[net2-wg] collection stuff

[Rodrigo Fonseca]

The tought was to avoid having the RoutingEngine as part of the data path.

I have a simpler suggestion:
 - there is no routing header
 - the BasicRouting interface becomes
         error_t command getNextHops(&nextHops, &len)
         uint8_t command getHopCount()

 All the info necessary for the data path lies in a header filled in
entirely by the FE.

The interaction between FE and RE is through the BasicRouting, for the
purposes of forwarding messages.

For address-free collection, I think that works well.

Thoughts?
Rodrigo

On 4/20/06, Philip Levis <pal at cs.stanford.edu> wrote:
>
> On Apr 20, 2006, at 7:34 PM, Rodrigo Fonseca wrote:
>
> > Hi Phil,
> >
> > just to clarify, you are talking solely about the data path, right?
> >
> > I am not sure we need anything in the Routing Header for simple
> > address free collection.
> > The only thing I can think of is the current hopcount, or the lowest
> > hopcount seen.
> >
> > The idea with the getNextHops command is that the FE just asks what
> > the addresses are for the next hop and adds them itself to the network
> > header.
> >
> > Notice that I am not considering the control messages at all, which
> > are totally separate.
> >
> > I agree that having send go to LE via the Routing will complicate the
> > Routing and may be unnecessary.
> >
> > All the FE needs to know it the size of the routing header if there is
> > one. This is what is in the RoutingEngineControl interface, but can be
> > changed.
> >
> >  interface RoutingEngineControl {
> >     command error_t initializeRH(message_t *msg, uint8_t tree_id);
> >     command uint8_t getHeaderSize();
> > }
> >
> > The FE.Packet interface takes the Routing header size in its
> > computations for the Packet interface.
>
>
> Why not follow the standard approach? It seems kind of weird to have
> this special "initializeRH" when you could just do it on the downward
> call path.
>
> Phil
>