Re: What protects rcu_dereference() in __in6_dev_get()?

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

On Fri, Jan 15, 2010 at 06:50:15AM +0100, Eric Dumazet wrote:
> Le 14/01/2010 19:32, Paul E. McKenney a écrit :
> > Hello, Yoshi,
> > 
> > Could you please tell me what protects the rcu_dereference() in
> > __in6_dev_get()?  I am adding lockdep-based checking to RCU, and
> > "git blame" said I should ask you about this one.
> > 
> > The current code, rcu_dereference(), assumes that this is protected only
> > by RCU.  My problem might be any of the following:
> > 
> > o	Some other flavor of RCU protects this, e.g., RCU-bh, which
> > 	would require rcu_dereference_bh().
> > 
> > o	This is called from updates as well as from readers, and
> > 	some lock protects the updates.
> > 
> > o	This is called during initialization, when this pointer is
> > 	inaccessible to readers.
> > 	
> > Please note that I can add a check to cover multiple possibilities.
> > For a real example in include/linux/fdtable.h:
> > 
> > 	file = rcu_dereference_check(fdt->fd[fd],
> > 				     rcu_read_lock_held() ||
> > 				     lockdep_is_held(&files->file_lock) ||
> > 				     atomic_read(&files->count) == 1);
> > 
> > The first argument is the pointer, and the second argument says that
> > this may be protected by either RCU (as opposed to RCU-bh, RCU-sched,
> > or SRCU), the files->file_lock as recorded by lockdep, or by being in
> > a single-threaded process as noted by the value of files->count.
> > (Please see http://lwn.net/Articles/368683/ for a recent patch, another
> > will go out soon.)
> > 
> > So, could you please tell me what protects the rcu_dereference()
> > in __in6_dev_get() so that I can craft the appropriate form of
> > rcu_dereference()?
> > 
> > 							Thanx, Paul
> 
> Hi Paul
> 
> __in6_dev_get() is called either with rcu_read_lock()/rcu_read_unlock() protection,
> or with the RTNL mutex held.

Very good!  So I make a lockdep_rtnl_is_held() in net/core/rtnetlink.c:

	#ifdef CONFIG_PROVE_LOCKING
	int lockdep_rtnl_is_held(void)
	{
		return lockdep_is_held(&rtnl_mutex);
	}
	EXPORT_SYMBOL(lockdep_rtnl_is_held);
	#endif /* #ifdef CONFIG_PROVE_LOCKING */

Then I make __in6_dev_get() look as follows:

	static inline struct inet6_dev *
	__in6_dev_get(struct net_device *dev)
	{
		return rcu_dereference_check(dev->ip6_ptr,
					     rcu_read_lock_held() ||
					     lockdep_rtnl_is_held());
	}

Seem reasonable?

> Well, thats the theory, we could have some bugs of course :)

I know that feeling!  ;-)

						Thanx, Paul