[linux-2.6-block.git] / net / bridge / br_forward.c

/*
 *	Forwarding decision
 *	Linux ethernet bridge
 *
 *	Authors:
 *	Lennert Buytenhek		<buytenh@gnu.org>
 *
 *	This program is free software; you can redistribute it and/or
 *	modify it under the terms of the GNU General Public License
 *	as published by the Free Software Foundation; either version
 *	2 of the License, or (at your option) any later version.
 */

#include <linux/err.h>
#include <linux/slab.h>
#include <linux/kernel.h>
#include <linux/netdevice.h>
#include <linux/netpoll.h>
#include <linux/skbuff.h>
#include <linux/if_vlan.h>
#include <linux/netfilter_bridge.h>
#include "br_private.h"

/* Don't forward packets to originating port or forwarding disabled */
static inline int should_deliver(const struct net_bridge_port *p,
				 const struct sk_buff *skb)
{
	struct net_bridge_vlan_group *vg;

	vg = nbp_vlan_group_rcu(p);
	return ((p->flags & BR_HAIRPIN_MODE) || skb->dev != p->dev) &&
		br_allowed_egress(vg, skb) && p->state == BR_STATE_FORWARDING &&
		nbp_switchdev_allowed_egress(p, skb) &&
		!br_skb_isolated(p, skb);
}

int br_dev_queue_push_xmit(struct net *net, struct sock *sk, struct sk_buff *skb)
{
	skb_push(skb, ETH_HLEN);
	if (!is_skb_forwardable(skb->dev, skb))
		goto drop;

	br_drop_fake_rtable(skb);

	if (skb->ip_summed == CHECKSUM_PARTIAL &&
	    (skb->protocol == htons(ETH_P_8021Q) ||
	     skb->protocol == htons(ETH_P_8021AD))) {
		int depth;

		if (!__vlan_get_protocol(skb, skb->protocol, &depth))
			goto drop;

		skb_set_network_header(skb, depth);
	}

	dev_queue_xmit(skb);

	return 0;

drop:
	kfree_skb(skb);
	return 0;
}
EXPORT_SYMBOL_GPL(br_dev_queue_push_xmit);

int br_forward_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
{
	skb->tstamp = 0;
	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_POST_ROUTING,
		       net, sk, skb, NULL, skb->dev,
		       br_dev_queue_push_xmit);

}
EXPORT_SYMBOL_GPL(br_forward_finish);

static void __br_forward(const struct net_bridge_port *to,
			 struct sk_buff *skb, bool local_orig)
{
	struct net_bridge_vlan_group *vg;
	struct net_device *indev;
	struct net *net;
	int br_hook;

	vg = nbp_vlan_group_rcu(to);
	skb = br_handle_vlan(to->br, to, vg, skb);
	if (!skb)
		return;

	indev = skb->dev;
	skb->dev = to->dev;
	if (!local_orig) {
		if (skb_warn_if_lro(skb)) {
			kfree_skb(skb);
			return;
		}
		br_hook = NF_BR_FORWARD;
		skb_forward_csum(skb);
		net = dev_net(indev);
	} else {
		if (unlikely(netpoll_tx_running(to->br->dev))) {
			skb_push(skb, ETH_HLEN);
			if (!is_skb_forwardable(skb->dev, skb))
				kfree_skb(skb);
			else
				br_netpoll_send_skb(to, skb);
			return;
		}
		br_hook = NF_BR_LOCAL_OUT;
		net = dev_net(skb->dev);
		indev = NULL;
	}

	NF_HOOK(NFPROTO_BRIDGE, br_hook,
		net, NULL, skb, indev, skb->dev,
		br_forward_finish);
}

static int deliver_clone(const struct net_bridge_port *prev,
			 struct sk_buff *skb, bool local_orig)
{
	struct net_device *dev = BR_INPUT_SKB_CB(skb)->brdev;

	skb = skb_clone(skb, GFP_ATOMIC);
	if (!skb) {
		dev->stats.tx_dropped++;
		return -ENOMEM;
	}

	__br_forward(prev, skb, local_orig);
	return 0;
}

/**
 * br_forward - forward a packet to a specific port
 * @to: destination port
 * @skb: packet being forwarded
 * @local_rcv: packet will be received locally after forwarding
 * @local_orig: packet is locally originated
 *
 * Should be called with rcu_read_lock.
 */
void br_forward(const struct net_bridge_port *to,
		struct sk_buff *skb, bool local_rcv, bool local_orig)
{
	if (unlikely(!to))
		goto out;

	/* redirect to backup link if the destination port is down */
	if (rcu_access_pointer(to->backup_port) && !netif_carrier_ok(to->dev)) {
		struct net_bridge_port *backup_port;

		backup_port = rcu_dereference(to->backup_port);
		if (unlikely(!backup_port))
			goto out;
		to = backup_port;
	}

	if (should_deliver(to, skb)) {
		if (local_rcv)
			deliver_clone(to, skb, local_orig);
		else
			__br_forward(to, skb, local_orig);
		return;
	}

out:
	if (!local_rcv)
		kfree_skb(skb);
}
EXPORT_SYMBOL_GPL(br_forward);

static struct net_bridge_port *maybe_deliver(
	struct net_bridge_port *prev, struct net_bridge_port *p,
	struct sk_buff *skb, bool local_orig)
{
	int err;

	if (!should_deliver(p, skb))
		return prev;

	if (!prev)
		goto out;

	err = deliver_clone(prev, skb, local_orig);
	if (err)
		return ERR_PTR(err);

out:
	return p;
}

/* called under rcu_read_lock */
void br_flood(struct net_bridge *br, struct sk_buff *skb,
	      enum br_pkt_type pkt_type, bool local_rcv, bool local_orig)
{
	u8 igmp_type = br_multicast_igmp_type(skb);
	struct net_bridge_port *prev = NULL;
	struct net_bridge_port *p;

	list_for_each_entry_rcu(p, &br->port_list, list) {
		/* Do not flood unicast traffic to ports that turn it off, nor
		 * other traffic if flood off, except for traffic we originate
		 */
		switch (pkt_type) {
		case BR_PKT_UNICAST:
			if (!(p->flags & BR_FLOOD))
				continue;
			break;
		case BR_PKT_MULTICAST:
			if (!(p->flags & BR_MCAST_FLOOD) && skb->dev != br->dev)
				continue;
			break;
		case BR_PKT_BROADCAST:
			if (!(p->flags & BR_BCAST_FLOOD) && skb->dev != br->dev)
				continue;
			break;
		}

		/* Do not flood to ports that enable proxy ARP */
		if (p->flags & BR_PROXYARP)
			continue;
		if ((p->flags & (BR_PROXYARP_WIFI | BR_NEIGH_SUPPRESS)) &&
		    BR_INPUT_SKB_CB(skb)->proxyarp_replied)
			continue;

		prev = maybe_deliver(prev, p, skb, local_orig);
		if (IS_ERR(prev))
			goto out;
		if (prev == p)
			br_multicast_count(p->br, p, skb, igmp_type,
					   BR_MCAST_DIR_TX);
	}

	if (!prev)
		goto out;

	if (local_rcv)
		deliver_clone(prev, skb, local_orig);
	else
		__br_forward(prev, skb, local_orig);
	return;

out:
	if (!local_rcv)
		kfree_skb(skb);
}

#ifdef CONFIG_BRIDGE_IGMP_SNOOPING
static void maybe_deliver_addr(struct net_bridge_port *p, struct sk_buff *skb,
			       const unsigned char *addr, bool local_orig)
{
	struct net_device *dev = BR_INPUT_SKB_CB(skb)->brdev;
	const unsigned char *src = eth_hdr(skb)->h_source;

	if (!should_deliver(p, skb))
		return;

	/* Even with hairpin, no soliloquies - prevent breaking IPv6 DAD */
	if (skb->dev == p->dev && ether_addr_equal(src, addr))
		return;

	skb = skb_copy(skb, GFP_ATOMIC);
	if (!skb) {
		dev->stats.tx_dropped++;
		return;
	}

	if (!is_broadcast_ether_addr(addr))
		memcpy(eth_hdr(skb)->h_dest, addr, ETH_ALEN);

	__br_forward(p, skb, local_orig);
}

/* called with rcu_read_lock */
void br_multicast_flood(struct net_bridge_mdb_entry *mdst,
			struct sk_buff *skb,
			bool local_rcv, bool local_orig)
{
	struct net_device *dev = BR_INPUT_SKB_CB(skb)->brdev;
	u8 igmp_type = br_multicast_igmp_type(skb);
	struct net_bridge *br = netdev_priv(dev);
	struct net_bridge_port *prev = NULL;
	struct net_bridge_port_group *p;
	struct hlist_node *rp;

	rp = rcu_dereference(hlist_first_rcu(&br->router_list));
	p = mdst ? rcu_dereference(mdst->ports) : NULL;
	while (p || rp) {
		struct net_bridge_port *port, *lport, *rport;

		lport = p ? p->port : NULL;
		rport = hlist_entry_safe(rp, struct net_bridge_port, rlist);

		if ((unsigned long)lport > (unsigned long)rport) {
			port = lport;

			if (port->flags & BR_MULTICAST_TO_UNICAST) {
				maybe_deliver_addr(lport, skb, p->eth_addr,
						   local_orig);
				goto delivered;
			}
		} else {
			port = rport;
		}

		prev = maybe_deliver(prev, port, skb, local_orig);
delivered:
		if (IS_ERR(prev))
			goto out;
		if (prev == port)
			br_multicast_count(port->br, port, skb, igmp_type,
					   BR_MCAST_DIR_TX);

		if ((unsigned long)lport >= (unsigned long)port)
			p = rcu_dereference(p->next);
		if ((unsigned long)rport >= (unsigned long)port)
			rp = rcu_dereference(hlist_next_rcu(rp));
	}

	if (!prev)
		goto out;

	if (local_rcv)
		deliver_clone(prev, skb, local_orig);
	else
		__br_forward(prev, skb, local_orig);
	return;

out:
	if (!local_rcv)
		kfree_skb(skb);
}
#endif
Commit	Line	Data
1da177e4 LT	1	/*
	2	* Forwarding decision
	3	* Linux ethernet bridge
	4	*
	5	* Authors:
	6	* Lennert Buytenhek <buytenh@gnu.org>
	7	*
1da177e4 LT	8	* This program is free software; you can redistribute it and/or
	9	* modify it under the terms of the GNU General Public License
	10	* as published by the Free Software Foundation; either version
	11	* 2 of the License, or (at your option) any later version.
	12	*/
	13
025d89c2	14	#include <linux/err.h>
5a0e3ad6	15	#include <linux/slab.h>
1da177e4 LT	16	#include <linux/kernel.h>
1da177e4 LT	17	#include <linux/netdevice.h>
c06ee961	18	#include <linux/netpoll.h>
1da177e4	19	#include <linux/skbuff.h>
85ca719e	20	#include <linux/if_vlan.h>
1da177e4 LT	21	#include <linux/netfilter_bridge.h>
	22	#include "br_private.h"
	23
1a81a2e0	24	/* Don't forward packets to originating port or forwarding disabled */
9d6f229f	25	static inline int should_deliver(const struct net_bridge_port *p,
1da177e4 LT	26	const struct sk_buff *skb)
1da177e4 LT	27	{
2594e906 NA	28	struct net_bridge_vlan_group *vg;
2594e906 NA	29
907b1e6e	30	vg = nbp_vlan_group_rcu(p);
a97bfc1d	31	return ((p->flags & BR_HAIRPIN_MODE) \|\| skb->dev != p->dev) &&
6bc506b4	32	br_allowed_egress(vg, skb) && p->state == BR_STATE_FORWARDING &&
7d850abd NA	33	nbp_switchdev_allowed_egress(p, skb) &&
7d850abd NA	34	!br_skb_isolated(p, skb);
1da177e4 LT	35	}
1da177e4 LT	36
0c4b51f0	37	int br_dev_queue_push_xmit(struct net net, struct sock sk, struct sk_buff *skb)
1da177e4	38	{
28c1382f	39	skb_push(skb, ETH_HLEN);
df356d5e TM	40	if (!is_skb_forwardable(skb->dev, skb))
	41	goto drop;
	42
df356d5e	43	br_drop_fake_rtable(skb);
df356d5e TM	44
	45	if (skb->ip_summed == CHECKSUM_PARTIAL &&
	46	(skb->protocol == htons(ETH_P_8021Q) \|\|
	47	skb->protocol == htons(ETH_P_8021AD))) {
	48	int depth;
	49
	50	if (!__vlan_get_protocol(skb, skb->protocol, &depth))
	51	goto drop;
	52
	53	skb_set_network_header(skb, depth);
1da177e4 LT	54	}
1da177e4 LT	55
df356d5e TM	56	dev_queue_xmit(skb);
	57
	58	return 0;
	59
	60	drop:
	61	kfree_skb(skb);
1da177e4 LT	62	return 0;
1da177e4 LT	63	}
34666d46	64	EXPORT_SYMBOL_GPL(br_dev_queue_push_xmit);
1da177e4	65
0c4b51f0	66	int br_forward_finish(struct net net, struct sock sk, struct sk_buff *skb)
1da177e4	67	{
41d1c883	68	skb->tstamp = 0;
29a26a56 EB	69	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_POST_ROUTING,
29a26a56 EB	70	net, sk, skb, NULL, skb->dev,
9ef513be	71	br_dev_queue_push_xmit);
1da177e4	72
1da177e4	73	}
34666d46	74	EXPORT_SYMBOL_GPL(br_forward_finish);
1da177e4	75
37b090e6 NA	76	static void __br_forward(const struct net_bridge_port *to,
37b090e6 NA	77	struct sk_buff *skb, bool local_orig)
1da177e4	78	{
2594e906	79	struct net_bridge_vlan_group *vg;
37b090e6 NA	80	struct net_device *indev;
	81	struct net *net;
	82	int br_hook;
2594e906	83
907b1e6e	84	vg = nbp_vlan_group_rcu(to);
11538d03	85	skb = br_handle_vlan(to->br, to, vg, skb);
78851988 VY	86	if (!skb)
	87	return;
	88
37b090e6	89	indev = skb->dev;
1da177e4	90	skb->dev = to->dev;
37b090e6 NA	91	if (!local_orig) {
37b090e6 NA	92	if (skb_warn_if_lro(skb)) {
91d2c34a	93	kfree_skb(skb);
37b090e6	94	return;
91d2c34a	95	}
37b090e6 NA	96	br_hook = NF_BR_FORWARD;
	97	skb_forward_csum(skb);
	98	net = dev_net(indev);
	99	} else {
	100	if (unlikely(netpoll_tx_running(to->br->dev))) {
28c1382f YW	101	skb_push(skb, ETH_HLEN);
28c1382f YW	102	if (!is_skb_forwardable(skb->dev, skb))
37b090e6	103	kfree_skb(skb);
28c1382f	104	else
37b090e6	105	br_netpoll_send_skb(to, skb);
37b090e6 NA	106	return;
	107	}
	108	br_hook = NF_BR_LOCAL_OUT;
	109	net = dev_net(skb->dev);
	110	indev = NULL;
91d2c34a HX	111	}
91d2c34a HX	112
37b090e6 NA	113	NF_HOOK(NFPROTO_BRIDGE, br_hook,
37b090e6 NA	114	net, NULL, skb, indev, skb->dev,
713aefa3	115	br_forward_finish);
1da177e4 LT	116	}
1da177e4 LT	117
37b090e6 NA	118	static int deliver_clone(const struct net_bridge_port *prev,
37b090e6 NA	119	struct sk_buff *skb, bool local_orig)
1da177e4	120	{
37b090e6	121	struct net_device *dev = BR_INPUT_SKB_CB(skb)->brdev;
1da177e4	122
37b090e6 NA	123	skb = skb_clone(skb, GFP_ATOMIC);
	124	if (!skb) {
	125	dev->stats.tx_dropped++;
	126	return -ENOMEM;
1da177e4 LT	127	}
1da177e4 LT	128
37b090e6 NA	129	__br_forward(prev, skb, local_orig);
37b090e6 NA	130	return 0;
1da177e4 LT	131	}
1da177e4 LT	132
37b090e6 NA	133	/**
	134	* br_forward - forward a packet to a specific port
	135	* @to: destination port
	136	* @skb: packet being forwarded
	137	* @local_rcv: packet will be received locally after forwarding
	138	* @local_orig: packet is locally originated
	139	*
	140	* Should be called with rcu_read_lock.
	141	*/
	142	void br_forward(const struct net_bridge_port *to,
	143	struct sk_buff *skb, bool local_rcv, bool local_orig)
1da177e4	144	{
2756f68c NA	145	if (unlikely(!to))
	146	goto out;
	147
	148	/* redirect to backup link if the destination port is down */
	149	if (rcu_access_pointer(to->backup_port) && !netif_carrier_ok(to->dev)) {
	150	struct net_bridge_port *backup_port;
	151
	152	backup_port = rcu_dereference(to->backup_port);
	153	if (unlikely(!backup_port))
	154	goto out;
	155	to = backup_port;
	156	}
	157
	158	if (should_deliver(to, skb)) {
b35c5f63	159	if (local_rcv)
37b090e6	160	deliver_clone(to, skb, local_orig);
7f7708f0	161	else
37b090e6	162	__br_forward(to, skb, local_orig);
1da177e4 LT	163	return;
	164	}
	165
2756f68c	166	out:
b35c5f63	167	if (!local_rcv)
7f7708f0	168	kfree_skb(skb);
1da177e4	169	}
37b090e6	170	EXPORT_SYMBOL_GPL(br_forward);
025d89c2 HX	171
	172	static struct net_bridge_port *maybe_deliver(
	173	struct net_bridge_port prev, struct net_bridge_port p,
37b090e6	174	struct sk_buff *skb, bool local_orig)
025d89c2 HX	175	{
	176	int err;
	177
	178	if (!should_deliver(p, skb))
	179	return prev;
	180
	181	if (!prev)
	182	goto out;
	183
37b090e6	184	err = deliver_clone(prev, skb, local_orig);
025d89c2 HX	185	if (err)
	186	return ERR_PTR(err);
	187
	188	out:
	189	return p;
	190	}
	191
37b090e6 NA	192	/* called under rcu_read_lock */
37b090e6 NA	193	void br_flood(struct net_bridge br, struct sk_buff skb,
8addd5e7	194	enum br_pkt_type pkt_type, bool local_rcv, bool local_orig)
1da177e4	195	{
1080ab95	196	u8 igmp_type = br_multicast_igmp_type(skb);
37b090e6	197	struct net_bridge_port *prev = NULL;
1080ab95	198	struct net_bridge_port *p;
1da177e4	199
1da177e4	200	list_for_each_entry_rcu(p, &br->port_list, list) {
99f906e9 MM	201	/* Do not flood unicast traffic to ports that turn it off, nor
	202	* other traffic if flood off, except for traffic we originate
	203	*/
	204	switch (pkt_type) {
	205	case BR_PKT_UNICAST:
	206	if (!(p->flags & BR_FLOOD))
	207	continue;
	208	break;
	209	case BR_PKT_MULTICAST:
	210	if (!(p->flags & BR_MCAST_FLOOD) && skb->dev != br->dev)
	211	continue;
	212	break;
	213	case BR_PKT_BROADCAST:
	214	if (!(p->flags & BR_BCAST_FLOOD) && skb->dev != br->dev)
	215	continue;
	216	break;
	217	}
95850116 KP	218
	219	/* Do not flood to ports that enable proxy ARP */
	220	if (p->flags & BR_PROXYARP)
	221	continue;
821f1b21	222	if ((p->flags & (BR_PROXYARP_WIFI \| BR_NEIGH_SUPPRESS)) &&
842a9ae0 JM	223	BR_INPUT_SKB_CB(skb)->proxyarp_replied)
842a9ae0 JM	224	continue;
95850116	225
37b090e6	226	prev = maybe_deliver(prev, p, skb, local_orig);
025d89c2 HX	227	if (IS_ERR(prev))
025d89c2 HX	228	goto out;
1080ab95	229	if (prev == p)
a65056ec	230	br_multicast_count(p->br, p, skb, igmp_type,
1080ab95	231	BR_MCAST_DIR_TX);
1da177e4 LT	232	}
1da177e4 LT	233
b33084be HX	234	if (!prev)
	235	goto out;
	236
b35c5f63	237	if (local_rcv)
37b090e6	238	deliver_clone(prev, skb, local_orig);
025d89c2	239	else
37b090e6	240	__br_forward(prev, skb, local_orig);
b33084be	241	return;
1da177e4	242
b33084be	243	out:
b35c5f63	244	if (!local_rcv)
b33084be	245	kfree_skb(skb);
1da177e4 LT	246	}
1da177e4 LT	247
5cb5e947	248	#ifdef CONFIG_BRIDGE_IGMP_SNOOPING
5b9d6b15 AB	249	static void maybe_deliver_addr(struct net_bridge_port p, struct sk_buff skb,
	250	const unsigned char *addr, bool local_orig)
	251	{
	252	struct net_device *dev = BR_INPUT_SKB_CB(skb)->brdev;
	253	const unsigned char *src = eth_hdr(skb)->h_source;
	254
	255	if (!should_deliver(p, skb))
	256	return;
	257
	258	/* Even with hairpin, no soliloquies - prevent breaking IPv6 DAD */
	259	if (skb->dev == p->dev && ether_addr_equal(src, addr))
	260	return;
	261
	262	skb = skb_copy(skb, GFP_ATOMIC);
	263	if (!skb) {
	264	dev->stats.tx_dropped++;
	265	return;
	266	}
	267
	268	if (!is_broadcast_ether_addr(addr))
	269	memcpy(eth_hdr(skb)->h_dest, addr, ETH_ALEN);
	270
	271	__br_forward(p, skb, local_orig);
	272	}
	273
5cb5e947	274	/* called with rcu_read_lock */
37b090e6 NA	275	void br_multicast_flood(struct net_bridge_mdb_entry *mdst,
	276	struct sk_buff *skb,
	277	bool local_rcv, bool local_orig)
5cb5e947 HX	278	{
5cb5e947 HX	279	struct net_device *dev = BR_INPUT_SKB_CB(skb)->brdev;
1080ab95	280	u8 igmp_type = br_multicast_igmp_type(skb);
5cb5e947	281	struct net_bridge *br = netdev_priv(dev);
afe0159d	282	struct net_bridge_port *prev = NULL;
5cb5e947 HX	283	struct net_bridge_port_group *p;
	284	struct hlist_node *rp;
	285
e8051688	286	rp = rcu_dereference(hlist_first_rcu(&br->router_list));
83f6a740	287	p = mdst ? rcu_dereference(mdst->ports) : NULL;
5cb5e947	288	while (p \|\| rp) {
afe0159d	289	struct net_bridge_port port, lport, *rport;
afe0159d	290
5cb5e947	291	lport = p ? p->port : NULL;
d8fb1648	292	rport = hlist_entry_safe(rp, struct net_bridge_port, rlist);
5cb5e947	293
6db6f0ea FF	294	if ((unsigned long)lport > (unsigned long)rport) {
	295	port = lport;
	296
	297	if (port->flags & BR_MULTICAST_TO_UNICAST) {
	298	maybe_deliver_addr(lport, skb, p->eth_addr,
	299	local_orig);
	300	goto delivered;
	301	}
	302	} else {
	303	port = rport;
	304	}
5cb5e947	305
37b090e6	306	prev = maybe_deliver(prev, port, skb, local_orig);
6db6f0ea	307	delivered:
5cb5e947 HX	308	if (IS_ERR(prev))
5cb5e947 HX	309	goto out;
1080ab95	310	if (prev == port)
a65056ec	311	br_multicast_count(port->br, port, skb, igmp_type,
1080ab95	312	BR_MCAST_DIR_TX);
5cb5e947 HX	313
5cb5e947 HX	314	if ((unsigned long)lport >= (unsigned long)port)
83f6a740	315	p = rcu_dereference(p->next);
5cb5e947	316	if ((unsigned long)rport >= (unsigned long)port)
e8051688	317	rp = rcu_dereference(hlist_next_rcu(rp));
5cb5e947 HX	318	}
	319
	320	if (!prev)
	321	goto out;
	322
b35c5f63	323	if (local_rcv)
37b090e6	324	deliver_clone(prev, skb, local_orig);
5cb5e947	325	else
37b090e6	326	__br_forward(prev, skb, local_orig);
5cb5e947 HX	327	return;
	328
	329	out:
b35c5f63	330	if (!local_rcv)
5cb5e947 HX	331	kfree_skb(skb);
5cb5e947 HX	332	}
5cb5e947	333	#endif