[linux-block.git] / net / bridge / br_input.c

// SPDX-License-Identifier: GPL-2.0-or-later
/*
 *	Handle incoming frames
 *	Linux ethernet bridge
 *
 *	Authors:
 *	Lennert Buytenhek		<buytenh@gnu.org>
 */

#include <linux/slab.h>
#include <linux/kernel.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/netfilter_bridge.h>
#ifdef CONFIG_NETFILTER_FAMILY_BRIDGE
#include <net/netfilter/nf_queue.h>
#endif
#include <linux/neighbour.h>
#include <net/arp.h>
#include <net/dsa.h>
#include <linux/export.h>
#include <linux/rculist.h>
#include "br_private.h"
#include "br_private_tunnel.h"

static int
br_netif_receive_skb(struct net *net, struct sock *sk, struct sk_buff *skb)
{
	br_drop_fake_rtable(skb);
	return netif_receive_skb(skb);
}

static int br_pass_frame_up(struct sk_buff *skb)
{
	struct net_device *indev, *brdev = BR_INPUT_SKB_CB(skb)->brdev;
	struct net_bridge *br = netdev_priv(brdev);
	struct net_bridge_vlan_group *vg;

	dev_sw_netstats_rx_add(brdev, skb->len);

	vg = br_vlan_group_rcu(br);

	/* Reset the offload_fwd_mark because there could be a stacked
	 * bridge above, and it should not think this bridge it doing
	 * that bridge's work forwarding out its ports.
	 */
	br_switchdev_frame_unmark(skb);

	/* Bridge is just like any other port.  Make sure the
	 * packet is allowed except in promisc mode when someone
	 * may be running packet capture.
	 */
	if (!(brdev->flags & IFF_PROMISC) &&
	    !br_allowed_egress(vg, skb)) {
		kfree_skb(skb);
		return NET_RX_DROP;
	}

	indev = skb->dev;
	skb->dev = brdev;
	skb = br_handle_vlan(br, NULL, vg, skb);
	if (!skb)
		return NET_RX_DROP;
	/* update the multicast stats if the packet is IGMP/MLD */
	br_multicast_count(br, NULL, skb, br_multicast_igmp_type(skb),
			   BR_MCAST_DIR_TX);

	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN,
		       dev_net(indev), NULL, skb, indev, NULL,
		       br_netif_receive_skb);
}

/* note: already called with rcu_read_lock */
int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
{
	struct net_bridge_port *p = br_port_get_rcu(skb->dev);
	enum br_pkt_type pkt_type = BR_PKT_UNICAST;
	struct net_bridge_fdb_entry *dst = NULL;
	struct net_bridge_mcast_port *pmctx;
	struct net_bridge_mdb_entry *mdst;
	bool local_rcv, mcast_hit = false;
	struct net_bridge_mcast *brmctx;
	struct net_bridge_vlan *vlan;
	struct net_bridge *br;
	u16 vid = 0;
	u8 state;

	if (!p)
		goto drop;

	br = p->br;

	if (br_mst_is_enabled(br)) {
		state = BR_STATE_FORWARDING;
	} else {
		if (p->state == BR_STATE_DISABLED)
			goto drop;

		state = p->state;
	}

	brmctx = &p->br->multicast_ctx;
	pmctx = &p->multicast_ctx;
	if (!br_allowed_ingress(p->br, nbp_vlan_group_rcu(p), skb, &vid,
				&state, &vlan))
		goto out;

	if (p->flags & BR_PORT_LOCKED) {
		struct net_bridge_fdb_entry *fdb_src =
			br_fdb_find_rcu(br, eth_hdr(skb)->h_source, vid);

		if (!fdb_src || READ_ONCE(fdb_src->dst) != p ||
		    test_bit(BR_FDB_LOCAL, &fdb_src->flags))
			goto drop;
	}

	nbp_switchdev_frame_mark(p, skb);

	/* insert into forwarding database after filtering to avoid spoofing */
	if (p->flags & BR_LEARNING)
		br_fdb_update(br, p, eth_hdr(skb)->h_source, vid, 0);

	local_rcv = !!(br->dev->flags & IFF_PROMISC);
	if (is_multicast_ether_addr(eth_hdr(skb)->h_dest)) {
		/* by definition the broadcast is also a multicast address */
		if (is_broadcast_ether_addr(eth_hdr(skb)->h_dest)) {
			pkt_type = BR_PKT_BROADCAST;
			local_rcv = true;
		} else {
			pkt_type = BR_PKT_MULTICAST;
			if (br_multicast_rcv(&brmctx, &pmctx, vlan, skb, vid))
				goto drop;
		}
	}

	if (state == BR_STATE_LEARNING)
		goto drop;

	BR_INPUT_SKB_CB(skb)->brdev = br->dev;
	BR_INPUT_SKB_CB(skb)->src_port_isolated = !!(p->flags & BR_ISOLATED);

	if (IS_ENABLED(CONFIG_INET) &&
	    (skb->protocol == htons(ETH_P_ARP) ||
	     skb->protocol == htons(ETH_P_RARP))) {
		br_do_proxy_suppress_arp(skb, br, vid, p);
	} else if (IS_ENABLED(CONFIG_IPV6) &&
		   skb->protocol == htons(ETH_P_IPV6) &&
		   br_opt_get(br, BROPT_NEIGH_SUPPRESS_ENABLED) &&
		   pskb_may_pull(skb, sizeof(struct ipv6hdr) +
				 sizeof(struct nd_msg)) &&
		   ipv6_hdr(skb)->nexthdr == IPPROTO_ICMPV6) {
			struct nd_msg *msg, _msg;

			msg = br_is_nd_neigh_msg(skb, &_msg);
			if (msg)
				br_do_suppress_nd(skb, br, vid, p, msg);
	}

	switch (pkt_type) {
	case BR_PKT_MULTICAST:
		mdst = br_mdb_get(brmctx, skb, vid);
		if ((mdst || BR_INPUT_SKB_CB_MROUTERS_ONLY(skb)) &&
		    br_multicast_querier_exists(brmctx, eth_hdr(skb), mdst)) {
			if ((mdst && mdst->host_joined) ||
			    br_multicast_is_router(brmctx, skb)) {
				local_rcv = true;
				br->dev->stats.multicast++;
			}
			mcast_hit = true;
		} else {
			local_rcv = true;
			br->dev->stats.multicast++;
		}
		break;
	case BR_PKT_UNICAST:
		dst = br_fdb_find_rcu(br, eth_hdr(skb)->h_dest, vid);
		break;
	default:
		break;
	}

	if (dst) {
		unsigned long now = jiffies;

		if (test_bit(BR_FDB_LOCAL, &dst->flags))
			return br_pass_frame_up(skb);

		if (now != dst->used)
			dst->used = now;
		br_forward(dst->dst, skb, local_rcv, false);
	} else {
		if (!mcast_hit)
			br_flood(br, skb, pkt_type, local_rcv, false);
		else
			br_multicast_flood(mdst, skb, brmctx, local_rcv, false);
	}

	if (local_rcv)
		return br_pass_frame_up(skb);

out:
	return 0;
drop:
	kfree_skb(skb);
	goto out;
}
EXPORT_SYMBOL_GPL(br_handle_frame_finish);

static void __br_handle_local_finish(struct sk_buff *skb)
{
	struct net_bridge_port *p = br_port_get_rcu(skb->dev);
	u16 vid = 0;

	/* check if vlan is allowed, to avoid spoofing */
	if ((p->flags & BR_LEARNING) &&
	    nbp_state_should_learn(p) &&
	    !br_opt_get(p->br, BROPT_NO_LL_LEARN) &&
	    br_should_learn(p, skb, &vid))
		br_fdb_update(p->br, p, eth_hdr(skb)->h_source, vid, 0);
}

/* note: already called with rcu_read_lock */
static int br_handle_local_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
{
	__br_handle_local_finish(skb);

	/* return 1 to signal the okfn() was called so it's ok to use the skb */
	return 1;
}

static int nf_hook_bridge_pre(struct sk_buff *skb, struct sk_buff **pskb)
{
#ifdef CONFIG_NETFILTER_FAMILY_BRIDGE
	struct nf_hook_entries *e = NULL;
	struct nf_hook_state state;
	unsigned int verdict, i;
	struct net *net;
	int ret;

	net = dev_net(skb->dev);
#ifdef HAVE_JUMP_LABEL
	if (!static_key_false(&nf_hooks_needed[NFPROTO_BRIDGE][NF_BR_PRE_ROUTING]))
		goto frame_finish;
#endif

	e = rcu_dereference(net->nf.hooks_bridge[NF_BR_PRE_ROUTING]);
	if (!e)
		goto frame_finish;

	nf_hook_state_init(&state, NF_BR_PRE_ROUTING,
			   NFPROTO_BRIDGE, skb->dev, NULL, NULL,
			   net, br_handle_frame_finish);

	for (i = 0; i < e->num_hook_entries; i++) {
		verdict = nf_hook_entry_hookfn(&e->hooks[i], skb, &state);
		switch (verdict & NF_VERDICT_MASK) {
		case NF_ACCEPT:
			if (BR_INPUT_SKB_CB(skb)->br_netfilter_broute) {
				*pskb = skb;
				return RX_HANDLER_PASS;
			}
			break;
		case NF_DROP:
			kfree_skb(skb);
			return RX_HANDLER_CONSUMED;
		case NF_QUEUE:
			ret = nf_queue(skb, &state, i, verdict);
			if (ret == 1)
				continue;
			return RX_HANDLER_CONSUMED;
		default: /* STOLEN */
			return RX_HANDLER_CONSUMED;
		}
	}
frame_finish:
	net = dev_net(skb->dev);
	br_handle_frame_finish(net, NULL, skb);
#else
	br_handle_frame_finish(dev_net(skb->dev), NULL, skb);
#endif
	return RX_HANDLER_CONSUMED;
}

/* Return 0 if the frame was not processed otherwise 1
 * note: already called with rcu_read_lock
 */
static int br_process_frame_type(struct net_bridge_port *p,
				 struct sk_buff *skb)
{
	struct br_frame_type *tmp;

	hlist_for_each_entry_rcu(tmp, &p->br->frame_type_list, list)
		if (unlikely(tmp->type == skb->protocol))
			return tmp->frame_handler(p, skb);

	return 0;
}

/*
 * Return NULL if skb is handled
 * note: already called with rcu_read_lock
 */
static rx_handler_result_t br_handle_frame(struct sk_buff **pskb)
{
	struct net_bridge_port *p;
	struct sk_buff *skb = *pskb;
	const unsigned char *dest = eth_hdr(skb)->h_dest;

	if (unlikely(skb->pkt_type == PACKET_LOOPBACK))
		return RX_HANDLER_PASS;

	if (!is_valid_ether_addr(eth_hdr(skb)->h_source))
		goto drop;

	skb = skb_share_check(skb, GFP_ATOMIC);
	if (!skb)
		return RX_HANDLER_CONSUMED;

	memset(skb->cb, 0, sizeof(struct br_input_skb_cb));

	p = br_port_get_rcu(skb->dev);
	if (p->flags & BR_VLAN_TUNNEL)
		br_handle_ingress_vlan_tunnel(skb, p, nbp_vlan_group_rcu(p));

	if (unlikely(is_link_local_ether_addr(dest))) {
		u16 fwd_mask = p->br->group_fwd_mask_required;

		/*
		 * See IEEE 802.1D Table 7-10 Reserved addresses
		 *
		 * Assignment		 		Value
		 * Bridge Group Address		01-80-C2-00-00-00
		 * (MAC Control) 802.3		01-80-C2-00-00-01
		 * (Link Aggregation) 802.3	01-80-C2-00-00-02
		 * 802.1X PAE address		01-80-C2-00-00-03
		 *
		 * 802.1AB LLDP 		01-80-C2-00-00-0E
		 *
		 * Others reserved for future standardization
		 */
		fwd_mask |= p->group_fwd_mask;
		switch (dest[5]) {
		case 0x00:	/* Bridge Group Address */
			/* If STP is turned off,
			   then must forward to keep loop detection */
			if (p->br->stp_enabled == BR_NO_STP ||
			    fwd_mask & (1u << dest[5]))
				goto forward;
			*pskb = skb;
			__br_handle_local_finish(skb);
			return RX_HANDLER_PASS;

		case 0x01:	/* IEEE MAC (Pause) */
			goto drop;

		case 0x0E:	/* 802.1AB LLDP */
			fwd_mask |= p->br->group_fwd_mask;
			if (fwd_mask & (1u << dest[5]))
				goto forward;
			*pskb = skb;
			__br_handle_local_finish(skb);
			return RX_HANDLER_PASS;

		default:
			/* Allow selective forwarding for most other protocols */
			fwd_mask |= p->br->group_fwd_mask;
			if (fwd_mask & (1u << dest[5]))
				goto forward;
		}

		/* The else clause should be hit when nf_hook():
		 *   - returns < 0 (drop/error)
		 *   - returns = 0 (stolen/nf_queue)
		 * Thus return 1 from the okfn() to signal the skb is ok to pass
		 */
		if (NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN,
			    dev_net(skb->dev), NULL, skb, skb->dev, NULL,
			    br_handle_local_finish) == 1) {
			return RX_HANDLER_PASS;
		} else {
			return RX_HANDLER_CONSUMED;
		}
	}

	if (unlikely(br_process_frame_type(p, skb)))
		return RX_HANDLER_PASS;

forward:
	if (br_mst_is_enabled(p->br))
		goto defer_stp_filtering;

	switch (p->state) {
	case BR_STATE_FORWARDING:
	case BR_STATE_LEARNING:
defer_stp_filtering:
		if (ether_addr_equal(p->br->dev->dev_addr, dest))
			skb->pkt_type = PACKET_HOST;

		return nf_hook_bridge_pre(skb, pskb);
	default:
drop:
		kfree_skb(skb);
	}
	return RX_HANDLER_CONSUMED;
}

/* This function has no purpose other than to appease the br_port_get_rcu/rtnl
 * helpers which identify bridged ports according to the rx_handler installed
 * on them (so there _needs_ to be a bridge rx_handler even if we don't need it
 * to do anything useful). This bridge won't support traffic to/from the stack,
 * but only hardware bridging. So return RX_HANDLER_PASS so we don't steal
 * frames from the ETH_P_XDSA packet_type handler.
 */
static rx_handler_result_t br_handle_frame_dummy(struct sk_buff **pskb)
{
	return RX_HANDLER_PASS;
}

rx_handler_func_t *br_get_rx_handler(const struct net_device *dev)
{
	if (netdev_uses_dsa(dev))
		return br_handle_frame_dummy;

	return br_handle_frame;
}

void br_add_frame(struct net_bridge *br, struct br_frame_type *ft)
{
	hlist_add_head_rcu(&ft->list, &br->frame_type_list);
}

void br_del_frame(struct net_bridge *br, struct br_frame_type *ft)
{
	struct br_frame_type *tmp;

	hlist_for_each_entry(tmp, &br->frame_type_list, list)
		if (ft == tmp) {
			hlist_del_rcu(&ft->list);
			return;
		}
}
Commit	Line	Data
2874c5fd	1	// SPDX-License-Identifier: GPL-2.0-or-later
1da177e4 LT	2	/*
	3	* Handle incoming frames
	4	* Linux ethernet bridge
	5	*
	6	* Authors:
	7	* Lennert Buytenhek <buytenh@gnu.org>
1da177e4 LT	8	*/
1da177e4 LT	9
5a0e3ad6	10	#include <linux/slab.h>
1da177e4 LT	11	#include <linux/kernel.h>
	12	#include <linux/netdevice.h>
	13	#include <linux/etherdevice.h>
	14	#include <linux/netfilter_bridge.h>
dc2f4189	15	#ifdef CONFIG_NETFILTER_FAMILY_BRIDGE
971502d7	16	#include <net/netfilter/nf_queue.h>
dc2f4189	17	#endif
95850116 KP	18	#include <linux/neighbour.h>
95850116 KP	19	#include <net/arp.h>
9eb8eff0	20	#include <net/dsa.h>
bc3b2d7f	21	#include <linux/export.h>
a37b85c9	22	#include <linux/rculist.h>
1da177e4	23	#include "br_private.h"
11538d03	24	#include "br_private_tunnel.h"
1da177e4	25
0c4b51f0 EB	26	static int
0c4b51f0 EB	27	br_netif_receive_skb(struct net net, struct sock sk, struct sk_buff *skb)
04eb4489	28	{
a13b2082	29	br_drop_fake_rtable(skb);
04eb4489 EB	30	return netif_receive_skb(skb);
	31	}
	32
68b7c895	33	static int br_pass_frame_up(struct sk_buff *skb)
1da177e4	34	{
68b7c895	35	struct net_device indev, brdev = BR_INPUT_SKB_CB(skb)->brdev;
14bb4789	36	struct net_bridge *br = netdev_priv(brdev);
2594e906	37	struct net_bridge_vlan_group *vg;
1da177e4	38
7609ecb2	39	dev_sw_netstats_rx_add(brdev, skb->len);
1da177e4	40
907b1e6e	41	vg = br_vlan_group_rcu(br);
fbb3abdf AL	42
	43	/* Reset the offload_fwd_mark because there could be a stacked
	44	* bridge above, and it should not think this bridge it doing
	45	* that bridge's work forwarding out its ports.
	46	*/
	47	br_switchdev_frame_unmark(skb);
	48
85f46c6b	49	/* Bridge is just like any other port. Make sure the
efb5b338	50	* packet is allowed except in promisc mode when someone
85f46c6b VY	51	* may be running packet capture.
	52	*/
	53	if (!(brdev->flags & IFF_PROMISC) &&
2594e906	54	!br_allowed_egress(vg, skb)) {
85f46c6b VY	55	kfree_skb(skb);
	56	return NET_RX_DROP;
	57	}
	58
1da177e4	59	indev = skb->dev;
a339f1c8	60	skb->dev = brdev;
11538d03	61	skb = br_handle_vlan(br, NULL, vg, skb);
fc92f745 VY	62	if (!skb)
fc92f745 VY	63	return NET_RX_DROP;
1080ab95	64	/* update the multicast stats if the packet is IGMP/MLD */
a65056ec	65	br_multicast_count(br, NULL, skb, br_multicast_igmp_type(skb),
1080ab95	66	BR_MCAST_DIR_TX);
1da177e4	67
29a26a56 EB	68	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN,
29a26a56 EB	69	dev_net(indev), NULL, skb, indev, NULL,
04eb4489	70	br_netif_receive_skb);
1da177e4 LT	71	}
1da177e4 LT	72
eeaf61d8	73	/* note: already called with rcu_read_lock */
0c4b51f0	74	int br_handle_frame_finish(struct net net, struct sock sk, struct sk_buff *skb)
1da177e4	75	{
f350a0a8	76	struct net_bridge_port *p = br_port_get_rcu(skb->dev);
8addd5e7	77	enum br_pkt_type pkt_type = BR_PKT_UNICAST;
46c0772d	78	struct net_bridge_fdb_entry *dst = NULL;
adc47037	79	struct net_bridge_mcast_port *pmctx;
c4fcb78c	80	struct net_bridge_mdb_entry *mdst;
8addd5e7	81	bool local_rcv, mcast_hit = false;
adc47037	82	struct net_bridge_mcast *brmctx;
f4b7002a	83	struct net_bridge_vlan *vlan;
46c0772d	84	struct net_bridge *br;
78851988	85	u16 vid = 0;
a580c76d	86	u8 state;
1da177e4	87
ec7328b5	88	if (!p)
b3f1be4b SH	89	goto drop;
b3f1be4b SH	90
a21d9a67	91	br = p->br;
ec7328b5 TW	92
	93	if (br_mst_is_enabled(br)) {
	94	state = BR_STATE_FORWARDING;
	95	} else {
	96	if (p->state == BR_STATE_DISABLED)
	97	goto drop;
	98
	99	state = p->state;
	100	}
	101
adc47037 NA	102	brmctx = &p->br->multicast_ctx;
adc47037 NA	103	pmctx = &p->multicast_ctx;
a580c76d	104	if (!br_allowed_ingress(p->br, nbp_vlan_group_rcu(p), skb, &vid,
f4b7002a	105	&state, &vlan))
eb707618	106	goto out;
a37b85c9	107
a21d9a67 HS	108	if (p->flags & BR_PORT_LOCKED) {
	109	struct net_bridge_fdb_entry *fdb_src =
	110	br_fdb_find_rcu(br, eth_hdr(skb)->h_source, vid);
	111
	112	if (!fdb_src \|\| READ_ONCE(fdb_src->dst) != p \|\|
	113	test_bit(BR_FDB_LOCAL, &fdb_src->flags))
	114	goto drop;
	115	}
	116
6bc506b4 IS	117	nbp_switchdev_frame_mark(p, skb);
6bc506b4 IS	118
85967bb4	119	/* insert into forwarding database after filtering to avoid spoofing */
9ba18891	120	if (p->flags & BR_LEARNING)
be0c5677	121	br_fdb_update(br, p, eth_hdr(skb)->h_source, vid, 0);
85967bb4	122
8addd5e7	123	local_rcv = !!(br->dev->flags & IFF_PROMISC);
3d26eb8a	124	if (is_multicast_ether_addr(eth_hdr(skb)->h_dest)) {
8addd5e7	125	/* by definition the broadcast is also a multicast address */
3d26eb8a	126	if (is_broadcast_ether_addr(eth_hdr(skb)->h_dest)) {
8addd5e7 NA	127	pkt_type = BR_PKT_BROADCAST;
	128	local_rcv = true;
	129	} else {
	130	pkt_type = BR_PKT_MULTICAST;
f4b7002a	131	if (br_multicast_rcv(&brmctx, &pmctx, vlan, skb, vid))
8addd5e7 NA	132	goto drop;
	133	}
	134	}
c4fcb78c	135
a580c76d	136	if (state == BR_STATE_LEARNING)
b3f1be4b	137	goto drop;
0e5eabac	138
68b7c895	139	BR_INPUT_SKB_CB(skb)->brdev = br->dev;
7d850abd	140	BR_INPUT_SKB_CB(skb)->src_port_isolated = !!(p->flags & BR_ISOLATED);
68b7c895	141
057658cb RP	142	if (IS_ENABLED(CONFIG_INET) &&
	143	(skb->protocol == htons(ETH_P_ARP) \|\|
	144	skb->protocol == htons(ETH_P_RARP))) {
	145	br_do_proxy_suppress_arp(skb, br, vid, p);
ed842fae RP	146	} else if (IS_ENABLED(CONFIG_IPV6) &&
ed842fae RP	147	skb->protocol == htons(ETH_P_IPV6) &&
c69c2cd4	148	br_opt_get(br, BROPT_NEIGH_SUPPRESS_ENABLED) &&
ed842fae RP	149	pskb_may_pull(skb, sizeof(struct ipv6hdr) +
	150	sizeof(struct nd_msg)) &&
	151	ipv6_hdr(skb)->nexthdr == IPPROTO_ICMPV6) {
	152	struct nd_msg *msg, _msg;
	153
	154	msg = br_is_nd_neigh_msg(skb, &_msg);
	155	if (msg)
	156	br_do_suppress_nd(skb, br, vid, p, msg);
057658cb	157	}
95850116	158
8addd5e7 NA	159	switch (pkt_type) {
8addd5e7 NA	160	case BR_PKT_MULTICAST:
adc47037	161	mdst = br_mdb_get(brmctx, skb, vid);
b00589af	162	if ((mdst \|\| BR_INPUT_SKB_CB_MROUTERS_ONLY(skb)) &&
adc47037	163	br_multicast_querier_exists(brmctx, eth_hdr(skb), mdst)) {
ff0fd34e	164	if ((mdst && mdst->host_joined) \|\|
adc47037	165	br_multicast_is_router(brmctx, skb)) {
b35c5f63	166	local_rcv = true;
e151aab9 NA	167	br->dev->stats.multicast++;
	168	}
	169	mcast_hit = true;
46c0772d	170	} else {
b35c5f63	171	local_rcv = true;
e151aab9	172	br->dev->stats.multicast++;
46c0772d	173	}
8addd5e7 NA	174	break;
8addd5e7 NA	175	case BR_PKT_UNICAST:
3d26eb8a	176	dst = br_fdb_find_rcu(br, eth_hdr(skb)->h_dest, vid);
ecd1c6a5	177	break;
8addd5e7 NA	178	default:
8addd5e7 NA	179	break;
1da177e4 LT	180	}
1da177e4 LT	181
e151aab9	182	if (dst) {
ca6d4480	183	unsigned long now = jiffies;
ca6d4480	184
6869c3b0	185	if (test_bit(BR_FDB_LOCAL, &dst->flags))
8addd5e7 NA	186	return br_pass_frame_up(skb);
8addd5e7 NA	187
ca6d4480	188	if (now != dst->used)
ca6d4480	189	dst->used = now;
37b090e6	190	br_forward(dst->dst, skb, local_rcv, false);
e151aab9 NA	191	} else {
e151aab9 NA	192	if (!mcast_hit)
8addd5e7	193	br_flood(br, skb, pkt_type, local_rcv, false);
e151aab9	194	else
adc47037	195	br_multicast_flood(mdst, skb, brmctx, local_rcv, false);
e081e1e3	196	}
1da177e4	197
b35c5f63 NA	198	if (local_rcv)
b35c5f63 NA	199	return br_pass_frame_up(skb);
87557c18	200
1da177e4 LT	201	out:
1da177e4 LT	202	return 0;
b3f1be4b SH	203	drop:
	204	kfree_skb(skb);
	205	goto out;
1da177e4	206	}
34666d46	207	EXPORT_SYMBOL_GPL(br_handle_frame_finish);
1da177e4	208
56fae404	209	static void __br_handle_local_finish(struct sk_buff *skb)
cf0f02d0	210	{
f350a0a8	211	struct net_bridge_port *p = br_port_get_rcu(skb->dev);
2ba071ec	212	u16 vid = 0;
cf0f02d0	213
e0d7968a	214	/* check if vlan is allowed, to avoid spoofing */
70e4272b	215	if ((p->flags & BR_LEARNING) &&
5d1fcaf3	216	nbp_state_should_learn(p) &&
70e4272b NA	217	!br_opt_get(p->br, BROPT_NO_LL_LEARN) &&
70e4272b NA	218	br_should_learn(p, skb, &vid))
be0c5677	219	br_fdb_update(p->br, p, eth_hdr(skb)->h_source, vid, 0);
56fae404 IS	220	}
	221
	222	/* note: already called with rcu_read_lock */
	223	static int br_handle_local_finish(struct net net, struct sock sk, struct sk_buff *skb)
	224	{
56fae404	225	__br_handle_local_finish(skb);
8626c56c	226
3b2e2904 NA	227	/* return 1 to signal the okfn() was called so it's ok to use the skb */
3b2e2904 NA	228	return 1;
cf0f02d0 SH	229	}
cf0f02d0 SH	230
971502d7 FW	231	static int nf_hook_bridge_pre(struct sk_buff skb, struct sk_buff *pskb)
	232	{
	233	#ifdef CONFIG_NETFILTER_FAMILY_BRIDGE
	234	struct nf_hook_entries *e = NULL;
	235	struct nf_hook_state state;
	236	unsigned int verdict, i;
	237	struct net *net;
	238	int ret;
	239
	240	net = dev_net(skb->dev);
	241	#ifdef HAVE_JUMP_LABEL
	242	if (!static_key_false(&nf_hooks_needed[NFPROTO_BRIDGE][NF_BR_PRE_ROUTING]))
	243	goto frame_finish;
	244	#endif
	245
	246	e = rcu_dereference(net->nf.hooks_bridge[NF_BR_PRE_ROUTING]);
	247	if (!e)
	248	goto frame_finish;
	249
	250	nf_hook_state_init(&state, NF_BR_PRE_ROUTING,
	251	NFPROTO_BRIDGE, skb->dev, NULL, NULL,
	252	net, br_handle_frame_finish);
	253
	254	for (i = 0; i < e->num_hook_entries; i++) {
	255	verdict = nf_hook_entry_hookfn(&e->hooks[i], skb, &state);
	256	switch (verdict & NF_VERDICT_MASK) {
	257	case NF_ACCEPT:
223fd0ad FW	258	if (BR_INPUT_SKB_CB(skb)->br_netfilter_broute) {
	259	*pskb = skb;
	260	return RX_HANDLER_PASS;
	261	}
971502d7 FW	262	break;
	263	case NF_DROP:
	264	kfree_skb(skb);
	265	return RX_HANDLER_CONSUMED;
	266	case NF_QUEUE:
0d9cb300	267	ret = nf_queue(skb, &state, i, verdict);
971502d7 FW	268	if (ret == 1)
	269	continue;
	270	return RX_HANDLER_CONSUMED;
	271	default: /* STOLEN */
	272	return RX_HANDLER_CONSUMED;
	273	}
	274	}
	275	frame_finish:
	276	net = dev_net(skb->dev);
	277	br_handle_frame_finish(net, NULL, skb);
	278	#else
	279	br_handle_frame_finish(dev_net(skb->dev), NULL, skb);
	280	#endif
	281	return RX_HANDLER_CONSUMED;
	282	}
	283
90c628dd HB	284	/* Return 0 if the frame was not processed otherwise 1
	285	* note: already called with rcu_read_lock
	286	*/
	287	static int br_process_frame_type(struct net_bridge_port *p,
	288	struct sk_buff *skb)
	289	{
	290	struct br_frame_type *tmp;
	291
	292	hlist_for_each_entry_rcu(tmp, &p->br->frame_type_list, list)
	293	if (unlikely(tmp->type == skb->protocol))
	294	return tmp->frame_handler(p, skb);
	295
	296	return 0;
	297	}
	298
1da177e4	299	/*
6229e362	300	* Return NULL if skb is handled
eeaf61d8	301	* note: already called with rcu_read_lock
1da177e4	302	*/
9eb8eff0	303	static rx_handler_result_t br_handle_frame(struct sk_buff **pskb)
1da177e4	304	{
ab95bfe0	305	struct net_bridge_port *p;
8a4eb573	306	struct sk_buff skb = pskb;
1da177e4 LT	307	const unsigned char *dest = eth_hdr(skb)->h_dest;
1da177e4 LT	308
c2368e79	309	if (unlikely(skb->pkt_type == PACKET_LOOPBACK))
8a4eb573	310	return RX_HANDLER_PASS;
ab95bfe0	311
1da177e4	312	if (!is_valid_ether_addr(eth_hdr(skb)->h_source))
467aea0d	313	goto drop;
1da177e4	314
7b995651 HX	315	skb = skb_share_check(skb, GFP_ATOMIC);
7b995651 HX	316	if (!skb)
8a4eb573	317	return RX_HANDLER_CONSUMED;
7b995651	318
f12064d1 FW	319	memset(skb->cb, 0, sizeof(struct br_input_skb_cb));
f12064d1 FW	320
f350a0a8	321	p = br_port_get_rcu(skb->dev);
a37c5c26 KP	322	if (p->flags & BR_VLAN_TUNNEL)
a37c5c26 KP	323	br_handle_ingress_vlan_tunnel(skb, p, nbp_vlan_group_rcu(p));
ab95bfe0	324
46acc460	325	if (unlikely(is_link_local_ether_addr(dest))) {
f2808d22 TM	326	u16 fwd_mask = p->br->group_fwd_mask_required;
f2808d22 TM	327
515853cc	328	/*
	329	* See IEEE 802.1D Table 7-10 Reserved addresses
	330	*
	331	* Assignment Value
	332	* Bridge Group Address 01-80-C2-00-00-00
	333	* (MAC Control) 802.3 01-80-C2-00-00-01
	334	* (Link Aggregation) 802.3 01-80-C2-00-00-02
	335	* 802.1X PAE address 01-80-C2-00-00-03
	336	*
	337	* 802.1AB LLDP 01-80-C2-00-00-0E
	338	*
	339	* Others reserved for future standardization
	340	*/
5af48b59	341	fwd_mask \|= p->group_fwd_mask;
515853cc	342	switch (dest[5]) {
	343	case 0x00: /* Bridge Group Address */
	344	/* If STP is turned off,
	345	then must forward to keep loop detection */
f2808d22 TM	346	if (p->br->stp_enabled == BR_NO_STP \|\|
f2808d22 TM	347	fwd_mask & (1u << dest[5]))
515853cc	348	goto forward;
56fae404 IS	349	*pskb = skb;
	350	__br_handle_local_finish(skb);
	351	return RX_HANDLER_PASS;
515853cc	352
515853cc	353	case 0x01: /* IEEE MAC (Pause) */
2111f8b9 SH	354	goto drop;
2111f8b9 SH	355
baedbe55 IS	356	case 0x0E: /* 802.1AB LLDP */
	357	fwd_mask \|= p->br->group_fwd_mask;
	358	if (fwd_mask & (1u << dest[5]))
	359	goto forward;
	360	*pskb = skb;
	361	__br_handle_local_finish(skb);
	362	return RX_HANDLER_PASS;
	363
515853cc	364	default:
515853cc	365	/* Allow selective forwarding for most other protocols */
f2808d22 TM	366	fwd_mask \|= p->br->group_fwd_mask;
f2808d22 TM	367	if (fwd_mask & (1u << dest[5]))
515853cc	368	goto forward;
515853cc	369	}
a598f6ae	370
3b2e2904 NA	371	/* The else clause should be hit when nf_hook():
	372	* - returns < 0 (drop/error)
	373	* - returns = 0 (stolen/nf_queue)
	374	* Thus return 1 from the okfn() to signal the skb is ok to pass
	375	*/
	376	if (NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN,
	377	dev_net(skb->dev), NULL, skb, skb->dev, NULL,
	378	br_handle_local_finish) == 1) {
	379	return RX_HANDLER_PASS;
	380	} else {
	381	return RX_HANDLER_CONSUMED;
	382	}
2111f8b9	383	}
1da177e4	384
90c628dd	385	if (unlikely(br_process_frame_type(p, skb)))
65369933 HV	386	return RX_HANDLER_PASS;
65369933 HV	387
a598f6ae	388	forward:
ec7328b5 TW	389	if (br_mst_is_enabled(p->br))
	390	goto defer_stp_filtering;
	391
467aea0d SH	392	switch (p->state) {
467aea0d SH	393	case BR_STATE_FORWARDING:
467aea0d	394	case BR_STATE_LEARNING:
ec7328b5	395	defer_stp_filtering:
9a7b6ef9	396	if (ether_addr_equal(p->br->dev->dev_addr, dest))
1da177e4 LT	397	skb->pkt_type = PACKET_HOST;
1da177e4 LT	398
971502d7	399	return nf_hook_bridge_pre(skb, pskb);
467aea0d SH	400	default:
	401	drop:
	402	kfree_skb(skb);
1da177e4	403	}
8a4eb573	404	return RX_HANDLER_CONSUMED;
1da177e4	405	}
9eb8eff0 VO	406
	407	/* This function has no purpose other than to appease the br_port_get_rcu/rtnl
	408	* helpers which identify bridged ports according to the rx_handler installed
	409	* on them (so there _needs_ to be a bridge rx_handler even if we don't need it
	410	* to do anything useful). This bridge won't support traffic to/from the stack,
	411	* but only hardware bridging. So return RX_HANDLER_PASS so we don't steal
	412	* frames from the ETH_P_XDSA packet_type handler.
	413	*/
	414	static rx_handler_result_t br_handle_frame_dummy(struct sk_buff **pskb)
	415	{
	416	return RX_HANDLER_PASS;
	417	}
	418
	419	rx_handler_func_t br_get_rx_handler(const struct net_device dev)
	420	{
	421	if (netdev_uses_dsa(dev))
	422	return br_handle_frame_dummy;
	423
	424	return br_handle_frame;
	425	}
90c628dd HB	426
	427	void br_add_frame(struct net_bridge br, struct br_frame_type ft)
	428	{
	429	hlist_add_head_rcu(&ft->list, &br->frame_type_list);
	430	}
	431
	432	void br_del_frame(struct net_bridge br, struct br_frame_type ft)
	433	{
	434	struct br_frame_type *tmp;
	435
	436	hlist_for_each_entry(tmp, &br->frame_type_list, list)
	437	if (ft == tmp) {
	438	hlist_del_rcu(&ft->list);
	439	return;
	440	}
	441	}