On (03/05/08 15:10), Peter Memishian wrote:
> 
>  > > This all seems reasonable for what's proposed, but I wonder how it might
>  > > be extended to handle an additional case that's come up in another 
> thread.
>  > > In particular, ndd currently allows one to effectively change the default
>  > > value,
>  > 
>  > Not sure I follow- could you provide an example? At least for datalink
>  > drivers, afaik, ndd allows one to set or get values (typically the MII
>  > parameters), and the "default" here is intended to be the value 
>  > that the driver would come up with when it attaches.
> 
> I'm not talking specifically about drivers, but about ndd in general.  As
> we were recently discussing, most of the IP ndd tunables effectively
> change the per-stack default.  I think that same concept has merit at the
> datalink layer and as such I was hoping our design could account for that

I think there are a lot of items that may not be transferrable to the
TCP/IP model.  It's still not clear if all the TCP/IP tunables 
can really be tied to a linkid, for example. Moreover, the ndd model
is still just the simple pair of  set and  get operations, and 
to reset a default paramter "foo", one needs a parameter called
"foo-default". 

> (But given that this is consolidation private at this
> point, we have time to evolve this.)

right.

--Sowmini


Reply via email to