Re: [PATCH] lockdep: annotate nfs/nfsd in-kernel sockets

[Peter Zijlstra <a.p.zijlstra@chello.nl>]

On Mon, 2006-10-09 at 11:00 +1000, Neil Brown wrote:
> On Friday October 6, a.p.zijlstra@chello.nl wrote:
> > 
> > Stick NFS sockets in their own class to avoid some lockdep warnings.
> > NFS sockets are never exposed to user-space, and will hence not trigger
> > certain code paths that would otherwise pose deadlock scenarios.
> 
> I'm a bit bothered that the changelog entry does mention what sort of
> lockdep warning are begin avoided, 

These:
  http://lkml.org/lkml/2006/7/13/84

> and that 'svc_reclassify_socket'
> doesn't contain the work 'lock', yet is it really the locks that are
> being reclassified.

Hmm, good point, shall I do s/reclassify_socket/reclassify_sock_lock/ ?

>   However
> 
> Acked-by: NeilBrown <neilb@suse.de>
> 
> NeilBrown
> 
> > +
> > +static inline void svc_reclassify_socket(struct socket *sock)
> > +{
> > +	struct sock *sk = sock->sk;
> > +	BUG_ON(sk->sk_lock.owner != NULL);
> > +	switch (sk->sk_family) {
> > +		case AF_INET:
> > +			sock_lock_init_class_and_name(sk,
> > +				"slock-AF_INET-NFSD", &svc_slock_key[0],
> > +				"sk_lock-AF_INET-NFSD", &svc_key[0]);
> > +			break;
> > +
> > +		case AF_INET6:
> > +			sock_lock_init_class_and_name(sk,
> > +				"slock-AF_INET6-NFSD", &svc_slock_key[1],
> > +				"sk_lock-AF_INET6-NFSD", &svc_key[1]);
> > +			break;
> > +
> > +		default:
> > +			BUG();
> > +	}
> > +}