cuda.bindings._v2.driver.GPUDirectRDMAWritesOrdering

class cuda.bindings._v2.driver.GPUDirectRDMAWritesOrdering(value: int)

Platform native ordering for GPUDirect RDMA writes

See CUGPUDirectRDMAWritesOrdering.

NONE: 0

The device does not natively support ordering of remote writes. cuFlushGPUDirectRDMAWrites() can be leveraged if supported.

OWNER: 100

Natively, the device can consistently consume remote writes, although other CUDA devices may not.

ALL_DEVICES: 200

Any CUDA device in the system can consistently consume remote writes to this device.