On Sun, Oct 25, 2009 at 01:25:21PM +0200, Or Gerlitz wrote:

> well, you didn't address some of my comments (not the ice-cream 
> ones...), which come to say that this wouldn't be inter-operable if for 
> one side you convert INET/TCP to IB/IB and for the other side you don't 
> (e.g userA/userB user/kernel kernel/user etc schemes). 

I don't think there will be any compatability issues. Most IP CM
active side requirements are trivially met through user space
generation of the private data.

> Also the functionality added under the bonding scheme is lost, etc.

Considering that ACM gets rid of the ND process I don't see how full
bonding functionality could have ever been maintained. That said, I
think within what Sean has designed there could be something analogous
to IP bonding within ACM - features like this are why it is important
the name resolution have control over source device selection, not
just outgoing route.

> I am asking you to have INET/TCP apps enjoy both ACM's DGID and route 
> resolution without being converted to IB/IB, simple as that. If needed 
> I'd be happy to assist in making this flow happen.

If they are 100% interoperable, and the conversion is transparent to
apps that use this new rdma_getaddrinfo thingy, why do you care at all?

Jason
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to