CUlaunchAttributeValue

Defined in /dvs/p4/build/sw/gpgpu/bin/x86_64_Linux_release/stub/cuda.h

union CUlaunchAttributeValue

Launch attributes union; used as value field of CUlaunchAttribute.

Public Members

char pad[64]

CUaccessPolicyWindow accessPolicyWindow

Value of launch attribute CU_LAUNCH_ATTRIBUTE_ACCESS_POLICY_WINDOW.

int cooperative

Value of launch attribute CU_LAUNCH_ATTRIBUTE_COOPERATIVE.

Nonzero indicates a cooperative kernel (see cuLaunchCooperativeKernel).

CUsynchronizationPolicy syncPolicy

Value of launch attribute CU_LAUNCH_ATTRIBUTE_SYNCHRONIZATION_POLICY.

CUsynchronizationPolicy for work queued up in this stream

unsigned int x

unsigned int y

unsigned int z

struct CUlaunchAttributeValue::[anonymous] clusterDim

Value of launch attribute CU_LAUNCH_ATTRIBUTE_CLUSTER_DIMENSION that represents the desired cluster dimensions for the kernel.

Opaque type with the following fields:

• x - The X dimension of the cluster, in blocks. Must be a divisor of the grid X dimension.

• y - The Y dimension of the cluster, in blocks. Must be a divisor of the grid Y dimension.

• z - The Z dimension of the cluster, in blocks. Must be a divisor of the grid Z dimension.

CUclusterSchedulingPolicy clusterSchedulingPolicyPreference

Value of launch attribute CU_LAUNCH_ATTRIBUTE_CLUSTER_SCHEDULING_POLICY_PREFERENCE.

Cluster scheduling policy preference for the kernel.

int programmaticStreamSerializationAllowed

Value of launch attribute CU_LAUNCH_ATTRIBUTE_PROGRAMMATIC_STREAM_SERIALIZATION.

CUevent event

int flags

int triggerAtBlockStart

struct CUlaunchAttributeValue::[anonymous] programmaticEvent

Value of launch attribute CU_LAUNCH_ATTRIBUTE_PROGRAMMATIC_EVENT with the following fields:

• CUevent event - Event to fire when all blocks trigger it.

• Event record flags, see cuEventRecordWithFlags. Does not accept :CU_EVENT_RECORD_EXTERNAL.

• triggerAtBlockStart - If this is set to non-0, each block launch will automatically trigger the event.

struct CUlaunchAttributeValue::[anonymous] launchCompletionEvent

Value of launch attribute CU_LAUNCH_ATTRIBUTE_LAUNCH_COMPLETION_EVENT with the following fields:

• CUevent event - Event to fire when the last block launches

• int flags; - Event record flags, see cuEventRecordWithFlags. Does not accept CU_EVENT_RECORD_EXTERNAL.

int priority

Value of launch attribute CU_LAUNCH_ATTRIBUTE_PRIORITY.

Execution priority of the kernel.

CUlaunchMemSyncDomainMap memSyncDomainMap

Value of launch attribute CU_LAUNCH_ATTRIBUTE_MEM_SYNC_DOMAIN_MAP.

See CUlaunchMemSyncDomainMap.

CUlaunchMemSyncDomain memSyncDomain

Value of launch attribute CU_LAUNCH_ATTRIBUTE_MEM_SYNC_DOMAIN.

See::CUlaunchMemSyncDomain

struct CUlaunchAttributeValue::[anonymous] preferredClusterDim

Value of launch attribute CU_LAUNCH_ATTRIBUTE_PREFERRED_CLUSTER_DIMENSION that represents the desired preferred cluster dimensions for the kernel.

Opaque type with the following fields:

• x - The X dimension of the preferred cluster, in blocks. Must be a divisor of the grid X dimension, and must be a multiple of the x field of CUlaunchAttributeValue::clusterDim.

• y - The Y dimension of the preferred cluster, in blocks. Must be a divisor of the grid Y dimension, and must be a multiple of the y field of CUlaunchAttributeValue::clusterDim.

• z - The Z dimension of the preferred cluster, in blocks. Must be equal to the z field of CUlaunchAttributeValue::clusterDim.

int deviceUpdatable

CUgraphDeviceNode devNode

struct CUlaunchAttributeValue::[anonymous] deviceUpdatableKernelNode

Value of launch attribute CU_LAUNCH_ATTRIBUTE_DEVICE_UPDATABLE_KERNEL_NODE.

with the following fields:

• int deviceUpdatable - Whether or not the resulting kernel node should be device-updatable.

• CUgraphDeviceNode devNode - Returns a handle to pass to the various device-side update functions.

unsigned int sharedMemCarveout

Value of launch attribute CU_LAUNCH_ATTRIBUTE_PREFERRED_SHARED_MEMORY_CARVEOUT.

void *commObj

CUhostCommunicationsVtable *commVtable

unsigned int localCount

void *localId

CUmultiKernelCooperativeDomain domain

struct CUlaunchAttributeValue::[anonymous] multiKernelCooperativeLaunch

Value of the launch attribute CU_LAUNCH_ATTRIBUTE_MULTI_KERNEL_COOPERATIVE_LAUNCH with the following fields:

• commObj - The communicator object (see CUhostCommunicationsVtable)

• commVtable - The vtable of the communicator object

• localCount - How many participants of the cooperative launch to expect in the local process. In-process coordination will be handled locally by CUDA. Calls to the communicator object are only made for inter-process operations and each operation is initiated only once per process.

• localId - This value is used to match calls from local participants. Calls for different collective launches that may race must use different IDs.

• domain - The domain the cooperative launch belongs to. See CUmultiKernelCooperativeDomain. This parameter is not nullable, in contrast to cuMultiKernelCooperativeDomainCreate.

unsigned int nvlinkUtilCentricScheduling

CUlaunchAttributePortableClusterMode portableClusterSizeMode

Value of launch attribute CU_LAUNCH_ATTRIBUTE_PORTABLE_CLUSTER_SIZE_MODE.

CUsharedMemoryMode sharedMemoryMode

Value of launch attribute CU_LAUNCH_ATTRIBUTE_SHARED_MEMORY_MODE.

See CUsharedMemoryMode for acceptable values.