Subject: Re: [PATCH] ipv4: fix the rcu race between free_fib_info and
 ip_route_output_slow
From: Eric Dumazet <eric.dumazet@gmail.com>
To: Yanmin Zhang <yanmin_zhang@linux.intel.com>
Cc: David Miller <davem@davemloft.net>, kunx.jiang@intel.com,
        netdev@vger.kernel.org, linux-kernel@vger.kernel.org
In-Reply-To: <1337748897.14538.184.camel@ymzhang.sh.intel.com>
References: <4FBB6105.2060808@intel.com>
	 <20120522.151554.106838106733194160.davem@davemloft.net>
	 <1337742123.14538.175.camel@ymzhang.sh.intel.com>
	 <20120522.232310.911242148705021745.davem@davemloft.net>
	 <1337747829.3361.1599.camel@edumazet-glaptop>
	 <1337748897.14538.184.camel@ymzhang.sh.intel.com>
Content-Type: text/plain; charset="UTF-8"
Date: Wed, 23 May 2012 07:02:19 +0200
Message-ID: <1337749339.3361.1655.camel@edumazet-glaptop>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 881
Lines: 35

On Wed, 2012-05-23 at 12:54 +0800, Yanmin Zhang wrote:

> > fi->fib_dev (aka fib_nh[0].nh_dev) need full RCU protection.
> The new patch posted at https://lkml.org/lkml/2012/5/22/558 does move the
> resetting to RCU protection.

Its not enough.

We must take care that all users are in a RCU protected region.

They might be already, but a full check is needed.

For example 

net/ipv4/fib_trie.c:2563:    fi->fib_dev ? fi->fib_dev->name : "*"

looks to be safe (because already in a rcu_read_lock())

But its not.

Right thing would be to do :

struct net_device *ndev = rcu_dereference(fi->fib_dev)

	...
	ndev ? ndev->name : "*"


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/