RDS: TCP: avoid bad page reference in rds_tcp_listen_data_ready

As the existing comments in rds_tcp_listen_data_ready() indicate,
it is possible under some race-windows to get to this function with the
accept() socket. If that happens, we could run into a sequence whereby

   thread 1				thread 2

rds_tcp_accept_one() thread
sets up new_sock via ->accept().
The sk_user_data is now
sock_def_readable
					data comes in for new_sock,
					->sk_data_ready is called, and
					we land in rds_tcp_listen_data_ready
rds_tcp_set_callbacks()
takes the sk_callback_lock and
sets up sk_user_data to be the cp
					read_lock sk_callback_lock
					ready = cp
					unlock sk_callback_lock
					page fault on ready

In the above sequence, we end up with a panic on a bad page reference
when trying to execute (*ready)(). Instead we need to call
sock_def_readable() safely, which is what this patch achieves.

Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/tcp.c b/net/rds/tcp.c
index d24f6c1..b411bb7 100644 (file)
--- a/net/rds/tcp.c
+++ b/net/rds/tcp.c
@@ -551,6 +551,13 @@ static void rds_tcp_kill_sock(struct net *net)
        }
 }
 
+void *rds_tcp_listen_sock_def_readable(struct net *net)
+{
+       struct rds_tcp_net *rtn = net_generic(net, rds_tcp_netid);
+
+       return rtn->rds_tcp_listen_sock->sk->sk_user_data;
+}
+
 static int rds_tcp_dev_event(struct notifier_block *this,
                             unsigned long event, void *ptr)
 {

diff --git a/net/rds/tcp.h b/net/rds/tcp.h
index 1c3160f..9a1cc89 100644 (file)
--- a/net/rds/tcp.h
+++ b/net/rds/tcp.h
@@ -70,6 +70,7 @@ void rds_tcp_listen_stop(struct socket *);
 void rds_tcp_listen_data_ready(struct sock *sk);
 int rds_tcp_accept_one(struct socket *sock);
 int rds_tcp_keepalive(struct socket *sock);
+void *rds_tcp_listen_sock_def_readable(struct net *net);
 
 /* tcp_recv.c */
 int rds_tcp_recv_init(void);

diff --git a/net/rds/tcp_listen.c b/net/rds/tcp_listen.c
index ca975a2..73040e3 100644 (file)
--- a/net/rds/tcp_listen.c
+++ b/net/rds/tcp_listen.c
@@ -183,6 +183,8 @@ void rds_tcp_listen_data_ready(struct sock *sk)
         */
        if (sk->sk_state == TCP_LISTEN)
                rds_tcp_accept_work(sk);
+       else
+               ready = rds_tcp_listen_sock_def_readable(sock_net(sk));
 
 out:
        read_unlock_bh(&sk->sk_callback_lock);