tipc: fix race condition at topology server receive

We have identified a race condition during reception of socket
events and messages in the topology server.

- The function tipc_close_conn() is releasing the corresponding
  struct tipc_subscriber instance without considering that there
  may still be items in the receive work queue. When those are
  scheduled, in the function tipc_receive_from_work(), they are
  using the subscriber pointer stored in struct tipc_conn, without
  first checking if this is valid or not. This will sometimes
  lead to crashes, as the next call of tipc_conn_recvmsg() will
  access the now deleted item.
  We fix this by making the usage of this pointer conditional on
  whether the connection is active or not. I.e., we check the condition
  test_bit(CF_CONNECTED) before making the call tipc_conn_recvmsg().

- Since the two functions may be running on different cores, the
  condition test described above is not enough. tipc_close_conn()
  may come in between and delete the subscriber item after the condition
  test is done, but before tipc_conn_recv_msg() is finished. This
  happens less frequently than the problem described above, but leads
  to the same symptoms.

  We fix this by using the existing sk_callback_lock for mutual
  exclusion in the two functions. In addition, we have to move
  a call to tipc_conn_terminate() outside the mentioned lock to
  avoid deadlock.

Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 11 insertions, 10 deletions

diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c
index 44df528ed6ab..68e26470c516 100644
--- a/net/tipc/subscr.c
+++ b/net/tipc/subscr.c
@@ -289,17 +289,16 @@ static struct tipc_subscription *tipc_subscrp_create(struct net *net,
 	return sub;
 }
 
-static void tipc_subscrp_subscribe(struct net *net, struct tipc_subscr *s,
-				   struct tipc_subscriber *subscriber, int swap,
-				   bool status)
+static int tipc_subscrp_subscribe(struct net *net, struct tipc_subscr *s,
+				  struct tipc_subscriber *subscriber, int swap,
+				  bool status)
 {
-	struct tipc_net *tn = net_generic(net, tipc_net_id);
 	struct tipc_subscription *sub = NULL;
 	u32 timeout;
 
 	sub = tipc_subscrp_create(net, s, swap);
 	if (!sub)
-		return tipc_conn_terminate(tn->topsrv, subscriber->conid);
+		return -1;
 
 	spin_lock_bh(&subscriber->lock);
 	list_add(&sub->subscrp_list, &subscriber->subscrp_list);
@@ -313,6 +312,7 @@ static void tipc_subscrp_subscribe(struct net *net, struct tipc_subscr *s,
 
 	if (timeout != TIPC_WAIT_FOREVER)
 		mod_timer(&sub->timer, jiffies + msecs_to_jiffies(timeout));
+	return 0;
 }
 
 /* Handle one termination request for the subscriber */
@@ -322,9 +322,9 @@ static void tipc_subscrb_release_cb(int conid, void *usr_data)
 }
 
 /* Handle one request to create a new subscription for the subscriber */
-static void tipc_subscrb_rcv_cb(struct net *net, int conid,
-				struct sockaddr_tipc *addr, void *usr_data,
-				void *buf, size_t len)
+static int tipc_subscrb_rcv_cb(struct net *net, int conid,
+			       struct sockaddr_tipc *addr, void *usr_data,
+			       void *buf, size_t len)
 {
 	struct tipc_subscriber *subscriber = usr_data;
 	struct tipc_subscr *s = (struct tipc_subscr *)buf;
@@ -338,10 +338,11 @@ static void tipc_subscrb_rcv_cb(struct net *net, int conid,
 	/* Detect & process a subscription cancellation request */
 	if (s->filter & htohl(TIPC_SUB_CANCEL, swap)) {
 		s->filter &= ~htohl(TIPC_SUB_CANCEL, swap);
-		return tipc_subscrp_cancel(s, subscriber);
+		tipc_subscrp_cancel(s, subscriber);
+		return 0;
 	}
 	status = !(s->filter & htohl(TIPC_SUB_NO_STATUS, swap));
-	tipc_subscrp_subscribe(net, s, subscriber, swap, status);
+	return tipc_subscrp_subscribe(net, s, subscriber, swap, status);
 }
 
 /* Handle one request to establish a new subscriber */