[net2-wg] Packet interface

[Kyle Jamieson]

> Whether or not this is less difficult to deal with than if we allow 
> per-packet changes on the same SendPoolEntry really depends on the 
> data structure that is used to queue packets, but it doesn't seem obvious 
> that per-packet metadata adds complexity to the provider of SendPoolEntry.

I think Phil is referring to the act of changing a packet's urgent
status while it is owned by and pending transmission by the network
stack.  This is indeed a sticky mess, and I agree with him on this
point.


Kyle