onnxruntime

3a8611ea - avoid wait on notification whoes tensor is generated on the same device (#12778)

avoid wait on notification whoes tensor is generated on the same device (#12778) * avoid wait on notification whoes tensor is generated on the same device for case cpu kernel wait on cuda kernel's cpu output, don't need to wait on cuda notification * fix linux break Co-authored-by: Cheng Tang <chenta@microsoft.com> Co-authored-by: Cheng Tang <chenta@microsoft.com@orttrainingdev9.d32nl1ml4oruzj4qz3bqlggovf.px.internal.cloudapp.net>