netfilter: Pass socket pointer down through okfn().

On the output paths in particular, we have to sometimes deal with two
socket contexts.  First, and usually skb->sk, is the local socket that
generated the frame.

And second, is potentially the socket used to control a tunneling
socket, such as one the encapsulates using UDP.

We do not want to disassociate skb->sk when encapsulating in order
to fix this, because that would break socket memory accounting.

The most extreme case where this can cause huge problems is an
AF_PACKET socket transmitting over a vxlan device.  We hit code
paths doing checks that assume they are dealing with an ipv4
socket, but are actually operating upon the AF_PACKET one.

Signed-off-by: David S. Miller <davem@davemloft.net>

6 files changed, 56 insertions, 43 deletions

diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index 3304a5442331..e97572b5d2cc 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -35,7 +35,7 @@ static inline int should_deliver(const struct net_bridge_port *p,
 		p->state == BR_STATE_FORWARDING;
 }
 
-int br_dev_queue_push_xmit(struct sk_buff *skb)
+int br_dev_queue_push_xmit(struct sock *sk, struct sk_buff *skb)
 {
 	if (!is_skb_forwardable(skb->dev, skb)) {
 		kfree_skb(skb);
@@ -49,9 +49,10 @@ int br_dev_queue_push_xmit(struct sk_buff *skb)
 }
 EXPORT_SYMBOL_GPL(br_dev_queue_push_xmit);
 
-int br_forward_finish(struct sk_buff *skb)
+int br_forward_finish(struct sock *sk, struct sk_buff *skb)
 {
-	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_POST_ROUTING, skb, NULL, skb->dev,
+	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_POST_ROUTING, sk, skb,
+		       NULL, skb->dev,
 		       br_dev_queue_push_xmit);
 
 }
@@ -75,7 +76,8 @@ static void __br_deliver(const struct net_bridge_port *to, struct sk_buff *skb)
 		return;
 	}
 
-	NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, skb, NULL, skb->dev,
+	NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, NULL, skb,
+		NULL, skb->dev,
 		br_forward_finish);
 }
 
@@ -96,7 +98,8 @@ static void __br_forward(const struct net_bridge_port *to, struct sk_buff *skb)
 	skb->dev = to->dev;
 	skb_forward_csum(skb);
 
-	NF_HOOK(NFPROTO_BRIDGE, NF_BR_FORWARD, skb, indev, skb->dev,
+	NF_HOOK(NFPROTO_BRIDGE, NF_BR_FORWARD, NULL, skb,
+		indev, skb->dev,
 		br_forward_finish);
 }
 
diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c
index 052c5ebbc947..f921a5dce22d 100644
--- a/net/bridge/br_input.c
+++ b/net/bridge/br_input.c
@@ -55,8 +55,9 @@ static int br_pass_frame_up(struct sk_buff *skb)
 	if (!skb)
 		return NET_RX_DROP;
 
-	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, skb, indev, NULL,
-		       netif_receive_skb);
+	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, NULL, skb,
+		       indev, NULL,
+		       netif_receive_skb_sk);
 }
 
 static void br_do_proxy_arp(struct sk_buff *skb, struct net_bridge *br,
@@ -119,7 +120,7 @@ static void br_do_proxy_arp(struct sk_buff *skb, struct net_bridge *br,
 }
 
 /* note: already called with rcu_read_lock */
-int br_handle_frame_finish(struct sk_buff *skb)
+int br_handle_frame_finish(struct sock *sk, struct sk_buff *skb)
 {
 	const unsigned char *dest = eth_hdr(skb)->h_dest;
 	struct net_bridge_port *p = br_port_get_rcu(skb->dev);
@@ -207,7 +208,7 @@ drop:
 EXPORT_SYMBOL_GPL(br_handle_frame_finish);
 
 /* note: already called with rcu_read_lock */
-static int br_handle_local_finish(struct sk_buff *skb)
+static int br_handle_local_finish(struct sock *sk, struct sk_buff *skb)
 {
 	struct net_bridge_port *p = br_port_get_rcu(skb->dev);
 	u16 vid = 0;
@@ -277,8 +278,8 @@ rx_handler_result_t br_handle_frame(struct sk_buff **pskb)
 		}
 
 		/* Deliver packet to local host only */
-		if (NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, skb, skb->dev,
-			    NULL, br_handle_local_finish)) {
+		if (NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, NULL, skb,
+			    skb->dev, NULL, br_handle_local_finish)) {
 			return RX_HANDLER_CONSUMED; /* consumed by filter */
 		} else {
 			*pskb = skb;
@@ -302,7 +303,8 @@ forward:
 		if (ether_addr_equal(p->br->dev->dev_addr, dest))
 			skb->pkt_type = PACKET_HOST;
 
-		NF_HOOK(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, skb, skb->dev, NULL,
+		NF_HOOK(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, NULL, skb,
+			skb->dev, NULL,
 			br_handle_frame_finish);
 		break;
 	default:
diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c
index c465876c7861..4b6722f8f179 100644
--- a/net/bridge/br_multicast.c
+++ b/net/bridge/br_multicast.c
@@ -814,7 +814,8 @@ static void __br_multicast_send_query(struct net_bridge *br,
 
 	if (port) {
 		skb->dev = port->dev;
-		NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, skb, NULL, skb->dev,
+		NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, NULL, skb,
+			NULL, skb->dev,
 			br_dev_queue_push_xmit);
 	} else {
 		br_multicast_select_own_querier(br, ip, skb);
diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c
index 7527e94dd5dc..acd31c9f2116 100644
--- a/net/bridge/br_netfilter.c
+++ b/net/bridge/br_netfilter.c
@@ -261,7 +261,7 @@ static void nf_bridge_update_protocol(struct sk_buff *skb)
 /* PF_BRIDGE/PRE_ROUTING *********************************************/
 /* Undo the changes made for ip6tables PREROUTING and continue the
  * bridge PRE_ROUTING hook. */
-static int br_nf_pre_routing_finish_ipv6(struct sk_buff *skb)
+static int br_nf_pre_routing_finish_ipv6(struct sock *sk, struct sk_buff *skb)
 {
 	struct nf_bridge_info *nf_bridge = skb->nf_bridge;
 	struct rtable *rt;
@@ -282,7 +282,8 @@ static int br_nf_pre_routing_finish_ipv6(struct sk_buff *skb)
 	skb->dev = nf_bridge->physindev;
 	nf_bridge_update_protocol(skb);
 	nf_bridge_push_encap_header(skb);
-	NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, skb, skb->dev, NULL,
+	NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, sk, skb,
+		       skb->dev, NULL,
 		       br_handle_frame_finish, 1);
 
 	return 0;
@@ -293,7 +294,7 @@ static int br_nf_pre_routing_finish_ipv6(struct sk_buff *skb)
  * don't, we use the neighbour framework to find out. In both cases, we make
  * sure that br_handle_frame_finish() is called afterwards.
  */
-static int br_nf_pre_routing_finish_bridge(struct sk_buff *skb)
+static int br_nf_pre_routing_finish_bridge(struct sock *sk, struct sk_buff *skb)
 {
 	struct nf_bridge_info *nf_bridge = skb->nf_bridge;
 	struct neighbour *neigh;
@@ -310,7 +311,7 @@ static int br_nf_pre_routing_finish_bridge(struct sk_buff *skb)
 		if (neigh->hh.hh_len) {
 			neigh_hh_bridge(&neigh->hh, skb);
 			skb->dev = nf_bridge->physindev;
-			ret = br_handle_frame_finish(skb);
+			ret = br_handle_frame_finish(sk, skb);
 		} else {
 			/* the neighbour function below overwrites the complete
 			 * MAC header, so we save the Ethernet source address and
@@ -387,7 +388,7 @@ static bool dnat_took_place(const struct sk_buff *skb)
  * device, we proceed as if ip_route_input() succeeded. If it differs from the
  * logical bridge port or if ip_route_output_key() fails we drop the packet.
  */
-static int br_nf_pre_routing_finish(struct sk_buff *skb)
+static int br_nf_pre_routing_finish(struct sock *sk, struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dev;
 	struct iphdr *iph = ip_hdr(skb);
@@ -440,7 +441,7 @@ bridged_dnat:
 				nf_bridge_push_encap_header(skb);
 				NF_HOOK_THRESH(NFPROTO_BRIDGE,
 					       NF_BR_PRE_ROUTING,
-					       skb, skb->dev, NULL,
+					       sk, skb, skb->dev, NULL,
 					       br_nf_pre_routing_finish_bridge,
 					       1);
 				return 0;
@@ -460,7 +461,8 @@ bridged_dnat:
 	skb->dev = nf_bridge->physindev;
 	nf_bridge_update_protocol(skb);
 	nf_bridge_push_encap_header(skb);
-	NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, skb, skb->dev, NULL,
+	NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, sk, skb,
+		       skb->dev, NULL,
 		       br_handle_frame_finish, 1);
 
 	return 0;
@@ -596,7 +598,8 @@ static unsigned int br_nf_pre_routing_ipv6(const struct nf_hook_ops *ops,
 		return NF_DROP;
 
 	skb->protocol = htons(ETH_P_IPV6);
-	NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING, skb, skb->dev, NULL,
+	NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING, state->sk, skb,
+		skb->dev, NULL,
 		br_nf_pre_routing_finish_ipv6);
 
 	return NF_STOLEN;
@@ -651,7 +654,8 @@ static unsigned int br_nf_pre_routing(const struct nf_hook_ops *ops,
 
 	skb->protocol = htons(ETH_P_IP);
 
-	NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING, skb, skb->dev, NULL,
+	NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING, state->sk, skb,
+		skb->dev, NULL,
 		br_nf_pre_routing_finish);
 
 	return NF_STOLEN;
@@ -674,7 +678,7 @@ static unsigned int br_nf_local_in(const struct nf_hook_ops *ops,
 }
 
 /* PF_BRIDGE/FORWARD *************************************************/
-static int br_nf_forward_finish(struct sk_buff *skb)
+static int br_nf_forward_finish(struct sock *sk, struct sk_buff *skb)
 {
 	struct nf_bridge_info *nf_bridge = skb->nf_bridge;
 	struct net_device *in;
@@ -691,8 +695,8 @@ static int br_nf_forward_finish(struct sk_buff *skb)
 	}
 	nf_bridge_push_encap_header(skb);
 
-	NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_FORWARD, skb, in,
-		       skb->dev, br_forward_finish, 1);
+	NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_FORWARD, sk, skb,
+		       in, skb->dev, br_forward_finish, 1);
 	return 0;
 }
 
@@ -746,7 +750,8 @@ static unsigned int br_nf_forward_ip(const struct nf_hook_ops *ops,
 	else
 		skb->protocol = htons(ETH_P_IPV6);
 
-	NF_HOOK(pf, NF_INET_FORWARD, skb, brnf_get_logical_dev(skb, state->in),
+	NF_HOOK(pf, NF_INET_FORWARD, NULL, skb,
+		brnf_get_logical_dev(skb, state->in),
 		parent,	br_nf_forward_finish);
 
 	return NF_STOLEN;
@@ -780,8 +785,8 @@ static unsigned int br_nf_forward_arp(const struct nf_hook_ops *ops,
 		return NF_ACCEPT;
 	}
 	*d = state->in;
-	NF_HOOK(NFPROTO_ARP, NF_ARP_FORWARD, skb, state->in,
-		state->out, br_nf_forward_finish);
+	NF_HOOK(NFPROTO_ARP, NF_ARP_FORWARD, state->sk, skb,
+		state->in, state->out, br_nf_forward_finish);
 
 	return NF_STOLEN;
 }
@@ -804,24 +809,24 @@ static bool nf_bridge_copy_header(struct sk_buff *skb)
 	return true;
 }
 
-static int br_nf_push_frag_xmit(struct sk_buff *skb)
+static int br_nf_push_frag_xmit(struct sock *sk, struct sk_buff *skb)
 {
 	if (!nf_bridge_copy_header(skb)) {
 		kfree_skb(skb);
 		return 0;
 	}
 
-	return br_dev_queue_push_xmit(skb);
+	return br_dev_queue_push_xmit(sk, skb);
 }
 
-static int br_nf_dev_queue_xmit(struct sk_buff *skb)
+static int br_nf_dev_queue_xmit(struct sock *sk, struct sk_buff *skb)
 {
 	int ret;
 	int frag_max_size;
 	unsigned int mtu_reserved;
 
 	if (skb_is_gso(skb) || skb->protocol != htons(ETH_P_IP))
-		return br_dev_queue_push_xmit(skb);
+		return br_dev_queue_push_xmit(sk, skb);
 
 	mtu_reserved = nf_bridge_mtu_reduction(skb);
 	/* This is wrong! We should preserve the original fragment
@@ -833,16 +838,16 @@ static int br_nf_dev_queue_xmit(struct sk_buff *skb)
 			/* Drop invalid packet */
 			return NF_DROP;
 		IPCB(skb)->frag_max_size = frag_max_size;
-		ret = ip_fragment(skb, br_nf_push_frag_xmit);
+		ret = ip_fragment(sk, skb, br_nf_push_frag_xmit);
 	} else
-		ret = br_dev_queue_push_xmit(skb);
+		ret = br_dev_queue_push_xmit(sk, skb);
 
 	return ret;
 }
 #else
-static int br_nf_dev_queue_xmit(struct sk_buff *skb)
+static int br_nf_dev_queue_xmit(struct sock *sk, struct sk_buff *skb)
 {
-        return br_dev_queue_push_xmit(skb);
+        return br_dev_queue_push_xmit(sk, skb);
 }
 #endif
 
@@ -887,7 +892,8 @@ static unsigned int br_nf_post_routing(const struct nf_hook_ops *ops,
 	else
 		skb->protocol = htons(ETH_P_IPV6);
 
-	NF_HOOK(pf, NF_INET_POST_ROUTING, skb, NULL, realoutdev,
+	NF_HOOK(pf, NF_INET_POST_ROUTING, state->sk, skb,
+		NULL, realoutdev,
 		br_nf_dev_queue_xmit);
 
 	return NF_STOLEN;
@@ -927,7 +933,7 @@ static void br_nf_pre_routing_finish_bridge_slow(struct sk_buff *skb)
 	skb_copy_to_linear_data_offset(skb, -(ETH_HLEN-ETH_ALEN),
 				       skb->nf_bridge->data, ETH_HLEN-ETH_ALEN);
 	skb->dev = nf_bridge->physindev;
-	br_handle_frame_finish(skb);
+	br_handle_frame_finish(NULL, skb);
 }
 
 static int br_nf_dev_xmit(struct sk_buff *skb)
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index b46fa0c5b8ec..6ca0251cb478 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -410,10 +410,10 @@ int br_fdb_external_learn_del(struct net_bridge *br, struct net_bridge_port *p,
 
 /* br_forward.c */
 void br_deliver(const struct net_bridge_port *to, struct sk_buff *skb);
-int br_dev_queue_push_xmit(struct sk_buff *skb);
+int br_dev_queue_push_xmit(struct sock *sk, struct sk_buff *skb);
 void br_forward(const struct net_bridge_port *to,
 		struct sk_buff *skb, struct sk_buff *skb0);
-int br_forward_finish(struct sk_buff *skb);
+int br_forward_finish(struct sock *sk, struct sk_buff *skb);
 void br_flood_deliver(struct net_bridge *br, struct sk_buff *skb, bool unicast);
 void br_flood_forward(struct net_bridge *br, struct sk_buff *skb,
 		      struct sk_buff *skb2, bool unicast);
@@ -431,7 +431,7 @@ void br_port_flags_change(struct net_bridge_port *port, unsigned long mask);
 void br_manage_promisc(struct net_bridge *br);
 
 /* br_input.c */
-int br_handle_frame_finish(struct sk_buff *skb);
+int br_handle_frame_finish(struct sock *sk, struct sk_buff *skb);
 rx_handler_result_t br_handle_frame(struct sk_buff **pskb);
 
 static inline bool br_rx_handler_check_rcu(const struct net_device *dev)
diff --git a/net/bridge/br_stp_bpdu.c b/net/bridge/br_stp_bpdu.c
index bdb459d21ad8..534fc4cd263e 100644
--- a/net/bridge/br_stp_bpdu.c
+++ b/net/bridge/br_stp_bpdu.c
@@ -54,8 +54,9 @@ static void br_send_bpdu(struct net_bridge_port *p,
 
 	skb_reset_mac_header(skb);
 
-	NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, skb, NULL, skb->dev,
-		dev_queue_xmit);
+	NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, NULL, skb,
+		NULL, skb->dev,
+		dev_queue_xmit_sk);
 }
 
 static inline void br_set_ticks(unsigned char *dest, int j)