I think it's best to remove the timer tick from UdpCommandDemuxer.
I looked at it again and it doesn't actually help with responsiveness.
Whatever it contributes is no different from what stop() does.
They both just call timer.cancel and cmdsocket.cancel.
So if that doesn't stop the socket in stop(), it won't magically
stop it more effectively if I call it from a timer handler.
We no longer try to enable pcloud data as part of the connectReq()
sequence. Instead we separate them so that a device can be connected
but not be issuing pcloud data.
We were canceling the registered handler CBs in their branch
segments instead of in the unifying oracle fn.
This caused a bug where handlers were left behind in the
udpCmdDemuxer and thus a handler was invoked twice when a UDP
msg came in.
Sadly we don't get to immediately see the results of our
work because we have to do a unified dispatcher for the incoming
UDP messages on the command channel.
Async: Use new [Non]PostedAsyncCont and callOriginalCb
This new hierarchy of classes gives us a central mechanism for
managing both reply-posting and lockSpec unlocking.
* callOriginalCb: Now uses a modern C++ variadic template design
enabling it to handle both direct calling and std::bind()
re-binding of an arbitrary number of arguments from the caller.
This enables us to mostly eliminate the repeated, bespoke
definitions of callOriginalCb littered throughout the codebase.
We've also propagated these changes throughout the codebase in
this patch.
This class enables us to consistently represent continuations
that are intended to be posted on a particular target handling
thread. It hols a sh_ptr to the caller so that the target thread
can re-enqueue the response on the caller after processing the
REQ/IND op.
We tested it.
It's important to note that between test runs, we need to take
into account the fact that the Avia stops sending bcast adverts
when it's been handshaken.
So the retry-delay-ms may be longer due to the fact that the Avia
may not be sending adverts for a good portion of that retry-delay-ms
time.
Updated Boost dependency to version 1.73.0 to address segfault issues with boost::asio in dynamic libraries. Refactored heartbeat socket management to use raw UDP sockets instead of boost::asio, improving compatibility and error handling during socket operations.
Handshake: We must wait for the handshake ACK on the same port as
the one we sent the handshake REQ from.
Heartbeat: We must send the heartbeats from the same source port
as the one we sent the handshake REQ from.
hayodea
latentprion