[ofa-general] Re: Re: IPoIB-CM UC mode

[Michael S. Tsirkin]

> Quoting Or Gerlitz <ogerlitz at voltaire.com>:
> Subject: Re: [ofa-general] Re: Re: IPoIB-CM UC mode
> 
> Michael S. Tsirkin wrote:
> >>Quoting Or Gerlitz <ogerlitz at voltaire.com>:
> 
> >>Yes, I know that there is a way to register for kernel level neighbour 
> >>update events, so on each neighbour update, ipoib cm reconnects, plus 
> >>you can remove the fast path memcmp we do today on the remote GUID, and 
> >>we done :)
> 
> >In the typical case (remote side reboots) both the GID and the UD QPN stay 
> >the
> >same, so it seems there won't be any neighbour update, right?  If so, while
> >playing with neighbour update events might get us data path speed-up, it 
> >will
> >not solve the problem of detecting the connection is alive.
> 
> I don't think we should give up here, first there might be a way (event) 
> and if not lets change the kernel :) to know that the neighbouring 
> subsystem issued a broadcast arp on a nieghbour.
> Second, let me think...

Frankly, I like the idea of using our own keepalive better: it will also
work if we have e.g. multiple connections per neighbour.

> What did the people who wrote the RFC said about the need / 
> implementation of liveness protocol?

That it's a general IB problem and should be addressed at IB level.
Which it seems to be - with CM.

-- 
MST