz_params.

See also

cuMemPrefetchAsync_v2_ptsz_params

count

int:

dev_ptr

int:

flags_

int:

static from_buffer(buffer)

Create an CuMemPrefetchAsyncV2PtszParams instance with the memory from the given buffer.

static from_data(data)

Create an CuMemPrefetchAsyncV2PtszParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype cu_mem_prefetch_async_v2ptsz_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an CuMemPrefetchAsyncV2PtszParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

h_stream

int:

ptr

Get the pointer address to the data as Python int.

Function Parameter Classes for NVTX APIs

class cupti.cupti.NvtxMarkExParams

Bases: object

Empty-initialize an instance of nvtxMarkEx_params.

See also

nvtxMarkEx_params

event_attrib

int:

static from_buffer(buffer)

Create an NvtxMarkExParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxMarkExParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_mark_ex_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxMarkExParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxMarkAParams

Bases: object

Empty-initialize an instance of nvtxMarkA_params.

See also

nvtxMarkA_params

static from_buffer(buffer)

Create an NvtxMarkAParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxMarkAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_mark_a_params_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an NvtxMarkAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

message

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxMarkWParams

Bases: object

Empty-initialize an instance of nvtxMarkW_params.

See also

nvtxMarkW_params

static from_buffer(buffer)

Create an NvtxMarkWParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxMarkWParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_mark_w_params_dtype holding the data.

static from_ptr(ptr: int, readonly: bool = False, owner=None)

Create an NvtxMarkWParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

message

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxRangeStartExParams

Bases: object

Empty-initialize an instance of nvtxRangeStartEx_params.

See also

nvtxRangeStartEx_params

event_attrib

int:

static from_buffer(buffer)

Create an NvtxRangeStartExParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxRangeStartExParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_range_start_ex_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxRangeStartExParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxRangeStartAParams

Bases: object

Empty-initialize an instance of nvtxRangeStartA_params.

See also

nvtxRangeStartA_params

static from_buffer(buffer)

Create an NvtxRangeStartAParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxRangeStartAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_range_start_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxRangeStartAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

message

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxRangeStartWParams

Bases: object

Empty-initialize an instance of nvtxRangeStartW_params.

See also

nvtxRangeStartW_params

static from_buffer(buffer)

Create an NvtxRangeStartWParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxRangeStartWParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_range_start_w_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxRangeStartWParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

message

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxRangeEndParams

Bases: object

Empty-initialize an instance of nvtxRangeEnd_params.

See also

nvtxRangeEnd_params

static from_buffer(buffer)

Create an NvtxRangeEndParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxRangeEndParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_range_end_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxRangeEndParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

id

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxRangePushExParams

Bases: object

Empty-initialize an instance of nvtxRangePushEx_params.

See also

nvtxRangePushEx_params

event_attrib

int:

static from_buffer(buffer)

Create an NvtxRangePushExParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxRangePushExParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_range_push_ex_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxRangePushExParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxRangePushAParams

Bases: object

Empty-initialize an instance of nvtxRangePushA_params.

See also

nvtxRangePushA_params

static from_buffer(buffer)

Create an NvtxRangePushAParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxRangePushAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_range_push_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxRangePushAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

message

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxRangePushWParams

Bases: object

Empty-initialize an instance of nvtxRangePushW_params.

See also

nvtxRangePushW_params

static from_buffer(buffer)

Create an NvtxRangePushWParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxRangePushWParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_range_push_w_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxRangePushWParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

message

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxRangePopParams

Bases: object

Empty-initialize an instance of nvtxRangePop_params.

See also

nvtxRangePop_params

dummy

int:

static from_buffer(buffer)

Create an NvtxRangePopParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxRangePopParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_range_pop_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxRangePopParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxNameCategoryAParams

Bases: object

Empty-initialize an instance of nvtxNameCategoryA_params.

See also

nvtxNameCategoryA_params

category

int:

static from_buffer(buffer)

Create an NvtxNameCategoryAParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCategoryAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_category_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCategoryAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxNameCategoryWParams

Bases: object

Empty-initialize an instance of nvtxNameCategoryW_params.

See also

nvtxNameCategoryW_params

category

int:

static from_buffer(buffer)

Create an NvtxNameCategoryWParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCategoryWParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_category_w_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCategoryWParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxNameOsThreadAParams

Bases: object

Empty-initialize an instance of nvtxNameOsThreadA_params.

See also

nvtxNameOsThreadA_params

static from_buffer(buffer)

Create an NvtxNameOsThreadAParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameOsThreadAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_os_thread_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameOsThreadAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

str:

ptr

Get the pointer address to the data as Python int.

thread_id

int:

class cupti.cupti.NvtxNameOsThreadWParams

Bases: object

Empty-initialize an instance of nvtxNameOsThreadW_params.

See also

nvtxNameOsThreadW_params

static from_buffer(buffer)

Create an NvtxNameOsThreadWParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameOsThreadWParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_os_thread_w_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameOsThreadWParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

int:

ptr

Get the pointer address to the data as Python int.

thread_id

int:

class cupti.cupti.NvtxNameCuDeviceAParams

Bases: object

Empty-initialize an instance of nvtxNameCuDeviceA_params.

See also

nvtxNameCuDeviceA_params

device_

int:

static from_buffer(buffer)

Create an NvtxNameCuDeviceAParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCuDeviceAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_cu_device_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCuDeviceAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxNameCuDeviceWParams

Bases: object

Empty-initialize an instance of nvtxNameCuDeviceW_params.

See also

nvtxNameCuDeviceW_params

device_

int:

static from_buffer(buffer)

Create an NvtxNameCuDeviceWParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCuDeviceWParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_cu_device_w_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCuDeviceWParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxNameCuContextAParams

Bases: object

Empty-initialize an instance of nvtxNameCuContextA_params.

See also

nvtxNameCuContextA_params

context

int:

static from_buffer(buffer)

Create an NvtxNameCuContextAParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCuContextAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_cu_context_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCuContextAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxNameCuContextWParams

Bases: object

Empty-initialize an instance of nvtxNameCuContextW_params.

See also

nvtxNameCuContextW_params

context

int:

static from_buffer(buffer)

Create an NvtxNameCuContextWParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCuContextWParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_cu_context_w_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCuContextWParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxNameCuStreamAParams

Bases: object

Empty-initialize an instance of nvtxNameCuStreamA_params.

See also

nvtxNameCuStreamA_params

static from_buffer(buffer)

Create an NvtxNameCuStreamAParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCuStreamAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_cu_stream_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCuStreamAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

str:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.NvtxNameCuStreamWParams

Bases: object

Empty-initialize an instance of nvtxNameCuStreamW_params.

See also

nvtxNameCuStreamW_params

static from_buffer(buffer)

Create an NvtxNameCuStreamWParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCuStreamWParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_cu_stream_w_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCuStreamWParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.NvtxNameCuEventAParams

Bases: object

Empty-initialize an instance of nvtxNameCuEventA_params.

See also

nvtxNameCuEventA_params

event

int:

static from_buffer(buffer)

Create an NvtxNameCuEventAParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCuEventAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_cu_event_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCuEventAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxNameCuEventWParams

Bases: object

Empty-initialize an instance of nvtxNameCuEventW_params.

See also

nvtxNameCuEventW_params

event

int:

static from_buffer(buffer)

Create an NvtxNameCuEventWParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCuEventWParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_cu_event_w_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCuEventWParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxNameCudaDeviceAParams

Bases: object

Empty-initialize an instance of nvtxNameCudaDeviceA_params.

See also

nvtxNameCudaDeviceA_params

device_

int:

static from_buffer(buffer)

Create an NvtxNameCudaDeviceAParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCudaDeviceAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_cuda_device_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCudaDeviceAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxNameCudaDeviceWParams

Bases: object

Empty-initialize an instance of nvtxNameCudaDeviceW_params.

See also

nvtxNameCudaDeviceW_params

device_

int:

static from_buffer(buffer)

Create an NvtxNameCudaDeviceWParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCudaDeviceWParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_cuda_device_w_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCudaDeviceWParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxNameCudaStreamAParams

Bases: object

Empty-initialize an instance of nvtxNameCudaStreamA_params.

See also

nvtxNameCudaStreamA_params

static from_buffer(buffer)

Create an NvtxNameCudaStreamAParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCudaStreamAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_cuda_stream_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCudaStreamAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

str:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.NvtxNameCudaStreamWParams

Bases: object

Empty-initialize an instance of nvtxNameCudaStreamW_params.

See also

nvtxNameCudaStreamW_params

static from_buffer(buffer)

Create an NvtxNameCudaStreamWParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCudaStreamWParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_cuda_stream_w_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCudaStreamWParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

int:

ptr

Get the pointer address to the data as Python int.

stream

int:

class cupti.cupti.NvtxNameCudaEventAParams

Bases: object

Empty-initialize an instance of nvtxNameCudaEventA_params.

See also

nvtxNameCudaEventA_params

event

int:

static from_buffer(buffer)

Create an NvtxNameCudaEventAParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCudaEventAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_cuda_event_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCudaEventAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxNameCudaEventWParams

Bases: object

Empty-initialize an instance of nvtxNameCudaEventW_params.

See also

nvtxNameCudaEventW_params

event

int:

static from_buffer(buffer)

Create an NvtxNameCudaEventWParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxNameCudaEventWParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_name_cuda_event_w_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxNameCudaEventWParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxDomainCreateAParams

Bases: object

Empty-initialize an instance of nvtxDomainCreateA_params.

See also

nvtxDomainCreateA_params

static from_buffer(buffer)

Create an NvtxDomainCreateAParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxDomainCreateAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_domain_create_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxDomainCreateAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

name

str:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxDomainDestroyParams

Bases: object

Empty-initialize an instance of nvtxDomainDestroy_params.

See also

nvtxDomainDestroy_params

domain

int:

static from_buffer(buffer)

Create an NvtxDomainDestroyParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxDomainDestroyParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_domain_destroy_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxDomainDestroyParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxDomainMarkExParams

Bases: object

Empty-initialize an instance of nvtxDomainMarkEx_params.

See also

nvtxDomainMarkEx_params

core

NvtxMarkExParams:

domain

int:

static from_buffer(buffer)

Create an NvtxDomainMarkExParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxDomainMarkExParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_domain_mark_ex_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxDomainMarkExParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxDomainRangeStartExParams

Bases: object

Empty-initialize an instance of nvtxDomainRangeStartEx_params.

See also

nvtxDomainRangeStartEx_params

core

NvtxRangeStartExParams:

domain

int:

static from_buffer(buffer)

Create an NvtxDomainRangeStartExParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxDomainRangeStartExParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_domain_range_start_ex_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxDomainRangeStartExParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxDomainRangeEndParams

Bases: object

Empty-initialize an instance of nvtxDomainRangeEnd_params.

See also

nvtxDomainRangeEnd_params

core

NvtxRangeEndParams:

domain

int:

static from_buffer(buffer)

Create an NvtxDomainRangeEndParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxDomainRangeEndParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_domain_range_end_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxDomainRangeEndParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxDomainRangePushExParams

Bases: object

Empty-initialize an instance of nvtxDomainRangePushEx_params.

See also

nvtxDomainRangePushEx_params

core

NvtxRangePushExParams:

domain

int:

static from_buffer(buffer)

Create an NvtxDomainRangePushExParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxDomainRangePushExParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_domain_range_push_ex_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxDomainRangePushExParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxDomainRangePopParams

Bases: object

Empty-initialize an instance of nvtxDomainRangePop_params.

See also

nvtxDomainRangePop_params

domain

int:

static from_buffer(buffer)

Create an NvtxDomainRangePopParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxDomainRangePopParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_domain_range_pop_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxDomainRangePopParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxSyncUserCreateParams

Bases: object

Empty-initialize an instance of nvtxSyncUserCreate_params.

See also

nvtxSyncUserCreate_params

attribs

int:

domain

int:

static from_buffer(buffer)

Create an NvtxSyncUserCreateParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxSyncUserCreateParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_sync_user_create_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxSyncUserCreateParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxSyncUserCommonParams

Bases: object

Empty-initialize an instance of nvtxSyncUserCommon_params.

See also

nvtxSyncUserCommon_params

static from_buffer(buffer)

Create an NvtxSyncUserCommonParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxSyncUserCommonParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_sync_user_common_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxSyncUserCommonParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

handle

int:

ptr

Get the pointer address to the data as Python int.

class cupti.cupti.NvtxDomainRegisterStringAParams

Bases: object

Empty-initialize an instance of nvtxDomainRegisterStringA_params.

See also

nvtxDomainRegisterStringA_params

domain

int:

static from_buffer(buffer)

Create an NvtxDomainRegisterStringAParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxDomainRegisterStringAParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_domain_register_string_a_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxDomainRegisterStringAParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

string

str:

class cupti.cupti.NvtxDomainRegisterStringWParams

Bases: object

Empty-initialize an instance of nvtxDomainRegisterStringW_params.

See also

nvtxDomainRegisterStringW_params

domain

int:

static from_buffer(buffer)

Create an NvtxDomainRegisterStringWParams instance with the memory from the given buffer.

static from_data(data)

Create an NvtxDomainRegisterStringWParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype nvtx_domain_register_string_w_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an NvtxDomainRegisterStringWParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

ptr

Get the pointer address to the data as Python int.

string

str:

PM Sampling API

Functions

cupti.cupti.pm_sampling_counter_data_get_sample_info(p_params)

Get the sample info (start and end time stamp) for the given sample index. Each sample is distinguished by the start and end time stamp.

Parameters:

p_params (intptr_t) – A pointer to CUpti_PmSampling_CounterData_GetSampleInfo_Params.

See also

cuptiPmSamplingCounterDataGetSampleInfo

cupti.cupti.pm_sampling_counter_data_image_initialize(p_params)

Initialize the counter data to CUPTI record format for storing the metric data.

Parameters:

p_params (intptr_t) – A pointer to CUpti_PmSampling_CounterDataImage_Initialize_Params.

See also

cuptiPmSamplingCounterDataImageInitialize

cupti.cupti.pm_sampling_decode_data(p_params)

Decode the metrics data stored in the hardware buffer to the counter data image.

Parameters:

p_params (intptr_t) – A pointer to CUpti_PmSampling_DecodeData_Params.

See also

cuptiPmSamplingDecodeData

cupti.cupti.pm_sampling_disable(p_params)

Disable PM sampling on the CUDA device and destroy the PM sampling object.

Parameters:

p_params (intptr_t) – A pointer to CUpti_PmSampling_Disable_Params.

See also

cuptiPmSamplingDisable

cupti.cupti.pm_sampling_enable(p_params)

Create a PM sampling object and enable PM sampling on the CUDA device.

Parameters:

p_params (intptr_t) – A pointer to CUpti_PmSampling_Enable_Params.

See also

cuptiPmSamplingEnable

cupti.cupti.pm_sampling_get_counter_availability(p_params)

Query counter availibility information in a buffer which can be used to filter unavailable raw metrics on host. Note: This API may fail, if any profiling or sampling session is active on the specified device.

Parameters:

p_params (intptr_t) – A pointer to CUpti_PmSampling_GetCounterAvailability_Params.

See also

cuptiPmSamplingGetCounterAvailability

cupti.cupti.pm_sampling_get_counter_data_info(p_params)

Get the counter data info like number of samples, number of populated samples and number of completed samples in a counter data image.

Parameters:

p_params (intptr_t) – A pointer to CUpti_PmSampling_GetCounterDataInfo_Params.

See also

cuptiPmSamplingGetCounterDataInfo

cupti.cupti.pm_sampling_get_counter_data_size(p_params)

Query the size of the counter data image which will be used to store the metrics data. User need to allocate the memory for the counter data image based on the size returned by this API.

Parameters:

p_params (intptr_t) – A pointer to CUpti_PmSampling_GetCounterDataSize_Params.

See also

cuptiPmSamplingGetCounterDataSize

cupti.cupti.pm_sampling_set_config(p_params)

Set the configuration for PM sampling like sampling interval, HW buffer size, trigger mode and the config image which has scheduling info for metric collection.

Parameters:

p_params (intptr_t) – A pointer to CUpti_PmSampling_SetConfig_Params.

See also

cuptiPmSamplingSetConfig

cupti.cupti.pm_sampling_start(p_params)

Start the PM sampling. The GPU will start collecting the metrics data periodically based on trigger type and sampling interval passed in CUpti_PmSampling_SetConfig_Params. The collected data will be stored in the hardware buffer.

Parameters:

p_params (intptr_t) – A pointer to CUpti_PmSampling_Start_Params.

See also

cuptiPmSamplingStart

cupti.cupti.pm_sampling_stop(p_params)

Stop the PM sampling. The GPU will stop collecting the metrics data.

Parameters:

p_params (intptr_t) – A pointer to CUpti_PmSampling_Stop_Params.

See also

cuptiPmSamplingStop

Enums

cupti.cupti.PmSamplingDecodeStopReason

alias of DecodeStopReason

cupti.cupti.PmSamplingHardwareBufferAppendMode

alias of HardwareBufferAppendMode

cupti.cupti.PmSamplingTriggerMode

alias of TriggerMode

Classes

class cupti.cupti.PmSamplingCounterDataImageInitializeParams

Bases: object

Empty-initialize an instance of CUpti_PmSampling_CounterDataImage_Initialize_Params.

See also

CUpti_PmSampling_CounterDataImage_Initialize_Params

counter_data_size

[in] Size of the counter data image.

Type:

int

static from_buffer(buffer)

Create an PmSamplingCounterDataImageInitializeParams instance with the memory from the given buffer.

static from_data(data)

Create an PmSamplingCounterDataImageInitializeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype pm_sampling_counter_data_image_initialize_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an PmSamplingCounterDataImageInitializeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_counter_data

[in] Counter data image.

Type:

int

p_pm_sampling_object

[in] PM sampling object.

Type:

int

p_priv

[in] Set to NULL.

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.PmSamplingCounterDataGetSampleInfoParams

Bases: object

Empty-initialize an instance of CUpti_PmSampling_CounterData_GetSampleInfo_Params.

See also

CUpti_PmSampling_CounterData_GetSampleInfo_Params

counter_data_image_size

[in] Size of the counter data image.

Type:

int

end_timestamp

[out] End time of the sample.

Type:

int

static from_buffer(buffer)

Create an PmSamplingCounterDataGetSampleInfoParams instance with the memory from the given buffer.

static from_data(data)

Create an PmSamplingCounterDataGetSampleInfoParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype pm_sampling_counter_data_get_sample_info_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an PmSamplingCounterDataGetSampleInfoParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_counter_data_image

[in] Counter data image.

Type:

int

p_pm_sampling_object

[in] PM sampling object.

Type:

int

p_priv

[in] Set to NULL.

Type:

int

ptr

Get the pointer address to the data as Python int.

sample_index

[in] Index of the sample.

Type:

int

start_timestamp

[out] Start time of the sample.

Type:

int

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.PmSamplingDecodeDataParams

Bases: object

Empty-initialize an instance of CUpti_PmSampling_DecodeData_Params.

See also

CUpti_PmSampling_DecodeData_Params

counter_data_image_size

[in] Size of the counter data image.

Type:

int

decode_stop_reason

[out] decode stop reason

Type:

int

static from_buffer(buffer)

Create an PmSamplingDecodeDataParams instance with the memory from the given buffer.

static from_data(data)

Create an PmSamplingDecodeDataParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype pm_sampling_decode_data_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an PmSamplingDecodeDataParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

overflow

[out] overflow status for hardware buffer. To avoid overflow, either increase the hardwareBufferSize value in CUpti_PmSampling_SetConfig_Params or increase the sampling interval.

Type:

int

p_counter_data_image

[in] Counter data image.

Type:

int

p_pm_sampling_object

[in] PM sampling object.

Type:

int

p_priv

[in] Set to NULL.

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.PmSamplingDisableParams

Bases: object

Empty-initialize an instance of CUpti_PmSampling_Disable_Params.

See also

CUpti_PmSampling_Disable_Params

static from_buffer(buffer)

Create an PmSamplingDisableParams instance with the memory from the given buffer.

static from_data(data)

Create an PmSamplingDisableParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype pm_sampling_disable_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an PmSamplingDisableParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_pm_sampling_object

[in] PM sampling object.

Type:

int

p_priv

[in] Set to NULL.

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.PmSamplingEnableParams

Bases: object

Empty-initialize an instance of CUpti_PmSampling_Enable_Params.

See also

CUpti_PmSampling_Enable_Params

device_index

[in] Device index.

Type:

int

static from_buffer(buffer)

Create an PmSamplingEnableParams instance with the memory from the given buffer.

static from_data(data)

Create an PmSamplingEnableParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype pm_sampling_enable_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an PmSamplingEnableParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_pm_sampling_object

[out] PM sampling object.

Type:

int

p_priv

[in] Set to NULL.

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.PmSamplingGetCounterAvailabilityParams

Bases: object

Empty-initialize an instance of CUpti_PmSampling_GetCounterAvailability_Params.

See also

CUpti_PmSampling_GetCounterAvailability_Params

counter_availability_image_size

[inout] Size of the counter availability image. When pCounterAvailabilityImage is NULL, this field is used to return the size of the counter availability image.

Type:

int

device_index

[in] Device index.

Type:

int

static from_buffer(buffer)

Create an PmSamplingGetCounterAvailabilityParams instance with the memory from the given buffer.

static from_data(data)

Create an PmSamplingGetCounterAvailabilityParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype pm_sampling_get_counter_availability_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an PmSamplingGetCounterAvailabilityParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_counter_availability_image

[out] Counter availability image.

Type:

int

p_priv

[in] Set to NULL.

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.PmSamplingGetCounterDataInfoParams

Bases: object

Empty-initialize an instance of CUpti_PmSampling_GetCounterDataInfo_Params.

See also

CUpti_PmSampling_GetCounterDataInfo_Params

counter_data_image_size

[in] Size of the counter data image.

Type:

int

static from_buffer(buffer)

Create an PmSamplingGetCounterDataInfoParams instance with the memory from the given buffer.

static from_data(data)

Create an PmSamplingGetCounterDataInfoParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype pm_sampling_get_counter_data_info_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an PmSamplingGetCounterDataInfoParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_completed_samples

[out] Number of samples that have been completed.

Type:

int

num_populated_samples

[out] Number of populated samples.

Type:

int

num_total_samples

[out] Number of samples in the counter data image.

Type:

int

p_counter_data_image

[in] Counter data image.

Type:

int

p_priv

[in] Set to NULL.

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.PmSamplingGetCounterDataSizeParams

Bases: object

Empty-initialize an instance of CUpti_PmSampling_GetCounterDataSize_Params.

See also

CUpti_PmSampling_GetCounterDataSize_Params

counter_data_size

[out] Size of the counter data image.

Type:

int

static from_buffer(buffer)

Create an PmSamplingGetCounterDataSizeParams instance with the memory from the given buffer.

static from_data(data)

Create an PmSamplingGetCounterDataSizeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype pm_sampling_get_counter_data_size_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an PmSamplingGetCounterDataSizeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

max_samples

[in] Maximum number of samples to be stored in the counter data image.

Type:

int

num_metrics

[in] Number of metrics to be collected.

Type:

int

p_metric_names

[in] Names of the metrics to be collected.

Type:

str

p_pm_sampling_object

[in] PM sampling object.

Type:

int

p_priv

[in] Set to NULL.

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.PmSamplingSetConfigParams

Bases: object

Empty-initialize an instance of CUpti_PmSampling_SetConfig_Params.

See also

CUpti_PmSampling_SetConfig_Params

config_size

[in] Size of the config image.

Type:

int

static from_buffer(buffer)

Create an PmSamplingSetConfigParams instance with the memory from the given buffer.

static from_data(data)

Create an PmSamplingSetConfigParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype pm_sampling_set_config_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an PmSamplingSetConfigParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

hardware_buffer_size

[in] The hardware buffer size in which raw PM sampling data will be stored. These samples will be decoded to counter data image with cuptiPmSamplingDecodeData call.

Type:

int

hw_buffer_append_mode

[in] Append mode for the records in hardware buffer. For KEEP_OLDEST mode, all the records will be kept in the buffer and in case hardware buffer is getting filled up. overflow will be set to 1 in CUpti_PmSampling_DecodeData_Params. For KEEP_LATEST mode, the new records will overwrite the oldest records in the buffer in case of filled buffer.

Type:

int

p_config

[in] Config image.

Type:

int

p_pm_sampling_object

[in] PM sampling object.

Type:

int

p_priv

[in] Set to NULL.

Type:

int

ptr

Get the pointer address to the data as Python int.

sampling_interval

[in] For the trigger mode CUPTI_PM_SAMPLING_TRIGGER_MODE_GPU_SYSCLK_INTERVAL, sampling interval is the number of sys clock cycles. For the trigger mode CUPTI_PM_SAMPLING_TRIGGER_MODE_GPU_TIME_INTERVAL, sampling interval is in nanoseconds.

Type:

int

struct_size

[in] Size of the data structure.

Type:

int

trigger_mode

CUPTI_PM_SAMPLING_TRIGGER_MODE_GPU_TIME_INTERVAL is not supported in Turing and GA100. Supported from GA10x onwards.

Type:

int

Type:

[in] Trigger mode. Note

class cupti.cupti.PmSamplingStartParams

Bases: object

Empty-initialize an instance of CUpti_PmSampling_Start_Params.

See also

CUpti_PmSampling_Start_Params

static from_buffer(buffer)

Create an PmSamplingStartParams instance with the memory from the given buffer.

static from_data(data)

Create an PmSamplingStartParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype pm_sampling_start_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an PmSamplingStartParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_pm_sampling_object

[in] PM sampling object.

Type:

int

p_priv

[in] Set to NULL.

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.PmSamplingStopParams

Bases: object

Empty-initialize an instance of CUpti_PmSampling_Stop_Params.

See also

CUpti_PmSampling_Stop_Params

static from_buffer(buffer)

Create an PmSamplingStopParams instance with the memory from the given buffer.

static from_data(data)

Create an PmSamplingStopParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype pm_sampling_stop_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an PmSamplingStopParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_pm_sampling_object

[in] PM sampling object.

Type:

int

p_priv

[in] Set to NULL.

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

Profiler Host API

Functions

cupti.cupti.profiler_host_config_add_metrics(p_params)

Add the metrics to the profiler host object for generating the config image. The config image will have the required information to schedule the metrics for collecting the profiling data. Note:.

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_Host_ConfigAddMetrics_Params.

See also

cuptiProfilerHostConfigAddMetrics

cupti.cupti.profiler_host_deinitialize(p_params)

Deinitialize and destroy the profiler host object (CUpti_Profiler_Host_Object).

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_Host_Deinitialize_Params.

See also

cuptiProfilerHostDeinitialize

cupti.cupti.profiler_host_evaluate_to_gpu_values(p_params)

Evaluate the metric values for the range index stored in the counter data.

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_Host_EvaluateToGpuValues_Params.

See also

cuptiProfilerHostEvaluateToGpuValues

cupti.cupti.profiler_host_get_base_metrics(p_params)

Get the list of supported base metrics for the chip.

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_Host_GetBaseMetrics_Params.

See also

cuptiProfilerHostGetBaseMetrics

cupti.cupti.profiler_host_get_config_image(p_params)

Get the config image for the metrics added to the profiler host object. User will pass the allocated buffer to store the config image.

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_Host_GetConfigImage_Params.

See also

cuptiProfilerHostGetConfigImage

cupti.cupti.profiler_host_get_config_image_size(p_params)

Get the size of the config image for the metrics added to the profiler host object. Users need to allocate the buffer for storing the config image.

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_Host_GetConfigImageSize_Params.

See also

cuptiProfilerHostGetConfigImageSize

cupti.cupti.profiler_host_get_max_num_hardware_metrics_per_pass(p_params)

Get the maximum number of hardware metrics (metric names which doesn’t include sass keyword) that can be scheduled in a single pass for a chip. While this represents a theoretical upper limit, practical constraints may prevent reaching this threshold for a specific set of metrics. Furthermore, the maximum achievable value is contingent upon the characteristics and architecture of the chip in question.

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_Host_GetMaxNumHardwareMetricsPerPass_Params .

See also

cuptiProfilerHostGetMaxNumHardwareMetricsPerPass

cupti.cupti.profiler_host_get_metric_properties(p_params)

Get the properties of the metric.

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_Host_GetMetricProperties_Params.

See also

cuptiProfilerHostGetMetricProperties

cupti.cupti.profiler_host_get_metrics_in_single_pass_set(p_params)

Get all the metrics defined in the single pass metric set. Profiling data for the metrics in a single pass metric set can be collected in a single pass.

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_Host_GetMetricsInSinglePassSet_Params.

See also

cuptiProfilerHostGetMetricsInSinglePassSet

cupti.cupti.profiler_host_get_num_of_passes(p_params)

Get the number of passes required for profiling the scheduled metrics in the config image.

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_Host_GetNumOfPasses_Params.

See also

cuptiProfilerHostGetNumOfPasses

cupti.cupti.profiler_host_get_range_name(p_params)

Get the range name for the range index stored in the counter data. In Range profiler, for Auto range mode the range name will be numeric value assigned to the kernel based on execution order. For user range mode, the name of range will be based on the range name provided by the user using Push range API.

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_Host_GetRangeName_Params.

See also

cuptiProfilerHostGetRangeName

cupti.cupti.profiler_host_get_single_pass_sets(p_params)

Get the single pass metric sets defined in the metric config file. Profiling data for the metrics in a single pass metric set can be collected in a single pass.

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_Host_GetSinglePassSets_Params when ppSinglePassSets is NULL, the function will return the number of single pass metric sets and the caller needs to allocate the buffer for the single pass metric sets using the number of single pass metric sets. (i.e. ppSinglePassSets).

See also

cuptiProfilerHostGetSinglePassSets

cupti.cupti.profiler_host_get_sub_metrics(p_params)

Get the list of supported sub-metrics for the metric.

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_Host_GetSubMetrics_Params.

See also

cuptiProfilerHostGetSubMetrics

cupti.cupti.profiler_host_get_supported_chips(p_params)

Get the list of supported chips.

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_Host_GetSupportedChips_Params.

See also

cuptiProfilerHostGetSupportedChips

cupti.cupti.profiler_host_initialize(p_params)

Create and initialize the profiler host object (CUpti_Profiler_Host_Object).

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_Host_Initialize_Params.

See also

cuptiProfilerHostInitialize

Enums

class cupti.cupti.MetricCollectionScope(value)

Bases: IntEnum

See CUpti_MetricCollectionScope.

CONTEXT = 0

DEVICE = 1

INVALID = 2

class cupti.cupti.MetricType(value)

Bases: IntEnum

See CUpti_MetricType.

COUNTER = 0

RATIO = 1

THROUGHPUT = 2

class cupti.cupti.MetricValueKind(value)

Bases: IntEnum

Kinds of metric values. Metric values can be one of several different kinds. Corresponding to each kind is a member of the CUpti_MetricValue union. The metric value returned by cuptiMetricGetValue should be accessed using the appropriate member of that union based on its value kind.

See CUpti_MetricValueKind.

DOUBLE = 0

FORCE_INT = 2147483647

INT64 = 4

NVTX_EXTENDED_PAYLOAD = 6

PERCENT = 2

THROUGHPUT = 3

UINT64 = 1

UTILIZATION_LEVEL = 5

class cupti.cupti.MetricValueUtilizationLevel(value)

Bases: IntEnum

Enumeration of utilization levels for metrics values of kind CUPTI_METRIC_VALUE_KIND_UTILIZATION_LEVEL. Utilization values can vary from IDLE (0) to MAX (10) but the enumeration only provides specific names for a few values.

See CUpti_MetricValueUtilizationLevel.

FORCE_INT = 2147483647

HIGH = 8

IDLE = 0

LOW = 2

MAX = 10

MID = 5

Classes

class cupti.cupti.ProfilerHostConfigAddMetricsParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Host_ConfigAddMetrics_Params.

See also

CUpti_Profiler_Host_ConfigAddMetrics_Params

static from_buffer(buffer)

Create an ProfilerHostConfigAddMetricsParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerHostConfigAddMetricsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_host_config_add_metrics_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerHostConfigAddMetricsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_metrics

[in] number of metrics

Type:

int

p_host_object

[in] reference to the profiler host object allocated by CUPTI in cuptiProfilerHostInitialize

Type:

int

p_priv

[in] Assign to NULL

Type:

int

pp_metric_names

[in] metric names for which config image will be generated

Type:

str

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.ProfilerHostDeinitializeParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Host_Deinitialize_Params.

See also

CUpti_Profiler_Host_Deinitialize_Params

static from_buffer(buffer)

Create an ProfilerHostDeinitializeParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerHostDeinitializeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_host_deinitialize_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerHostDeinitializeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_host_object

[in] reference to the profiler host object allocated by CUPTI in cuptiProfilerHostInitialize

Type:

int

p_priv

[in] Assign to NULL

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure. CUPTI client should set the size of the structure. It will be used in CUPTI to check what fields are available in the structure. Used to preserve backward compatibility.

Type:

int

class cupti.cupti.ProfilerHostEvaluateToGpuValuesParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Host_EvaluateToGpuValues_Params.

See also

CUpti_Profiler_Host_EvaluateToGpuValues_Params

counter_data_image_size

[in] size of counter data image

Type:

int

static from_buffer(buffer)

Create an ProfilerHostEvaluateToGpuValuesParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerHostEvaluateToGpuValuesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_host_evaluate_to_gpu_values_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerHostEvaluateToGpuValuesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_metrics

[in] number of metrics

Type:

int

p_counter_data_image

[in] the counter data image where profiling data has been decoded

Type:

int

p_host_object

[in] reference to the profiler host object allocated by CUPTI in cuptiProfilerHostInitialize

Type:

int

p_metric_values

[out] output value for given metric and range index

Type:

int

p_priv

[in] Assign to NULL

Type:

int

pp_metric_names

[in] the metrics for which GPU values will be evaluated for the range

Type:

str

ptr

Get the pointer address to the data as Python int.

range_index

[in] range index for which the range name will be queried

Type:

int

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.ProfilerHostGetBaseMetricsParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Host_GetBaseMetrics_Params.

See also

CUpti_Profiler_Host_GetBaseMetrics_Params

static from_buffer(buffer)

Create an ProfilerHostGetBaseMetricsParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerHostGetBaseMetricsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_host_get_base_metrics_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerHostGetBaseMetricsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

metric_type

[in] metric type (counter, ratio, throughput)

Type:

int

num_metrics

[out] number of metrics

Type:

int

p_host_object

[in] reference to the profiler host object allocated by CUPTI in cuptiProfilerHostInitialize

Type:

int

p_priv

[in] Assign to NULL

Type:

int

pp_metric_names

[out] list of base metrics supported of queried metric type for the chip

Type:

str

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.ProfilerHostGetConfigImageSizeParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Host_GetConfigImageSize_Params.

See also

CUpti_Profiler_Host_GetConfigImageSize_Params

config_image_size

[out] the size of config image, users need to allocate the buffer for storing

Type:

int

static from_buffer(buffer)

Create an ProfilerHostGetConfigImageSizeParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerHostGetConfigImageSizeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_host_get_config_image_size_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerHostGetConfigImageSizeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_host_object

[in] reference to the profiler host object allocated by CUPTI in cuptiProfilerHostInitialize

Type:

int

p_priv

[in] Assign to NULL

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.ProfilerHostGetConfigImageParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Host_GetConfigImage_Params.

See also

CUpti_Profiler_Host_GetConfigImage_Params

config_image_size

[in] Number of bytes allocated for pBuffer

Type:

int

static from_buffer(buffer)

Create an ProfilerHostGetConfigImageParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerHostGetConfigImageParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_host_get_config_image_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerHostGetConfigImageParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_config_image

[out] Buffer receiving the config image

Type:

int

p_host_object

[in] reference to the profiler host object allocated by CUPTI in cuptiProfilerHostInitialize

Type:

int

p_priv

[in] Assign to NULL

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.ProfilerHostGetMaxNumHardwareMetricsPerPassParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Host_GetMaxNumHardwareMetricsPerPass_Params.

See also

CUpti_Profiler_Host_GetMaxNumHardwareMetricsPerPass_Params

static from_buffer(buffer)

Create an ProfilerHostGetMaxNumHardwareMetricsPerPassParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerHostGetMaxNumHardwareMetricsPerPassParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_host_get_max_num_hardware_metrics_per_pass_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerHostGetMaxNumHardwareMetricsPerPassParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

max_metrics_per_pass

[out] maximum number of metrics that can be scheduled in a pass

Type:

int

p_chip_name

[in] accepted for chips supported at the time-of-release.

Type:

str

p_counter_availability_image

[in] buffer with counter availability image - required for future chip support

Type:

int

p_priv

[in] Assign to NULL

Type:

int

profiler_type

[in] the profiler kind one from CUpti_ProfilerType

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.ProfilerHostGetMetricPropertiesParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Host_GetMetricProperties_Params.

See also

CUpti_Profiler_Host_GetMetricProperties_Params

static from_buffer(buffer)

Create an ProfilerHostGetMetricPropertiesParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerHostGetMetricPropertiesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_host_get_metric_properties_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerHostGetMetricPropertiesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

metric_collection_scope

[out] the metric collection scope (context, device)

Type:

int

metric_type

[out] the metric type (counter, ratio or throughput)

Type:

int

p_description

[out] a short description about the metric

Type:

str

p_dim_unit

[out] the dimension of the metric values

Type:

str

p_host_object

[in] reference to the profiler host object allocated by CUPTI in cuptiProfilerHostInitialize

Type:

int

p_hw_unit

[out] associated hw unit for the metric

Type:

str

p_metric_name

[in] metric name for which its properties will be listed. Metric name can be with or without extension (rollup or submetric)

Type:

str

p_priv

[in] Assign to NULL

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.ProfilerHostGetMetricsInSinglePassSetParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Host_GetMetricsInSinglePassSet_Params.

See also

CUpti_Profiler_Host_GetMetricsInSinglePassSet_Params

static from_buffer(buffer)

Create an ProfilerHostGetMetricsInSinglePassSetParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerHostGetMetricsInSinglePassSetParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_host_get_metrics_in_single_pass_set_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerHostGetMetricsInSinglePassSetParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

metrics_buffer_size

[inout] In Query mode, this will be returned as the buffer size needed for the metrics in the single pass metric set In Data retrieval mode, this must be set to the value returned in the query mode

Type:

int

num_of_metrics_in_single_pass_set

[inout] In Query mode, this will be returned as the number of metrics in the single pass metric set In Data retrieval mode, this must be set to the value returned in the query mode

Type:

int

p_chip_name

[in] the chip name for which the metrics in the single pass metric set will be queried

Type:

str

p_metrics_buffer

[out] When set to NULL, the function call will treat it as query mode, otherwise it will treat it as data retrieval mode. User need to allocate the buffer for the metrics in the single pass metric set. The buffer size should be metricsBufferSize bytes.

Type:

int

p_metrics_indices_buffer

[out] When set to NULL, the function call will treat it as query mode, otherwise it will treat it as data retrieval mode. User need to allocate the buffer for the metrics indices in the single pass metric set. The buffer size should be numOfMetricsInSinglePassSet * sizeof(size_t).

Type:

int

p_priv

[in] Assign to NULL

Type:

int

p_single_pass_set_name

[in] the single pass metric set name for which the metrics will be queried

Type:

str

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.ProfilerHostGetNumOfPassesParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Host_GetNumOfPasses_Params.

See also

CUpti_Profiler_Host_GetNumOfPasses_Params

config_image_size

[in] Number of bytes allocated for pConfigImage

Type:

int

static from_buffer(buffer)

Create an ProfilerHostGetNumOfPassesParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerHostGetNumOfPassesParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_host_get_num_of_passes_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerHostGetNumOfPassesParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_of_passes

[out] number of passes required for profiling scheduled metrics in the config image

Type:

int

p_config_image

[in] the config image buffer

Type:

int

p_priv

[in] Assign to NULL

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.ProfilerHostGetRangeNameParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Host_GetRangeName_Params.

See also

CUpti_Profiler_Host_GetRangeName_Params

counter_data_image_size

[in] size of counter data image

Type:

int

delimiter

[in] used in case of nested ranges, default=”/”. Range1<delimiter>Range2

Type:

str

static from_buffer(buffer)

Create an ProfilerHostGetRangeNameParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerHostGetRangeNameParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_host_get_range_name_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerHostGetRangeNameParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_counter_data_image

[in] the counter data image where profiling data has been decoded

Type:

int

p_priv

[in] Assign to NULL

Type:

int

p_range_name

that the CUPTI allocate the memory internal and its user responsibility to free up the allocated memory

Type:

str

Type:

[out] the range name. Note

ptr

Get the pointer address to the data as Python int.

range_index

[in] range index for which the range name will be queried

Type:

int

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.ProfilerHostGetSinglePassSetsParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Host_GetSinglePassSets_Params.

See also

CUpti_Profiler_Host_GetSinglePassSets_Params

static from_buffer(buffer)

Create an ProfilerHostGetSinglePassSetsParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerHostGetSinglePassSetsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_host_get_single_pass_sets_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerHostGetSinglePassSetsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_of_single_pass_sets

[out] number of single pass metric sets

Type:

int

p_chip_name

[in] the chip name for which the single pass metric sets will be queried

Type:

str

p_priv

[in] Assign to NULL

Type:

int

pp_single_pass_sets

[out] list of single pass metric sets.

Type:

str

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.ProfilerHostGetSubMetricsParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Host_GetSubMetrics_Params.

See also

CUpti_Profiler_Host_GetSubMetrics_Params

static from_buffer(buffer)

Create an ProfilerHostGetSubMetricsParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerHostGetSubMetricsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_host_get_sub_metrics_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerHostGetSubMetricsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

metric_type

[in] the metric type for queried metric

Type:

int

num_of_submetrics

[out] number of submetrics supported

Type:

int

p_host_object

[in] reference to the profiler host object allocated by CUPTI in cuptiProfilerHostInitialize

Type:

int

p_metric_name

[in] metric name for which sub-metric will be listed. Metric name can be with or without extension (rollup or submetric)

Type:

str

p_priv

[in] Assign to NULL

Type:

int

pp_sub_metrics

[out] list of submetrics supported for the metric.

Type:

str

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.ProfilerHostGetSupportedChipsParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Host_GetSupportedChips_Params.

See also

CUpti_Profiler_Host_GetSupportedChips_Params

static from_buffer(buffer)

Create an ProfilerHostGetSupportedChipsParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerHostGetSupportedChipsParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_host_get_supported_chips_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerHostGetSupportedChipsParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

num_chips

[out] number of supported chips

Type:

int

p_priv

[in] Assign to NULL

Type:

int

pp_chip_names

[out] list of supported chips

Type:

str

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure.

Type:

int

class cupti.cupti.ProfilerHostInitializeParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Host_Initialize_Params.

See also

CUpti_Profiler_Host_Initialize_Params

static from_buffer(buffer)

Create an ProfilerHostInitializeParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerHostInitializeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_host_initialize_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerHostInitializeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_chip_name

[in] accepted for chips supported at the time-of-release.

Type:

str

p_counter_availability_image

[in] buffer with counter availability image - required for future chip support

Type:

int

p_host_object

[out] binary blob allocated by CUPTI and operations associated with this object.

Type:

int

p_priv

[in] Assign to NULL

Type:

int

p_single_pass_metric_set_name

[in] the single pass metric set name, single pass metric set supported for a chip can be found using cuptiProfilerHostGetSinglePassSets API. Only valid for PM sampling, this can be set to NULL for range profiler.

Type:

str

profiler_type

[in] the profiler kind one from CUpti_ProfilerType

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] Size of the data structure. CUPTI client should set the size of the structure. It will be used in CUPTI to check what fields are available in the structure. Used to preserve backward compatibility.

Type:

int

Profiling API

Functions

cupti.cupti.profiler_deinitialize(p_params: int)

DeInitializes the profiler interface

Parameters:

p_params (intptr_t) – A pointer to CUpti_Profiler_DeInitialize_Params.

See also

cuptiProfilerDeInitialize

cupti.cupti.profiler_initialize(p_params: int)

Initializes the profiler interface

Parameters:

p_params (intptr_t) – A pointer to ``CUpti_Profiler_Initialize_Params ``.

See also

cuptiProfilerInitialize

Enums

class cupti.cupti.ProfilerType(value)

Bases: IntEnum

See CUpti_ProfilerType.

PM_SAMPLING = 1

PROFILER_INVALID = 2

RANGE_PROFILER = 0

Classes

class cupti.cupti.ProfilerDeInitializeParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_DeInitialize_Params.

See also

CUpti_Profiler_DeInitialize_Params

static from_buffer(buffer)

Create an ProfilerDeInitializeParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerDeInitializeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_de_initialize_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerDeInitializeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_priv

[in] assign to NULL

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] CUpti_Profiler_DeInitialize_Params_STRUCT_SIZE

Type:

int

class cupti.cupti.ProfilerInitializeParams

Bases: object

Empty-initialize an instance of CUpti_Profiler_Initialize_Params.

See also

CUpti_Profiler_Initialize_Params

static from_buffer(buffer)

Create an ProfilerInitializeParams instance with the memory from the given buffer.

static from_data(data)

Create an ProfilerInitializeParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype profiler_initialize_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an ProfilerInitializeParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_priv

[in] assign to NULL

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in] CUpti_Profiler_Initialize_Params_STRUCT_SIZE

Type:

int

Device query

Functions

cupti.cupti.device_get_chip_name(p_params)

Classes

class cupti.cupti.DeviceGetChipNameParams

Bases: object

Empty-initialize an instance of CUpti_Device_GetChipName_Params.

See also

CUpti_Device_GetChipName_Params

device_index

[in]

Type:

int

static from_buffer(buffer)

Create an DeviceGetChipNameParams instance with the memory from the given buffer.

static from_data(data)

Create an DeviceGetChipNameParams instance wrapping the given NumPy array.

Parameters:

data (_numpy.ndarray) – a single-element array of dtype device_get_chip_name_params_dtype holding the data.

static from_ptr(

ptr: int,

readonly: bool = False,

owner=None,

)

Create an DeviceGetChipNameParams instance wrapping the given pointer.

Parameters:

• ptr (intptr_t) – pointer address as Python int to the data.

• owner (object) – The Python object that owns the pointer. If not provided, data will be copied.

• readonly (bool) – whether the data is read-only (to the user). default is False.

p_chip_name

[out]

Type:

str

p_priv

[in] assign to NULL

Type:

int

ptr

Get the pointer address to the data as Python int.

struct_size

[in]

Type:

int

Clock Control

Functions

cupti.cupti.clock_control_get_status(ctx: int) → int

Query the device-global GPU clock-control status.

Parameters:

ctx (intptr_t) – A valid CUDA context on the device whose clock status should be queried.

Returns:

Returns the device clock-control status.

Return type:

int

See also

cuptiClockControlGetStatus

cupti.cupti.clock_control_lock(ctx: int, mode: int)

Lock GPU clocks to the requested frequency.

Parameters:

• ctx (intptr_t) – A valid CUDA context on the device whose clocks should be locked.

• mode (ClockControlMode) – Clock control mode (see CUpti_ClockControl_Mode).

See also

cuptiClockControlLock

cupti.cupti.clock_control_unlock()

Unlock GPU clocks previously locked via cuptiClockControlLock.

See also

cuptiClockControlUnlock

Enums

class cupti.cupti.ClockControlMode(value)

Bases: IntEnum

Clock control modes. Determines the frequency to which GPU clocks are pinned for a profiling session. Locking clocks produces reproducible metric values across kernel launches and replay passes. Clock locking is device-global – affects all contexts on the GPU. It is not supported on platforms where the underlying driver clock-control operation is unavailable, such as MIG-partitioned devices. Clock control is currently supported on Windows and Linux desktop platforms; on all other platforms the clock-control APIs return CUPTI_ERROR_NOT_SUPPORTED. Tegra SoC GPUs do not support clock control either, and report the same error regardless of which platform build of CUPTI is used to profile them. To skip clock locking entirely, simply do not call cuptiClockControlLock. Numeric values start at 1 so that an uninitialized / zero-valued CUpti_ClockControl_Mode is rejected with CUPTI_ERROR_INVALID_PARAMETER rather than silently locking clocks.

See CUpti_ClockControl_Mode.

BASE = 1

BOOST = 2

FORCE_INT = 2147483647

class cupti.cupti.ClockControlStatus(value)

Bases: IntEnum

Device clock-control status. This status describes the device-global clock-control state reported by the driver for the GPU associated with a CUDA context. It does not describe CUPTI ownership. For example, CUPTI_CLOCK_CONTROL_STATUS_LOCKED can be returned for a lock created by CUPTI, by another profiling tool, or by an administrator through an external mechanism such as nvidia-smi. Tools can query this state before calling cuptiClockControlLock to make an informed policy decision. A common use case is MIG: CUPTI cannot create a new clock lock on MIG-partitioned devices, but a tool may still choose to run if this API reports that clocks are already externally locked. A status of CUPTI_CLOCK_CONTROL_STATUS_UNLOCKED only means the driver currently reports no explicit clock lock. It does not guarantee fixed, deterministic, or non-boosting clocks.

See CUpti_ClockControl_Status.

BOOST_DISABLED = 5

BOOST_ENABLED = 4

FORCE_INT = 2147483647

LOCKED = 2

LOCKED_TO_FLOOR = 3

UNLOCKED = 1

Result Codes and Exceptions

class cupti.cupti.Result(value)

Bases: IntEnum

CUPTI result codes. Error and result codes returned by CUPTI functions.

See CUptiResult.

ERROR_API_NOT_IMPLEMENTED = 11

ERROR_CDP_TRACING_NOT_SUPPORTED = 32

ERROR_CMP_DEVICE_NOT_SUPPORTED = 42

ERROR_CONFIDENTIAL_COMPUTING_NOT_SUPPORTED = 41

ERROR_CUDA_COMPILER_NOT_COMPATIBLE = 34

ERROR_DISABLED = 23

ERROR_FORCE_INT = 2147483647

ERROR_HARDWARE = 9

ERROR_HARDWARE_BUSY = 26

ERROR_HES_TRACE_NOT_SUPPORTED_ON_MPS = 47

ERROR_INSUFFICIENT_PRIVILEGES = 35

ERROR_INVALID_CHIP_NAME = 46

ERROR_INVALID_CONTEXT = 3

ERROR_INVALID_DEVICE = 2

ERROR_INVALID_EVENT_DOMAIN_ID = 4

ERROR_INVALID_EVENT_ID = 5

ERROR_INVALID_EVENT_NAME = 6

ERROR_INVALID_EVENT_VALUE = 22

ERROR_INVALID_HANDLE = 19

ERROR_INVALID_KIND = 21

ERROR_INVALID_METRIC_ID = 16

ERROR_INVALID_METRIC_NAME = 17

ERROR_INVALID_METRIC_VALUE = 25

ERROR_INVALID_MODULE = 24

ERROR_INVALID_OPERATION = 7

ERROR_INVALID_PARAMETER = 1

ERROR_INVALID_STREAM = 20

ERROR_LEGACY_PROFILER_NOT_SUPPORTED = 38

ERROR_MAX_LIMIT_REACHED = 12

ERROR_MIG_DEVICE_NOT_SUPPORTED = 43

ERROR_MULTIPLE_SUBSCRIBERS_NOT_SUPPORTED = 39

ERROR_NOT_COMPATIBLE = 14

ERROR_NOT_INITIALIZED = 15

ERROR_NOT_READY = 13

ERROR_NOT_SUPPORTED = 27

ERROR_OLD_PROFILER_API_INITIALIZED = 36

ERROR_OPENACC_UNDEFINED_ROUTINE = 37

ERROR_OUT_OF_MEMORY = 8

ERROR_PARAMETER_SIZE_NOT_SUFFICIENT = 10

ERROR_QUEUE_EMPTY = 18

ERROR_SLI_DEVICE_NOT_SUPPORTED = 44

ERROR_UM_PROFILING_NOT_SUPPORTED = 28

ERROR_UM_PROFILING_NOT_SUPPORTED_ON_DEVICE = 29

ERROR_UM_PROFILING_NOT_SUPPORTED_ON_NON_P2P_DEVICES = 30

ERROR_UM_PROFILING_NOT_SUPPORTED_WITH_MPS = 31

ERROR_UNKNOWN = 999

ERROR_VIRTUALIZED_DEVICE_INSUFFICIENT_PRIVILEGES = 40

ERROR_VIRTUALIZED_DEVICE_NOT_SUPPORTED = 33

ERROR_WSL_DEVICE_NOT_SUPPORTED = 45

SUCCESS = 0

class cupti.cupti.cuptiError(status: int)

Bases: Exception