4 ms·
You would get all kinds of interesting packet reordering issues, which can cause performance problems all their own. I've sometimes contemplated just faking th
by zrm 8y ago
You would get all kinds of interesting packet reordering issues, which can cause performance problems all their own.
I've sometimes contemplated just faking the TCP protocol entirely. Do the TCP handshake, then send "TCP packets" but interpret each payload as a datagram, acknowledge everything up to the latest sequence number regardless of whether any of them were lost and don't bother with any flow control or retransmissions.
It's a terrible hack but it would probably work at least some of the time.
- caf 8y agoIn theory the re-ordering should only happen in the presence of packetloss, which is exactly what you want. Could be worth an experiment, anyway. Your suggestion is how some "WAN optimisation" network middleboxes work (transparently terminating the TCP in the middlebox itself). To do that with Wireguard you'd have to implement that in the Linux kernel.
- rahimnathwani 8y agovdm further down thread pointed at a repo which does something like this: https://github.com/wangyu-/udp2raw-tunnel/blob/master/README.md https://github.com/wangyu-/udp2raw-tunnel/blob/master/README...