diff mbox

[net-next] bridge: fix seq check in br_mdb_dump()

Message ID 1355141735-19576-1-git-send-email-amwang@redhat.com
State Accepted, archived
Delegated to: David Miller
Headers show

Commit Message

Amerigo Wang Dec. 10, 2012, 12:15 p.m. UTC
From: Cong Wang <amwang@redhat.com>

In case of rehashing, introduce a global variable 'br_mdb_rehash_seq'
which gets increased every time when rehashing, and assign
net->dev_base_seq + br_mdb_rehash_seq to cb->seq.

In theory cb->seq could be wrapped to zero, but this is not
easy to fix, as net->dev_base_seq is not visible inside
br_mdb_rehash(). In practice, this is rare.

Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Stephen Hemminger <shemminger@vyatta.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Thomas Graf <tgraf@suug.ch>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Cong Wang <amwang@redhat.com>
---
 net/bridge/br_mdb.c       |    6 ++----
 net/bridge/br_multicast.c |    2 ++
 net/bridge/br_private.h   |    1 +
 3 files changed, 5 insertions(+), 4 deletions(-)

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Comments

David Miller Dec. 10, 2012, 6:46 p.m. UTC | #1
From: Cong Wang <amwang@redhat.com>
Date: Mon, 10 Dec 2012 20:15:35 +0800

> From: Cong Wang <amwang@redhat.com>
> 
> In case of rehashing, introduce a global variable 'br_mdb_rehash_seq'
> which gets increased every time when rehashing, and assign
> net->dev_base_seq + br_mdb_rehash_seq to cb->seq.
> 
> In theory cb->seq could be wrapped to zero, but this is not
> easy to fix, as net->dev_base_seq is not visible inside
> br_mdb_rehash(). In practice, this is rare.
> 
> Cc: Herbert Xu <herbert@gondor.apana.org.au>
> Cc: Stephen Hemminger <shemminger@vyatta.com>
> Cc: "David S. Miller" <davem@davemloft.net>
> Cc: Thomas Graf <tgraf@suug.ch>
> Cc: Jesper Dangaard Brouer <brouer@redhat.com>
> Signed-off-by: Cong Wang <amwang@redhat.com>

No synchronization at all is applied to this variable, I can't
see how this is OK.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
David Miller Dec. 11, 2012, 6:45 p.m. UTC | #2
From: Cong Wang <amwang@redhat.com>
Date: Tue, 11 Dec 2012 11:49:32 +0800

> On Mon, 2012-12-10 at 13:46 -0500, David Miller wrote:
>> From: Cong Wang <amwang@redhat.com>
>> Date: Mon, 10 Dec 2012 20:15:35 +0800
>> 
>> > From: Cong Wang <amwang@redhat.com>
>> > 
>> > In case of rehashing, introduce a global variable 'br_mdb_rehash_seq'
>> > which gets increased every time when rehashing, and assign
>> > net->dev_base_seq + br_mdb_rehash_seq to cb->seq.
>> > 
>> > In theory cb->seq could be wrapped to zero, but this is not
>> > easy to fix, as net->dev_base_seq is not visible inside
>> > br_mdb_rehash(). In practice, this is rare.
>> > 
>> > Cc: Herbert Xu <herbert@gondor.apana.org.au>
>> > Cc: Stephen Hemminger <shemminger@vyatta.com>
>> > Cc: "David S. Miller" <davem@davemloft.net>
>> > Cc: Thomas Graf <tgraf@suug.ch>
>> > Cc: Jesper Dangaard Brouer <brouer@redhat.com>
>> > Signed-off-by: Cong Wang <amwang@redhat.com>
>> 
>> No synchronization at all is applied to this variable, I can't
>> see how this is OK.
> 
> br_mdb_rehash() is protected by the multicast spinlock, so increasing
> this variable is protected by it, and reading the variable doesn't need
> locking. Am I missing anything? Or should we use atomic_t?

I missed that locking, ok, looks good.

Applied, thanks.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c
index edc0d73..ccc43a9 100644
--- a/net/bridge/br_mdb.c
+++ b/net/bridge/br_mdb.c
@@ -117,10 +117,8 @@  static int br_mdb_dump(struct sk_buff *skb, struct netlink_callback *cb)
 
 	rcu_read_lock();
 
-	/* TODO: in case of rehashing, we need to check
-	 * consistency for dumping.
-	 */
-	cb->seq = net->dev_base_seq;
+	/* In theory this could be wrapped to 0... */
+	cb->seq = net->dev_base_seq + br_mdb_rehash_seq;
 
 	for_each_netdev_rcu(net, dev) {
 		if (dev->priv_flags & IFF_EBRIDGE) {
diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c
index 68e375a..847b98a1 100644
--- a/net/bridge/br_multicast.c
+++ b/net/bridge/br_multicast.c
@@ -37,6 +37,7 @@ 
 	rcu_dereference_protected(X, lockdep_is_held(&br->multicast_lock))
 
 static void br_multicast_start_querier(struct net_bridge *br);
+unsigned int br_mdb_rehash_seq;
 
 #if IS_ENABLED(CONFIG_IPV6)
 static inline int ipv6_is_transient_multicast(const struct in6_addr *addr)
@@ -338,6 +339,7 @@  static int br_mdb_rehash(struct net_bridge_mdb_htable __rcu **mdbp, int max,
 		return err;
 	}
 
+	br_mdb_rehash_seq++;
 	call_rcu_bh(&mdb->rcu, br_mdb_free);
 
 out:
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index ae0a6ec..f95b766 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -411,6 +411,7 @@  extern int br_ioctl_deviceless_stub(struct net *net, unsigned int cmd, void __us
 
 /* br_multicast.c */
 #ifdef CONFIG_BRIDGE_IGMP_SNOOPING
+extern unsigned int br_mdb_rehash_seq;
 extern int br_multicast_rcv(struct net_bridge *br,
 			    struct net_bridge_port *port,
 			    struct sk_buff *skb);