: Fix skb->nf_bridge lifetime issues

Bart, can you please have a look at this patch and ACK/NACK it? We have a bugreport in the netfilter bugzilla of broken conntrack with tunnels on top of bridge devices (#448), which should be cured by this patch.

There is also another report of broken conntrack with vlan on top of bridge devices (#400) that looks related, but probably this patch won't help.

[NETFILTER]: Fix skb->nf_bridge lifetime issues

The bridge netfilter code simulates the NF_IP_PRE_ROUTING hook and skips the real hook by registering with high priority and returning NF_STOP if skb->nf_bridge is present and the BRNF_NF_BRIDGE_PREROUTING flag is not set. The flag is only set during the simulated hook.

Because skb->nf_bridge is only freed when the packet is destroyed, the packet will not only skip the first invocation of NF_IP_PRE_ROUTING, but in the case of tunnel devices on top of the bridge also all further ones. Forwarded packets from a bridge encapsulated by a tunnel device and sent as locally outgoing packet will also still have the incorrect bridge information from the input path attached.

We already have nf_reset calls on all RX/TX paths of tunnel devices, so simply reset the nf_bridge field there too. As an added bonus, the bridge information for locally delivered packets is now also freed when the packet is queued to a socket.

---
commit 22e4dcb796d276db47a61c2382bdf2eaf76db32e
tree ccc5173acbe70506aa60e826e1881a4513f8c868
parent 337ba256a7e68f174a88ffba805b40622297fc22
author Patrick McHardy <[email protected]> Mon, 20 Feb 2006 08:41:48 +0100 committer Patrick McHardy <[email protected]> Mon, 20 Feb 2006 08:41:48 +0100

 include/linux/skbuff.h          |   24 ++++++++++++++----------
 net/ipv4/netfilter/ipt_REJECT.c |    4 ----
 2 files changed, 14 insertions(+), 14 deletions(-)

index 838ce0f..65bf1fa 100644
@@ -1351,16 +1351,6 @@ static inline void nf_conntrack_put_reas
 		kfree_skb(skb);
 }
 #endif
-static inline void nf_reset(struct sk_buff *skb)
-{
-	nf_conntrack_put(skb->nfct);
-	skb->nfct = NULL;
-#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
-	nf_conntrack_put_reasm(skb->nfct_reasm);
-	skb->nfct_reasm = NULL;
-#endif
-}
-
 #ifdef CONFIG_BRIDGE_NETFILTER
 static inline void nf_bridge_put(struct nf_bridge_info *nf_bridge)
 {

@@ -1373,6 +1363,20 @@ static inline void nf_bridge_get(struct 
 		atomic_inc(&nf_bridge->use);
 }
 #endif /* CONFIG_BRIDGE_NETFILTER */
+static inline void nf_reset(struct sk_buff *skb)
+{
+	nf_conntrack_put(skb->nfct);
+	skb->nfct = NULL;
+#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
+	nf_conntrack_put_reasm(skb->nfct_reasm);
+	skb->nfct_reasm = NULL;
+#endif
+#ifdef CONFIG_BRIDGE_NETFILTER
+	nf_bridge_put(skb->nfct);
+	skb->nfct = NULL;
+#endif
+}
+
 #else /* CONFIG_NETFILTER */
 static inline void nf_reset(struct sk_buff *skb) {}
 #endif /* CONFIG_NETFILTER */

index 26ea6c1..9d3b357 100644
@@ -154,10 +154,6 @@ static void send_reset(struct sk_buff *o
 	/* This packet will not be the same as the other: clear nf fields */
 	nf_reset(nskb);
 	nskb->nfmark = 0;
-#ifdef CONFIG_BRIDGE_NETFILTER
-	nf_bridge_put(nskb->nf_bridge);
-	nskb->nf_bridge = NULL;
-#endif
 
 	tcph = (struct tcphdr *)((u_int32_t*)nskb->nh.iph + nskb->nh.iph->ihl);
 

Op ma, 20-02-2006 te 10:33 +0100, schreef Patrick McHardy:

Looks fine to me. Nice work.

cheers,
Bart

From: Patrick McHardy <[email protected]>
Date: Mon, 20 Feb 2006 10:33:34 +0100

Want me to push this into net-2.6 now that Bart has ACK'd it?

Thanks.

I think net-2.6.17 is better, the bridging stuff is really to fragile for this to be sure it won't break anything and this has never worked so far.

From: Patrick McHardy <[email protected]>
Date: Tue, 21 Feb 2006 05:17:04 +0100

Ok, done.