Merge branch 'nf' of git://1984.lsi.us.es/net

3 files changed, 49 insertions, 36 deletions

diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c
index 611c335..2555816 100644
--- a/net/netfilter/ipvs/ip_vs_core.c
+++ b/net/netfilter/ipvs/ip_vs_core.c
@@ -232,6 +232,7 @@ ip_vs_sched_persist(struct ip_vs_service *svc,
 	__be16 dport = 0;		/* destination port to forward */
 	unsigned int flags;
 	struct ip_vs_conn_param param;
+	const union nf_inet_addr fwmark = { .ip = htonl(svc->fwmark) };
 	union nf_inet_addr snet;	/* source network of the client,
 					   after masking */
 
@@ -267,7 +268,6 @@ ip_vs_sched_persist(struct ip_vs_service *svc,
 	{
 		int protocol = iph.protocol;
 		const union nf_inet_addr *vaddr = &iph.daddr;
-		const union nf_inet_addr fwmark = { .ip = htonl(svc->fwmark) };
 		__be16 vport = 0;
 
 		if (dst_port == svc->port) {
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 9307b03..cc70517 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -1367,15 +1367,12 @@ ctnetlink_create_conntrack(struct net *net, u16 zone,
 						    nf_ct_protonum(ct));
 		if (helper == NULL) {
 			rcu_read_unlock();
-			spin_unlock_bh(&nf_conntrack_lock);
 #ifdef CONFIG_MODULES
 			if (request_module("nfct-helper-%s", helpname) < 0) {
-				spin_lock_bh(&nf_conntrack_lock);
 				err = -EOPNOTSUPP;
 				goto err1;
 			}
 
-			spin_lock_bh(&nf_conntrack_lock);
 			rcu_read_lock();
 			helper = __nf_conntrack_helper_find(helpname,
 							    nf_ct_l3num(ct),
@@ -1469,7 +1466,10 @@ ctnetlink_create_conntrack(struct net *net, u16 zone,
 		tstamp->start = ktime_to_ns(ktime_get_real());
 
 	add_timer(&ct->timeout);
+	spin_lock_bh(&nf_conntrack_lock);
 	nf_conntrack_hash_insert(ct);
+	nf_conntrack_get(&ct->ct_general);
+	spin_unlock_bh(&nf_conntrack_lock);
 	rcu_read_unlock();
 
 	return ct;
@@ -1490,6 +1490,7 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb,
 	struct nf_conntrack_tuple otuple, rtuple;
 	struct nf_conntrack_tuple_hash *h = NULL;
 	struct nfgenmsg *nfmsg = nlmsg_data(nlh);
+	struct nf_conn *ct;
 	u_int8_t u3 = nfmsg->nfgen_family;
 	u16 zone;
 	int err;
@@ -1512,25 +1513,22 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb,
 
 	spin_lock_bh(&nf_conntrack_lock);
 	if (cda[CTA_TUPLE_ORIG])
-		h = __nf_conntrack_find(net, zone, &otuple);
+		h = nf_conntrack_find_get(net, zone, &otuple);
 	else if (cda[CTA_TUPLE_REPLY])
-		h = __nf_conntrack_find(net, zone, &rtuple);
+		h = nf_conntrack_find_get(net, zone, &rtuple);
+	spin_unlock_bh(&nf_conntrack_lock);
 
 	if (h == NULL) {
 		err = -ENOENT;
 		if (nlh->nlmsg_flags & NLM_F_CREATE) {
-			struct nf_conn *ct;
 			enum ip_conntrack_events events;
 
 			ct = ctnetlink_create_conntrack(net, zone, cda, &otuple,
 							&rtuple, u3);
-			if (IS_ERR(ct)) {
-				err = PTR_ERR(ct);
-				goto out_unlock;
-			}
+			if (IS_ERR(ct))
+				return PTR_ERR(ct);
+
 			err = 0;
-			nf_conntrack_get(&ct->ct_general);
-			spin_unlock_bh(&nf_conntrack_lock);
 			if (test_bit(IPS_EXPECTED_BIT, &ct->status))
 				events = IPCT_RELATED;
 			else
@@ -1545,23 +1543,19 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb,
 						      ct, NETLINK_CB(skb).pid,
 						      nlmsg_report(nlh));
 			nf_ct_put(ct);
-		} else
-			spin_unlock_bh(&nf_conntrack_lock);
+		}
 
 		return err;
 	}
 	/* implicit 'else' */
 
-	/* We manipulate the conntrack inside the global conntrack table lock,
-	 * so there's no need to increase the refcount */
 	err = -EEXIST;
+	ct = nf_ct_tuplehash_to_ctrack(h);
 	if (!(nlh->nlmsg_flags & NLM_F_EXCL)) {
-		struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h);
-
+		spin_lock_bh(&nf_conntrack_lock);
 		err = ctnetlink_change_conntrack(ct, cda);
+		spin_unlock_bh(&nf_conntrack_lock);
 		if (err == 0) {
-			nf_conntrack_get(&ct->ct_general);
-			spin_unlock_bh(&nf_conntrack_lock);
 			nf_conntrack_eventmask_report((1 << IPCT_REPLY) |
 						      (1 << IPCT_ASSURED) |
 						      (1 << IPCT_HELPER) |
@@ -1570,15 +1564,10 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb,
 						      (1 << IPCT_MARK),
 						      ct, NETLINK_CB(skb).pid,
 						      nlmsg_report(nlh));
-			nf_ct_put(ct);
-		} else
-			spin_unlock_bh(&nf_conntrack_lock);
-
-		return err;
+		}
 	}
 
-out_unlock:
-	spin_unlock_bh(&nf_conntrack_lock);
+	nf_ct_put(ct);
 	return err;
 }
 
diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c
index b3a7db6..ce60cf0 100644
--- a/net/netfilter/nf_queue.c
+++ b/net/netfilter/nf_queue.c
@@ -203,6 +203,27 @@ err:
 	return status;
 }
 
+#ifdef CONFIG_BRIDGE_NETFILTER
+/* When called from bridge netfilter, skb->data must point to MAC header
+ * before calling skb_gso_segment(). Else, original MAC header is lost
+ * and segmented skbs will be sent to wrong destination.
+ */
+static void nf_bridge_adjust_skb_data(struct sk_buff *skb)
+{
+	if (skb->nf_bridge)
+		__skb_push(skb, skb->network_header - skb->mac_header);
+}
+
+static void nf_bridge_adjust_segmented_data(struct sk_buff *skb)
+{
+	if (skb->nf_bridge)
+		__skb_pull(skb, skb->network_header - skb->mac_header);
+}
+#else
+#define nf_bridge_adjust_skb_data(s) do {} while (0)
+#define nf_bridge_adjust_segmented_data(s) do {} while (0)
+#endif
+
 int nf_queue(struct sk_buff *skb,
 	     struct list_head *elem,
 	     u_int8_t pf, unsigned int hook,
@@ -212,7 +233,7 @@ int nf_queue(struct sk_buff *skb,
 	     unsigned int queuenum)
 {
 	struct sk_buff *segs;
-	int err;
+	int err = -EINVAL;
 	unsigned int queued;
 
 	if (!skb_is_gso(skb))
@@ -228,23 +249,25 @@ int nf_queue(struct sk_buff *skb,
 		break;
 	}
 
+	nf_bridge_adjust_skb_data(skb);
 	segs = skb_gso_segment(skb, 0);
 	/* Does not use PTR_ERR to limit the number of error codes that can be
 	 * returned by nf_queue.  For instance, callers rely on -ECANCELED to mean
 	 * 'ignore this hook'.
 	 */
 	if (IS_ERR(segs))
-		return -EINVAL;
-
+		goto out_err;
 	queued = 0;
 	err = 0;
 	do {
 		struct sk_buff *nskb = segs->next;
 
 		segs->next = NULL;
-		if (err == 0)
+		if (err == 0) {
+			nf_bridge_adjust_segmented_data(segs);
 			err = __nf_queue(segs, elem, pf, hook, indev,
 					   outdev, okfn, queuenum);
+		}
 		if (err == 0)
 			queued++;
 		else
@@ -252,11 +275,12 @@ int nf_queue(struct sk_buff *skb,
 		segs = nskb;
 	} while (segs);
 
-	/* also free orig skb if only some segments were queued */
-	if (unlikely(err && queued))
-		err = 0;
-	if (err == 0)
+	if (queued) {
 		kfree_skb(skb);
+		return 0;
+	}
+  out_err:
+	nf_bridge_adjust_segmented_data(skb);
 	return err;
 }