On Mar 13, 2013, at 10:37 AM, Darrel Lewis <[email protected]> wrote:

> 
> On Mar 13, 2013, at 7:21 AM, Noel Chiappa wrote:
> 
>> 
>> This is exactly my impression too, and for that very reason.
>> 
>> Perhaps I'm confused, but it seems to me that pretty much the only thing this
>> block buys someone is that one can tell, simply by looking at such IPv6
>> addresses, that they are EIDs - one don't need to do a mapping lookup cycle

Well you do if you need to populate your map-cache with RLOCs. 

As I said, the reason for the /12 is to know who aren't EIDs (with a supported 
configuration knob) so legacy forwarding in an ITR or PITR does not drop 
packets while doing a non-fruitful lookup. 

>> to see if a mapping exists. And maybe there are some other things too (e.g.
>> the whole block can be anycast routed to the nearest PITR).

This is an added benefit yes. 

> I don't think you are confused.

Noel is definitely not confused. 

> I think, as long as we support the 'conversion' of existing prefixes to EID 
> prefixes*, then the complexity in the implementation to support this behavior 
> must exist, and the corresponding value of having some destinations you don't 
> have to ask the mapping system about is reduced.

Right - the knob allows you to wholesale know if the lookup will provide RLOCs. 

Dino

> 
> -Darrel 
> 
> * and I think supporting this is a MUST
> _______________________________________________
> lisp mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/lisp
_______________________________________________
lisp mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/lisp

Reply via email to