tipc: resolve race problem at unicast message reception

TIPC handles message cardinality and sequencing at the link layer,
before passing messages upwards to the destination sockets. During the
upcall from link to socket no locks are held. It is therefore possible,
and we see it happen occasionally, that messages arriving in different
threads and delivered in sequence still bypass each other before they
reach the destination socket. This must not happen, since it violates
the sequentiality guarantee.

We solve this by adding a new input buffer queue to the link structure.
Arriving messages are added safely to the tail of that queue by the
link, while the head of the queue is consumed, also safely, by the
receiving socket. Sequentiality is secured per socket by only allowing
buffers to be dequeued inside the socket lock. Since there may be multiple
simultaneous readers of the queue, we use a 'filter' parameter to reduce
the risk that they peek the same buffer from the queue, hence also
reducing the risk of contention on the receiving socket locks.

This solves the sequentiality problem, and seems to cause no measurable
performance degradation.

A nice side effect of this change is that lock handling in the functions
tipc_rcv() and tipc_bcast_rcv() now becomes uniform, something that
will enable future simplifications of those functions.

Reviewed-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, 21 insertions, 12 deletions

diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c
index dd8564cd9dbb..fcb07915aaac 100644
--- a/net/tipc/name_distr.c
+++ b/net/tipc/name_distr.c
@@ -381,25 +381,34 @@ void tipc_named_process_backlog(struct net *net)
 }
 
 /**
- * tipc_named_rcv - process name table update message sent by another node
+ * tipc_named_rcv - process name table update messages sent by another node
  */
-void tipc_named_rcv(struct net *net, struct sk_buff *buf)
+void tipc_named_rcv(struct net *net, struct sk_buff_head *inputq)
 {
 	struct tipc_net *tn = net_generic(net, tipc_net_id);
-	struct tipc_msg *msg = buf_msg(buf);
-	struct distr_item *item = (struct distr_item *)msg_data(msg);
-	u32 count = msg_data_sz(msg) / ITEM_SIZE;
-	u32 node = msg_orignode(msg);
+	struct tipc_msg *msg;
+	struct distr_item *item;
+	uint count;
+	u32 node;
+	struct sk_buff *skb;
+	int mtype;
 
 	spin_lock_bh(&tn->nametbl_lock);
-	while (count--) {
-		if (!tipc_update_nametbl(net, item, node, msg_type(msg)))
-			tipc_named_add_backlog(item, msg_type(msg), node);
-		item++;
+	for (skb = skb_dequeue(inputq); skb; skb = skb_dequeue(inputq)) {
+		msg = buf_msg(skb);
+		mtype = msg_type(msg);
+		item = (struct distr_item *)msg_data(msg);
+		count = msg_data_sz(msg) / ITEM_SIZE;
+		node = msg_orignode(msg);
+		while (count--) {
+			if (!tipc_update_nametbl(net, item, node, mtype))
+				tipc_named_add_backlog(item, mtype, node);
+			item++;
+		}
+		kfree_skb(skb);
+		tipc_named_process_backlog(net);
 	}
-	tipc_named_process_backlog(net);
 	spin_unlock_bh(&tn->nametbl_lock);
-	kfree_skb(buf);
 }
 
 /**