> The spec permits one IPid sequence per <src,dst,protocol> triple,
> since those parameters are used in matching fragments.  The downside
> is implementation complexity; it won't break anything anywhere on
> the net.  You also don't need to consume IPid space -- or at least,
> you don't need to worry about preventing duplicates -- on packets
> that have  set.
            ^ dont fragment ???

Actually there is is little point separating on the src address [1]

I also suspect that very few IP packets are fragmented anywhere other
than the source system (pppoe is one place where it could be done...) [2].
So you don't really need to worry too much about the IPid on unfragmented
packets.

	David

[1] Only relevant if the system has a large number of source addresses,
but only talks to a small number of remote addresses.

[2] fddi - ethernet and token-ring ethernet gateways could have to split
long frames, but I actually suspect there is little (ie only a few %)
benefit in using 17k frames on 16M token ring.  Do many systems restrict
the size to the ethernet limit anyway? (we didn't....)

-- 
David Laight: david@l8s.co.uk