[SE,XLA] Switch to using multiple streams in xla_device_context

Instead of having one stream for compute, host-to-device and device-to-host transfers, switch to having separate streams, just like the GPU does.
Add a se::Event field to XlaTensor to allow accurate inter-stream dependencies to be created.

As part of this:
 - Fix TransferManager::TransferLiteralFrom/ToDevice to correctly make generated substreams wait on their master stream.
 - Fix Stream::BlockHostUntilDone() to not block on or return substreams. This behavior is completely broken and not only nondeterministically returns substreams to the pool but causes indefinite hangs with the HostStream.

PiperOrigin-RevId: 203726543

1 files changed, 3 insertions, 0 deletions

diff --git a/tensorflow/stream_executor/event.h b/tensorflow/stream_executor/event.h
index 1f37262c78..9cc87a7c12 100644
--- a/tensorflow/stream_executor/event.h
+++ b/tensorflow/stream_executor/event.h
@@ -61,6 +61,9 @@ class Event {
   // Returns a pointer to the underlying platform-specific implementation.
   internal::EventInterface* implementation() { return implementation_.get(); }
 
+  Event(Event&&) = default;
+  Event& operator=(Event&&) = default;
+
  private:
   friend class Stream;