[Ns-bugs] [Bug 735] OLSR doesn't deliver packets to upper layers

code@nsnam.ece.gatech.edu code at nsnam.ece.gatech.edu
Tue Nov 3 05:50:47 PST 2009


http://www.nsnam.org/bugzilla/show_bug.cgi?id=735





--- Comment #4 from Tom Henderson <tomh at tomh.org>  2009-11-03 08:50:47 EDT ---
(In reply to comment #3)
>   Tom,
> 
>   thank you for clarifying this point.
> 
> > I would clarify instead that OLSR routing does not work outside of the
> > ListRouting context (which has been documented).  
> 
>   Please point me that documentation, I've definitely missed it. 

It is in the manual, in the OLSR section:
http://www.nsnam.org/docs/manual/manual_95.html#SEC161

but I don't see it now documented in src/routing/olsr code.  So I see how this
detail could be missed.

> 
> > This logic is similar but not identical to ipv4-list-routing-protocol.cc.  I
> > didn't copy the ipv4-list-routing-protocol.cc logic into olsr a while back
> > because I wasn't sure that it was a good idea to clone this type of code all
> > around the routing protocols.  Instead, I think that some kind of helper
> > function that encapsulated the above in one place (such as non-member
> > non-friend in ipv4.{h,cc}) could be reusable across protocols.  
> 
>   I didn't catch your resolution. Do you prefer to leave local delivery to list
> routing? 

No, I support your proposal to make all of these protocols operate standalone--
I just would prefer to factor out the common logic such as deciding whether an
address belongs to the node.

> If so -- please document this decision in ipv4-routing-protocol.h and
> I will remove local delivery code from AODV. If not, do you plan to write this
> helper function to be used by all routing protocols to locally deliver packets?
> 
>   For now I have AODV which can work standalone and OLSR which can't, this is
> not convenient.  

Agreed.  I will suggest a small revision and post it here.


-- 
Configure bugmail: http://www.nsnam.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.


More information about the Ns-bugs mailing list