Re: [PATCH] NFS: Fix RCU warnings in nfs_inode_return_delegation_noreclaim() [ver #2]

["Paul E. McKenney" <paulmck@linux.vnet.ibm.com>]

On Thu, Apr 01, 2010 at 12:45:14PM +0100, David Howells wrote:
> Paul E. McKenney <paulmck@linux.vnet.ibm.com> wrote:
> 
> > > I think it is incorrectly used.  Given that the rcu_dereference() in:
> > > 
> > > 	if (rcu_dereference(nfsi->delegation) != NULL) {
> > > 		spin_lock(&clp->cl_lock);
> > > 		delegation = nfs_detach_delegation_locked(nfsi, NULL);
> > > 		spin_unlock(&clp->cl_lock);
> > > 		if (delegation != NULL)
> > > 			nfs_do_return_delegation(inode, delegation, 0);
> > > 	}
> > 
> > And nfs_detach_delegation_locked() rechecks nfsi->delegation() under
> > the lock, so this is a legitimate use.
> > 
> > The pointer is not held constant, but any changes will be accounted
> > for and handled correctly.  So I would argue that the pointer value is
> > in fact protected by the recheck-under-lock algorithm used here.
> 
> A legitimate use of what?

A legitimate use of loading an RCU-protected pointer without
smp_read_barrier_depends().  However, I could imagine some situations
where the ACCESS_ONCE() semantics were required -- though in this
particular situation, I am having a hard time seeing how the compiler
could mess us up.  That said, my time on the C++ standards committee
has given me new respect for the perversity of compiler writers.

So you have objected to needless memory barriers.  How do you feel
about possibly needless ACCESS_ONCE() calls?

							Thanx, Paul