The nvmath.device module is experimental and potentially subject to future changes.

copy_wait

nvmath.device.copy_wait()

Creates a synchronization point. It has to be called after nvmath.device.copy() to ensure that the copy operation has completed before any subsequent operations are executed.

Refer to the cuBLASDx documentation for more details on how to use this function: api/other_tensors.html#cooperative-global-shared-copying