TcpPort Reconnect should inform upper layer to start sending again

webrtc/p2p/base/tcpport.cc

Index: webrtc/p2p/base/tcpport.cc
diff --git a/webrtc/p2p/base/tcpport.cc b/webrtc/p2p/base/tcpport.cc
index 5c9e0425c80fc3d9ae138288d417ff41c08c7696..3360ce69fc101fea9e37c4f77e230cf9e060af91 100644
--- a/webrtc/p2p/base/tcpport.cc
+++ b/webrtc/p2p/base/tcpport.cc
@@ -355,8 +355,12 @@ int TCPConnection::GetError() {
 
 void TCPConnection::OnConnectionRequestResponse(ConnectionRequest* req,
                                                 StunMessage* response) {
-  // Once we receive a binding response, we are really writable, and not just
-  // pretending to be writable.
+  // If we're in the state of pretending to be writeable, we should inform the
+  // upper layer it's ready to send again as previous EWOULDLBLOCK from socket
+  // would have stopped the outgoing stream.
+  if (pretending_to_be_writable_) {
+    Connection::OnReadyToSend();

[pthatcher1, 2015/08/20 03:47:27]

If we fire OnReadyToSend() when we go from pretending to not pretending, should
we have a "not ready to send" signal that fires when we do the opposite (inside
of OnClose)?

[juberti1, 2015/08/20 05:33:00]

That is driven by EWOULDBLOCK being returned from Send().

But do we want to do this before letting Connection process the STUN response?
We might get called back with packets immediately.

[guoweis_webrtc, 2015/08/21 05:05:50]

Yes, this is possible. Moved this down

+  }
   pretending_to_be_writable_ = false;
   Connection::OnConnectionRequestResponse(req, response);
   ASSERT(write_state() == STATE_WRITABLE);