[NETNS]: Namespace stop vs 'ip r l' race.

During network namespace stop process kernel side netlink sockets
belonging to a namespace should be closed. They should not prevent
namespace to stop, so they do not increment namespace usage
counter. Though this counter will be put during last sock_put.

The raplacement of the correct netns for init_ns solves the problem
only partial as socket to be stoped until proper stop is a valid
netlink kernel socket and can be looked up by the user processes. This
is not a problem until it resides in initial namespace (no processes
inside this net), but this is not true for init_net.

So, hold the referrence for a socket, remove it from lookup tables and
only after that change namespace and perform a last put.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Tested-by: Alexey Dobriyan <adobriyan@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 2 insertions, 13 deletions

diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 02cf848f71d..ddbdde82a70 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -1368,25 +1368,14 @@ static int rtnetlink_net_init(struct net *net)
 				   rtnetlink_rcv, &rtnl_mutex, THIS_MODULE);
 	if (!sk)
 		return -ENOMEM;
-
-	/* Don't hold an extra reference on the namespace */
-	put_net(sk->sk_net);
 	net->rtnl = sk;
 	return 0;
 }
 
 static void rtnetlink_net_exit(struct net *net)
 {
-	struct sock *sk = net->rtnl;
-	if (sk) {
-		/* At the last minute lie and say this is a socket for the
-		 * initial network namespace.  So the socket will be safe to
-		 * free.
-		 */
-		sk->sk_net = get_net(&init_net);
-		netlink_kernel_release(net->rtnl);
-		net->rtnl = NULL;
-	}
+	netlink_kernel_release(net->rtnl);
+	net->rtnl = NULL;
 }
 
 static struct pernet_operations rtnetlink_net_ops = {