Re: [PATCH] Fix RCU warning in rt_cache_seq_show

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

On Tue, Aug 09, 2011 at 07:18:56PM +0200, Eric Dumazet wrote:
> Le mardi 09 août 2011 à 18:02 +0100, Mark Rutland a écrit :
> > Commit f2c31e32 ("net: fix NULL dereferences in check_peer_redir()")
> > added rcu protection to dst neighbour, and updated callsites for
> > dst_{get,set}_neighbour. Unfortunately, it missed rt_cache_seq_show.
> > 
> > This produces a warning on v3.1-rc1 (on a preemptible kernel, on an
> > ARM Vexpress A9x4):
> > 
> > ===================================================
> > [ INFO: suspicious rcu_dereference_check() usage. ]
> > ---------------------------------------------------
> > include/net/dst.h:91 invoked rcu_dereference_check() without protection!
> > 
> > other info that might help us debug this:
> > 
> > rcu_scheduler_active = 1, debug_locks = 0
> > 2 locks held by proc01/32159:

This is very strange.  It says that there are two locks held by the
task, but refuses to list them.  Maybe something stomped on the list
of held locks?

							Thanx, Paul

> > stack backtrace:
> > [<80014880>] (unwind_backtrace+0x0/0xf8) from [<802e5c78>] (rt_cache_seq_show+0x18c/0x1c4)
> > [<802e5c78>] (rt_cache_seq_show+0x18c/0x1c4) from [<800e0c5c>] (seq_read+0x324/0x4a4)
> > [<800e0c5c>] (seq_read+0x324/0x4a4) from [<8010786c>] (proc_reg_read+0x70/0x94)
> > [<8010786c>] (proc_reg_read+0x70/0x94) from [<800c0ba8>] (vfs_read+0xb0/0x144)
> > [<800c0ba8>] (vfs_read+0xb0/0x144) from [<800c0ea8>] (sys_read+0x40/0x70)
> > [<800c0ea8>] (sys_read+0x40/0x70) from [<8000e0c0>] (ret_fast_syscall+0x0/0x3c)
> > 
> > This patch adds calls to rcu_read_{lock,unlock} in rt_cache_seq_show,
> > protecting the dereferenced variable, and clearing the warning.
> > 
> > Signed-off-by: Mark Rutland <mark.rutland@arm.com>
> > Cc: David S. Miller <davem@davemloft.net>
> > Cc: Eric Dumazet <eric.dumazet@gmail.com>
> > Cc: Gergely Kalman <synapse@hippy.csoma.elte.hu>
> > ---
> >  net/ipv4/route.c |    2 ++
> >  1 files changed, 2 insertions(+), 0 deletions(-)
> > 
> > diff --git a/net/ipv4/route.c b/net/ipv4/route.c
> > index e3dec1c..6699ef7 100644
> > --- a/net/ipv4/route.c
> > +++ b/net/ipv4/route.c
> > @@ -419,6 +419,7 @@ static int rt_cache_seq_show(struct seq_file *seq, void *v)
> >  		struct neighbour *n;
> >  		int len;
> >  
> > +		rcu_read_lock();
> >  		n = dst_get_neighbour(&r->dst);
> >  		seq_printf(seq, "%s\t%08X\t%08X\t%8X\t%d\t%u\t%d\t"
> >  			      "%08X\t%d\t%u\t%u\t%02X\t%d\t%1d\t%08X%n",
> > @@ -435,6 +436,7 @@ static int rt_cache_seq_show(struct seq_file *seq, void *v)
> >  			-1,
> >  			(n && (n->nud_state & NUD_CONNECTED)) ? 1 : 0,
> >  			r->rt_spec_dst, &len);
> > +		rcu_read_unlock();
> >  
> >  		seq_printf(seq, "%*s\n", 127 - len, "");
> >  	}
> 
> 
> Hmm, I though rcu_read_lock_bh() (done by caller of this function) was
> protecting us here.
> 
> 
> Paul, is it really needed, or is it a lockdep artifact ?
> 
> 
> 
> 
>