[support] MCoA dynamic interfaces

[Romain KUNTZ]

Hi Vilmos,

On 2008/07/31, at 5:08, Vilmos Nebehaj wrote:
> please find attached a patch to support dynamic interfaces
> with MCoA.  It is based on the patch from Arnaud Ebalard
> (http://ml.nautilus6.org/pipermail/support/2007-December/000316.html).

Thanks for this patch, it's a very useful feature IMHO.

At the moment I'm thinking about reorganizing the MCoA development  
source versioning and switch to mercurial, which would allow me to put  
your patch on top of MCoA. I also plan to merge some of Arnaud's patch  
to NEPL. This to say that it may take little time before your patch  
appear but be sure that I'll take it into account.

> A question:
> - Why is the tunnel kept alive in mcoa_mn_chk_ho_verdict() around
>  line 3044?  For me it doesn't seem necessary.

I think we keep it alive to avoid to much tunnel deletion/addition  
(which is costly in terms of ioctl).

Cheers,
-- 
Romain KUNTZ
kuntz at lsiit.u-strasbg.fr
LSIIT - Networks and Protocols Team
http://clarinet.u-strasbg.fr/~kuntz/