[net-next-2.6.git] / drivers / net / macvlan.c

/*
 * Copyright (c) 2007 Patrick McHardy <kaber@trash.net>
 *
 * 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.
 *
 * The code this is based on carried the following copyright notice:
 * ---
 * (C) Copyright 2001-2006
 * Alex Zeffertt, Cambridge Broadband Ltd, ajz@cambridgebroadband.com
 * Re-worked by Ben Greear <greearb@candelatech.com>
 * ---
 */
#include <linux/kernel.h>
#include <linux/types.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/errno.h>
#include <linux/slab.h>
#include <linux/string.h>
#include <linux/rculist.h>
#include <linux/notifier.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/ethtool.h>
#include <linux/if_arp.h>
#include <linux/if_link.h>
#include <linux/if_macvlan.h>
#include <net/rtnetlink.h>

#define MACVLAN_HASH_SIZE	(1 << BITS_PER_BYTE)

struct macvlan_port {
	struct net_device	*dev;
	struct hlist_head	vlan_hash[MACVLAN_HASH_SIZE];
	struct list_head	vlans;
};

struct macvlan_dev {
	struct net_device	*dev;
	struct list_head	list;
	struct hlist_node	hlist;
	struct macvlan_port	*port;
	struct net_device	*lowerdev;
};


static struct macvlan_dev *macvlan_hash_lookup(const struct macvlan_port *port,
					       const unsigned char *addr)
{
	struct macvlan_dev *vlan;
	struct hlist_node *n;

	hlist_for_each_entry_rcu(vlan, n, &port->vlan_hash[addr[5]], hlist) {
		if (!compare_ether_addr(vlan->dev->dev_addr, addr))
			return vlan;
	}
	return NULL;
}

static void macvlan_broadcast(struct sk_buff *skb,
			      const struct macvlan_port *port)
{
	const struct ethhdr *eth = eth_hdr(skb);
	const struct macvlan_dev *vlan;
	struct hlist_node *n;
	struct net_device *dev;
	struct sk_buff *nskb;
	unsigned int i;

	for (i = 0; i < MACVLAN_HASH_SIZE; i++) {
		hlist_for_each_entry_rcu(vlan, n, &port->vlan_hash[i], hlist) {
			dev = vlan->dev;

			nskb = skb_clone(skb, GFP_ATOMIC);
			if (nskb == NULL) {
				dev->stats.rx_errors++;
				dev->stats.rx_dropped++;
				continue;
			}

			dev->stats.rx_bytes += skb->len + ETH_HLEN;
			dev->stats.rx_packets++;
			dev->stats.multicast++;

			nskb->dev = dev;
			if (!compare_ether_addr(eth->h_dest, dev->broadcast))
				nskb->pkt_type = PACKET_BROADCAST;
			else
				nskb->pkt_type = PACKET_MULTICAST;

			netif_rx(nskb);
		}
	}
}

/* called under rcu_read_lock() from netif_receive_skb */
static struct sk_buff *macvlan_handle_frame(struct sk_buff *skb)
{
	const struct ethhdr *eth = eth_hdr(skb);
	const struct macvlan_port *port;
	const struct macvlan_dev *vlan;
	struct net_device *dev;

	port = rcu_dereference(skb->dev->macvlan_port);
	if (port == NULL)
		return skb;

	if (is_multicast_ether_addr(eth->h_dest)) {
		macvlan_broadcast(skb, port);
		return skb;
	}

	vlan = macvlan_hash_lookup(port, eth->h_dest);
	if (vlan == NULL)
		return skb;

	dev = vlan->dev;
	if (unlikely(!(dev->flags & IFF_UP))) {
		kfree_skb(skb);
		return NULL;
	}

	skb = skb_share_check(skb, GFP_ATOMIC);
	if (skb == NULL) {
		dev->stats.rx_errors++;
		dev->stats.rx_dropped++;
		return NULL;
	}

	dev->stats.rx_bytes += skb->len + ETH_HLEN;
	dev->stats.rx_packets++;

	skb->dev = dev;
	skb->pkt_type = PACKET_HOST;

	netif_rx(skb);
	return NULL;
}

static int macvlan_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
	const struct macvlan_dev *vlan = netdev_priv(dev);
	unsigned int len = skb->len;
	int ret;

	skb->dev = vlan->lowerdev;
	ret = dev_queue_xmit(skb);

	if (likely(ret == NET_XMIT_SUCCESS)) {
		dev->stats.tx_packets++;
		dev->stats.tx_bytes += len;
	} else {
		dev->stats.tx_errors++;
		dev->stats.tx_aborted_errors++;
	}
	return NETDEV_TX_OK;
}

static int macvlan_hard_header(struct sk_buff *skb, struct net_device *dev,
			       unsigned short type, const void *daddr,
			       const void *saddr, unsigned len)
{
	const struct macvlan_dev *vlan = netdev_priv(dev);
	struct net_device *lowerdev = vlan->lowerdev;

	return dev_hard_header(skb, lowerdev, type, daddr,
			       saddr ? : dev->dev_addr, len);
}

static const struct header_ops macvlan_hard_header_ops = {
	.create  	= macvlan_hard_header,
	.rebuild	= eth_rebuild_header,
	.parse		= eth_header_parse,
	.cache		= eth_header_cache,
	.cache_update	= eth_header_cache_update,
};

static int macvlan_open(struct net_device *dev)
{
	struct macvlan_dev *vlan = netdev_priv(dev);
	struct macvlan_port *port = vlan->port;
	struct net_device *lowerdev = vlan->lowerdev;
	int err;

	err = dev_unicast_add(lowerdev, dev->dev_addr, ETH_ALEN);
	if (err < 0)
		goto out;
	if (dev->flags & IFF_ALLMULTI) {
		err = dev_set_allmulti(lowerdev, 1);
		if (err < 0)
			goto del_unicast;
	}

	hlist_add_head_rcu(&vlan->hlist, &port->vlan_hash[dev->dev_addr[5]]);
	return 0;

del_unicast:
	dev_unicast_delete(lowerdev, dev->dev_addr, ETH_ALEN);
out:
	return err;
}

static int macvlan_stop(struct net_device *dev)
{
	struct macvlan_dev *vlan = netdev_priv(dev);
	struct net_device *lowerdev = vlan->lowerdev;

	dev_mc_unsync(lowerdev, dev);
	if (dev->flags & IFF_ALLMULTI)
		dev_set_allmulti(lowerdev, -1);

	dev_unicast_delete(lowerdev, dev->dev_addr, ETH_ALEN);

	hlist_del_rcu(&vlan->hlist);
	synchronize_rcu();
	return 0;
}

static int macvlan_set_mac_address(struct net_device *dev, void *p)
{
	struct macvlan_dev *vlan = netdev_priv(dev);
	struct net_device *lowerdev = vlan->lowerdev;
	struct sockaddr *addr = p;
	int err;

	if (!is_valid_ether_addr(addr->sa_data))
		return -EADDRNOTAVAIL;

	if (!(dev->flags & IFF_UP))
		goto out;

	err = dev_unicast_add(lowerdev, addr->sa_data, ETH_ALEN);
	if (err < 0)
		return err;
	dev_unicast_delete(lowerdev, dev->dev_addr, ETH_ALEN);

out:
	memcpy(dev->dev_addr, addr->sa_data, ETH_ALEN);
	return 0;
}

static void macvlan_change_rx_flags(struct net_device *dev, int change)
{
	struct macvlan_dev *vlan = netdev_priv(dev);
	struct net_device *lowerdev = vlan->lowerdev;

	if (change & IFF_ALLMULTI)
		dev_set_allmulti(lowerdev, dev->flags & IFF_ALLMULTI ? 1 : -1);
}

static void macvlan_set_multicast_list(struct net_device *dev)
{
	struct macvlan_dev *vlan = netdev_priv(dev);

	dev_mc_sync(vlan->lowerdev, dev);
}

static int macvlan_change_mtu(struct net_device *dev, int new_mtu)
{
	struct macvlan_dev *vlan = netdev_priv(dev);

	if (new_mtu < 68 || vlan->lowerdev->mtu < new_mtu)
		return -EINVAL;
	dev->mtu = new_mtu;
	return 0;
}

/*
 * macvlan network devices have devices nesting below it and are a special
 * "super class" of normal network devices; split their locks off into a
 * separate class since they always nest.
 */
static struct lock_class_key macvlan_netdev_xmit_lock_key;
static struct lock_class_key macvlan_netdev_addr_lock_key;

#define MACVLAN_FEATURES \
	(NETIF_F_SG | NETIF_F_ALL_CSUM | NETIF_F_HIGHDMA | NETIF_F_FRAGLIST | \
	 NETIF_F_GSO | NETIF_F_TSO | NETIF_F_UFO | NETIF_F_GSO_ROBUST | \
	 NETIF_F_TSO_ECN | NETIF_F_TSO6)

#define MACVLAN_STATE_MASK \
	((1<<__LINK_STATE_NOCARRIER) | (1<<__LINK_STATE_DORMANT))

static void macvlan_set_lockdep_class_one(struct net_device *dev,
					  struct netdev_queue *txq,
					  void *_unused)
{
	lockdep_set_class(&txq->_xmit_lock,
			  &macvlan_netdev_xmit_lock_key);
}

static void macvlan_set_lockdep_class(struct net_device *dev)
{
	lockdep_set_class(&dev->addr_list_lock,
			  &macvlan_netdev_addr_lock_key);
	netdev_for_each_tx_queue(dev, macvlan_set_lockdep_class_one, NULL);
}

static int macvlan_init(struct net_device *dev)
{
	struct macvlan_dev *vlan = netdev_priv(dev);
	const struct net_device *lowerdev = vlan->lowerdev;

	dev->state		= (dev->state & ~MACVLAN_STATE_MASK) |
				  (lowerdev->state & MACVLAN_STATE_MASK);
	dev->features 		= lowerdev->features & MACVLAN_FEATURES;
	dev->iflink		= lowerdev->ifindex;

	macvlan_set_lockdep_class(dev);

	return 0;
}

static void macvlan_ethtool_get_drvinfo(struct net_device *dev,
					struct ethtool_drvinfo *drvinfo)
{
	snprintf(drvinfo->driver, 32, "macvlan");
	snprintf(drvinfo->version, 32, "0.1");
}

static u32 macvlan_ethtool_get_rx_csum(struct net_device *dev)
{
	const struct macvlan_dev *vlan = netdev_priv(dev);
	struct net_device *lowerdev = vlan->lowerdev;

	if (lowerdev->ethtool_ops->get_rx_csum == NULL)
		return 0;
	return lowerdev->ethtool_ops->get_rx_csum(lowerdev);
}

static int macvlan_ethtool_get_settings(struct net_device *dev,
					struct ethtool_cmd *cmd)
{
	const struct macvlan_dev *vlan = netdev_priv(dev);
	struct net_device *lowerdev = vlan->lowerdev;

	if (!lowerdev->ethtool_ops->get_settings)
		return -EOPNOTSUPP;

	return lowerdev->ethtool_ops->get_settings(lowerdev, cmd);
}

static u32 macvlan_ethtool_get_flags(struct net_device *dev)
{
	const struct macvlan_dev *vlan = netdev_priv(dev);
	struct net_device *lowerdev = vlan->lowerdev;

	if (!lowerdev->ethtool_ops->get_flags)
		return 0;
	return lowerdev->ethtool_ops->get_flags(lowerdev);
}

static const struct ethtool_ops macvlan_ethtool_ops = {
	.get_link		= ethtool_op_get_link,
	.get_settings		= macvlan_ethtool_get_settings,
	.get_rx_csum		= macvlan_ethtool_get_rx_csum,
	.get_drvinfo		= macvlan_ethtool_get_drvinfo,
	.get_flags		= macvlan_ethtool_get_flags,
};

static const struct net_device_ops macvlan_netdev_ops = {
	.ndo_init		= macvlan_init,
	.ndo_open		= macvlan_open,
	.ndo_stop		= macvlan_stop,
	.ndo_change_mtu		= macvlan_change_mtu,
	.ndo_change_rx_flags	= macvlan_change_rx_flags,
	.ndo_set_mac_address	= macvlan_set_mac_address,
	.ndo_set_multicast_list	= macvlan_set_multicast_list,
	.ndo_validate_addr	= eth_validate_addr,
};

static void macvlan_setup(struct net_device *dev)
{
	ether_setup(dev);

	dev->netdev_ops		= &macvlan_netdev_ops;
	dev->hard_start_xmit	= macvlan_hard_start_xmit;
	dev->destructor		= free_netdev;
	dev->header_ops		= &macvlan_hard_header_ops,
	dev->ethtool_ops	= &macvlan_ethtool_ops;
	dev->tx_queue_len	= 0;
}

static int macvlan_port_create(struct net_device *dev)
{
	struct macvlan_port *port;
	unsigned int i;

	if (dev->type != ARPHRD_ETHER || dev->flags & IFF_LOOPBACK)
		return -EINVAL;

	port = kzalloc(sizeof(*port), GFP_KERNEL);
	if (port == NULL)
		return -ENOMEM;

	port->dev = dev;
	INIT_LIST_HEAD(&port->vlans);
	for (i = 0; i < MACVLAN_HASH_SIZE; i++)
		INIT_HLIST_HEAD(&port->vlan_hash[i]);
	rcu_assign_pointer(dev->macvlan_port, port);
	return 0;
}

static void macvlan_port_destroy(struct net_device *dev)
{
	struct macvlan_port *port = dev->macvlan_port;

	rcu_assign_pointer(dev->macvlan_port, NULL);
	synchronize_rcu();
	kfree(port);
}

static void macvlan_transfer_operstate(struct net_device *dev)
{
	struct macvlan_dev *vlan = netdev_priv(dev);
	const struct net_device *lowerdev = vlan->lowerdev;

	if (lowerdev->operstate == IF_OPER_DORMANT)
		netif_dormant_on(dev);
	else
		netif_dormant_off(dev);

	if (netif_carrier_ok(lowerdev)) {
		if (!netif_carrier_ok(dev))
			netif_carrier_on(dev);
	} else {
		if (netif_carrier_ok(dev))
			netif_carrier_off(dev);
	}
}

static int macvlan_validate(struct nlattr *tb[], struct nlattr *data[])
{
	if (tb[IFLA_ADDRESS]) {
		if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN)
			return -EINVAL;
		if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS])))
			return -EADDRNOTAVAIL;
	}
	return 0;
}

static int macvlan_newlink(struct net_device *dev,
			   struct nlattr *tb[], struct nlattr *data[])
{
	struct macvlan_dev *vlan = netdev_priv(dev);
	struct macvlan_port *port;
	struct net_device *lowerdev;
	int err;

	if (!tb[IFLA_LINK])
		return -EINVAL;

	lowerdev = __dev_get_by_index(dev_net(dev), nla_get_u32(tb[IFLA_LINK]));
	if (lowerdev == NULL)
		return -ENODEV;

	/* Don't allow macvlans on top of other macvlans - its not really
	 * wrong, but lockdep can't handle it and its not useful for anything
	 * you couldn't do directly on top of the real device.
	 */
	if (lowerdev->rtnl_link_ops == dev->rtnl_link_ops)
		return -ENODEV;

	if (!tb[IFLA_MTU])
		dev->mtu = lowerdev->mtu;
	else if (dev->mtu > lowerdev->mtu)
		return -EINVAL;

	if (!tb[IFLA_ADDRESS])
		random_ether_addr(dev->dev_addr);

	if (lowerdev->macvlan_port == NULL) {
		err = macvlan_port_create(lowerdev);
		if (err < 0)
			return err;
	}
	port = lowerdev->macvlan_port;

	vlan->lowerdev = lowerdev;
	vlan->dev      = dev;
	vlan->port     = port;

	err = register_netdevice(dev);
	if (err < 0)
		return err;

	list_add_tail(&vlan->list, &port->vlans);
	macvlan_transfer_operstate(dev);
	return 0;
}

static void macvlan_dellink(struct net_device *dev)
{
	struct macvlan_dev *vlan = netdev_priv(dev);
	struct macvlan_port *port = vlan->port;

	list_del(&vlan->list);
	unregister_netdevice(dev);

	if (list_empty(&port->vlans))
		macvlan_port_destroy(port->dev);
}

static struct rtnl_link_ops macvlan_link_ops __read_mostly = {
	.kind		= "macvlan",
	.priv_size	= sizeof(struct macvlan_dev),
	.setup		= macvlan_setup,
	.validate	= macvlan_validate,
	.newlink	= macvlan_newlink,
	.dellink	= macvlan_dellink,
};

static int macvlan_device_event(struct notifier_block *unused,
				unsigned long event, void *ptr)
{
	struct net_device *dev = ptr;
	struct macvlan_dev *vlan, *next;
	struct macvlan_port *port;

	port = dev->macvlan_port;
	if (port == NULL)
		return NOTIFY_DONE;

	switch (event) {
	case NETDEV_CHANGE:
		list_for_each_entry(vlan, &port->vlans, list)
			macvlan_transfer_operstate(vlan->dev);
		break;
	case NETDEV_FEAT_CHANGE:
		list_for_each_entry(vlan, &port->vlans, list) {
			vlan->dev->features = dev->features & MACVLAN_FEATURES;
			netdev_features_change(vlan->dev);
		}
		break;
	case NETDEV_UNREGISTER:
		list_for_each_entry_safe(vlan, next, &port->vlans, list)
			macvlan_dellink(vlan->dev);
		break;
	}
	return NOTIFY_DONE;
}

static struct notifier_block macvlan_notifier_block __read_mostly = {
	.notifier_call	= macvlan_device_event,
};

static int __init macvlan_init_module(void)
{
	int err;

	register_netdevice_notifier(&macvlan_notifier_block);
	macvlan_handle_frame_hook = macvlan_handle_frame;

	err = rtnl_link_register(&macvlan_link_ops);
	if (err < 0)
		goto err1;
	return 0;
err1:
	macvlan_handle_frame_hook = NULL;
	unregister_netdevice_notifier(&macvlan_notifier_block);
	return err;
}

static void __exit macvlan_cleanup_module(void)
{
	rtnl_link_unregister(&macvlan_link_ops);
	macvlan_handle_frame_hook = NULL;
	unregister_netdevice_notifier(&macvlan_notifier_block);
}

module_init(macvlan_init_module);
module_exit(macvlan_cleanup_module);

MODULE_LICENSE("GPL");
MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>");
MODULE_DESCRIPTION("Driver for MAC address based VLANs");
MODULE_ALIAS_RTNL_LINK("macvlan");
Commit	Line	Data
b863ceb7 PM	1	/*
	2	* Copyright (c) 2007 Patrick McHardy <kaber@trash.net>
	3	*
	4	* This program is free software; you can redistribute it and/or
	5	* modify it under the terms of the GNU General Public License as
	6	* published by the Free Software Foundation; either version 2 of
	7	* the License, or (at your option) any later version.
	8	*
	9	* The code this is based on carried the following copyright notice:
	10	* ---
	11	* (C) Copyright 2001-2006
	12	* Alex Zeffertt, Cambridge Broadband Ltd, ajz@cambridgebroadband.com
	13	* Re-worked by Ben Greear <greearb@candelatech.com>
	14	* ---
	15	*/
	16	#include <linux/kernel.h>
	17	#include <linux/types.h>
	18	#include <linux/module.h>
	19	#include <linux/init.h>
	20	#include <linux/errno.h>
	21	#include <linux/slab.h>
	22	#include <linux/string.h>
82524746	23	#include <linux/rculist.h>
b863ceb7 PM	24	#include <linux/notifier.h>
	25	#include <linux/netdevice.h>
	26	#include <linux/etherdevice.h>
	27	#include <linux/ethtool.h>
	28	#include <linux/if_arp.h>
	29	#include <linux/if_link.h>
	30	#include <linux/if_macvlan.h>
	31	#include <net/rtnetlink.h>
	32
	33	#define MACVLAN_HASH_SIZE (1 << BITS_PER_BYTE)
	34
	35	struct macvlan_port {
	36	struct net_device *dev;
	37	struct hlist_head vlan_hash[MACVLAN_HASH_SIZE];
	38	struct list_head vlans;
	39	};
	40
	41	struct macvlan_dev {
	42	struct net_device *dev;
	43	struct list_head list;
	44	struct hlist_node hlist;
	45	struct macvlan_port *port;
	46	struct net_device *lowerdev;
	47	};
	48
	49
	50	static struct macvlan_dev macvlan_hash_lookup(const struct macvlan_port port,
	51	const unsigned char *addr)
	52	{
	53	struct macvlan_dev *vlan;
	54	struct hlist_node *n;
	55
	56	hlist_for_each_entry_rcu(vlan, n, &port->vlan_hash[addr[5]], hlist) {
	57	if (!compare_ether_addr(vlan->dev->dev_addr, addr))
	58	return vlan;
	59	}
	60	return NULL;
	61	}
	62
	63	static void macvlan_broadcast(struct sk_buff *skb,
	64	const struct macvlan_port *port)
	65	{
	66	const struct ethhdr *eth = eth_hdr(skb);
	67	const struct macvlan_dev *vlan;
	68	struct hlist_node *n;
	69	struct net_device *dev;
	70	struct sk_buff *nskb;
	71	unsigned int i;
	72
	73	for (i = 0; i < MACVLAN_HASH_SIZE; i++) {
	74	hlist_for_each_entry_rcu(vlan, n, &port->vlan_hash[i], hlist) {
	75	dev = vlan->dev;
b863ceb7 PM	76
	77	nskb = skb_clone(skb, GFP_ATOMIC);
	78	if (nskb == NULL) {
	79	dev->stats.rx_errors++;
	80	dev->stats.rx_dropped++;
	81	continue;
	82	}
	83
	84	dev->stats.rx_bytes += skb->len + ETH_HLEN;
	85	dev->stats.rx_packets++;
	86	dev->stats.multicast++;
b863ceb7 PM	87
	88	nskb->dev = dev;
	89	if (!compare_ether_addr(eth->h_dest, dev->broadcast))
	90	nskb->pkt_type = PACKET_BROADCAST;
	91	else
	92	nskb->pkt_type = PACKET_MULTICAST;
	93
	94	netif_rx(nskb);
	95	}
	96	}
	97	}
	98
	99	/* called under rcu_read_lock() from netif_receive_skb */
	100	static struct sk_buff macvlan_handle_frame(struct sk_buff skb)
	101	{
	102	const struct ethhdr *eth = eth_hdr(skb);
	103	const struct macvlan_port *port;
	104	const struct macvlan_dev *vlan;
	105	struct net_device *dev;
	106
	107	port = rcu_dereference(skb->dev->macvlan_port);
	108	if (port == NULL)
	109	return skb;
	110
	111	if (is_multicast_ether_addr(eth->h_dest)) {
	112	macvlan_broadcast(skb, port);
	113	return skb;
	114	}
	115
	116	vlan = macvlan_hash_lookup(port, eth->h_dest);
	117	if (vlan == NULL)
	118	return skb;
	119
	120	dev = vlan->dev;
	121	if (unlikely(!(dev->flags & IFF_UP))) {
	122	kfree_skb(skb);
	123	return NULL;
	124	}
	125
	126	skb = skb_share_check(skb, GFP_ATOMIC);
	127	if (skb == NULL) {
	128	dev->stats.rx_errors++;
	129	dev->stats.rx_dropped++;
	130	return NULL;
	131	}
	132
	133	dev->stats.rx_bytes += skb->len + ETH_HLEN;
	134	dev->stats.rx_packets++;
b863ceb7 PM	135
	136	skb->dev = dev;
	137	skb->pkt_type = PACKET_HOST;
	138
	139	netif_rx(skb);
	140	return NULL;
	141	}
	142
	143	static int macvlan_hard_start_xmit(struct sk_buff skb, struct net_device dev)
	144	{
	145	const struct macvlan_dev *vlan = netdev_priv(dev);
	146	unsigned int len = skb->len;
	147	int ret;
	148
	149	skb->dev = vlan->lowerdev;
	150	ret = dev_queue_xmit(skb);
	151
	152	if (likely(ret == NET_XMIT_SUCCESS)) {
	153	dev->stats.tx_packets++;
	154	dev->stats.tx_bytes += len;
	155	} else {
	156	dev->stats.tx_errors++;
	157	dev->stats.tx_aborted_errors++;
	158	}
	159	return NETDEV_TX_OK;
	160	}
	161
	162	static int macvlan_hard_header(struct sk_buff skb, struct net_device dev,
3b04ddde SH	163	unsigned short type, const void *daddr,
3b04ddde SH	164	const void *saddr, unsigned len)
b863ceb7 PM	165	{
	166	const struct macvlan_dev *vlan = netdev_priv(dev);
	167	struct net_device *lowerdev = vlan->lowerdev;
	168
0c4e8581 SH	169	return dev_hard_header(skb, lowerdev, type, daddr,
0c4e8581 SH	170	saddr ? : dev->dev_addr, len);
b863ceb7 PM	171	}
b863ceb7 PM	172
3b04ddde SH	173	static const struct header_ops macvlan_hard_header_ops = {
	174	.create = macvlan_hard_header,
	175	.rebuild = eth_rebuild_header,
	176	.parse = eth_header_parse,
3b04ddde SH	177	.cache = eth_header_cache,
	178	.cache_update = eth_header_cache_update,
	179	};
	180
b863ceb7 PM	181	static int macvlan_open(struct net_device *dev)
	182	{
	183	struct macvlan_dev *vlan = netdev_priv(dev);
	184	struct macvlan_port *port = vlan->port;
	185	struct net_device *lowerdev = vlan->lowerdev;
	186	int err;
	187
	188	err = dev_unicast_add(lowerdev, dev->dev_addr, ETH_ALEN);
	189	if (err < 0)
b89fb7da WC	190	goto out;
	191	if (dev->flags & IFF_ALLMULTI) {
	192	err = dev_set_allmulti(lowerdev, 1);
	193	if (err < 0)
	194	goto del_unicast;
	195	}
b863ceb7 PM	196
	197	hlist_add_head_rcu(&vlan->hlist, &port->vlan_hash[dev->dev_addr[5]]);
	198	return 0;
b89fb7da WC	199
	200	del_unicast:
	201	dev_unicast_delete(lowerdev, dev->dev_addr, ETH_ALEN);
	202	out:
	203	return err;
b863ceb7 PM	204	}
	205
	206	static int macvlan_stop(struct net_device *dev)
	207	{
	208	struct macvlan_dev *vlan = netdev_priv(dev);
	209	struct net_device *lowerdev = vlan->lowerdev;
	210
	211	dev_mc_unsync(lowerdev, dev);
	212	if (dev->flags & IFF_ALLMULTI)
	213	dev_set_allmulti(lowerdev, -1);
	214
	215	dev_unicast_delete(lowerdev, dev->dev_addr, ETH_ALEN);
	216
	217	hlist_del_rcu(&vlan->hlist);
	218	synchronize_rcu();
	219	return 0;
	220	}
	221
ad5d20a6 PM	222	static int macvlan_set_mac_address(struct net_device dev, void p)
	223	{
	224	struct macvlan_dev *vlan = netdev_priv(dev);
	225	struct net_device *lowerdev = vlan->lowerdev;
	226	struct sockaddr *addr = p;
	227	int err;
	228
	229	if (!is_valid_ether_addr(addr->sa_data))
	230	return -EADDRNOTAVAIL;
	231
	232	if (!(dev->flags & IFF_UP))
	233	goto out;
	234
	235	err = dev_unicast_add(lowerdev, addr->sa_data, ETH_ALEN);
	236	if (err < 0)
	237	return err;
	238	dev_unicast_delete(lowerdev, dev->dev_addr, ETH_ALEN);
	239
	240	out:
	241	memcpy(dev->dev_addr, addr->sa_data, ETH_ALEN);
	242	return 0;
	243	}
	244
b863ceb7 PM	245	static void macvlan_change_rx_flags(struct net_device *dev, int change)
	246	{
	247	struct macvlan_dev *vlan = netdev_priv(dev);
	248	struct net_device *lowerdev = vlan->lowerdev;
	249
	250	if (change & IFF_ALLMULTI)
	251	dev_set_allmulti(lowerdev, dev->flags & IFF_ALLMULTI ? 1 : -1);
	252	}
	253
	254	static void macvlan_set_multicast_list(struct net_device *dev)
	255	{
	256	struct macvlan_dev *vlan = netdev_priv(dev);
	257
	258	dev_mc_sync(vlan->lowerdev, dev);
	259	}
	260
	261	static int macvlan_change_mtu(struct net_device *dev, int new_mtu)
	262	{
	263	struct macvlan_dev *vlan = netdev_priv(dev);
	264
	265	if (new_mtu < 68 \|\| vlan->lowerdev->mtu < new_mtu)
	266	return -EINVAL;
	267	dev->mtu = new_mtu;
	268	return 0;
	269	}
	270
	271	/*
	272	* macvlan network devices have devices nesting below it and are a special
	273	* "super class" of normal network devices; split their locks off into a
	274	* separate class since they always nest.
	275	*/
	276	static struct lock_class_key macvlan_netdev_xmit_lock_key;
cf508b12	277	static struct lock_class_key macvlan_netdev_addr_lock_key;
b863ceb7 PM	278
	279	#define MACVLAN_FEATURES \
	280	(NETIF_F_SG \| NETIF_F_ALL_CSUM \| NETIF_F_HIGHDMA \| NETIF_F_FRAGLIST \| \
	281	NETIF_F_GSO \| NETIF_F_TSO \| NETIF_F_UFO \| NETIF_F_GSO_ROBUST \| \
	282	NETIF_F_TSO_ECN \| NETIF_F_TSO6)
	283
	284	#define MACVLAN_STATE_MASK \
	285	((1<<__LINK_STATE_NOCARRIER) \| (1<<__LINK_STATE_DORMANT))
	286
e8a0464c DM	287	static void macvlan_set_lockdep_class_one(struct net_device *dev,
	288	struct netdev_queue *txq,
	289	void *_unused)
c773e847 DM	290	{
	291	lockdep_set_class(&txq->_xmit_lock,
	292	&macvlan_netdev_xmit_lock_key);
	293	}
	294
	295	static void macvlan_set_lockdep_class(struct net_device *dev)
	296	{
cf508b12 DM	297	lockdep_set_class(&dev->addr_list_lock,
cf508b12 DM	298	&macvlan_netdev_addr_lock_key);
e8a0464c	299	netdev_for_each_tx_queue(dev, macvlan_set_lockdep_class_one, NULL);
c773e847 DM	300	}
c773e847 DM	301
b863ceb7 PM	302	static int macvlan_init(struct net_device *dev)
	303	{
	304	struct macvlan_dev *vlan = netdev_priv(dev);
	305	const struct net_device *lowerdev = vlan->lowerdev;
	306
	307	dev->state = (dev->state & ~MACVLAN_STATE_MASK) \|
	308	(lowerdev->state & MACVLAN_STATE_MASK);
	309	dev->features = lowerdev->features & MACVLAN_FEATURES;
	310	dev->iflink = lowerdev->ifindex;
	311
c773e847 DM	312	macvlan_set_lockdep_class(dev);
c773e847 DM	313
b863ceb7 PM	314	return 0;
	315	}
	316
	317	static void macvlan_ethtool_get_drvinfo(struct net_device *dev,
	318	struct ethtool_drvinfo *drvinfo)
	319	{
	320	snprintf(drvinfo->driver, 32, "macvlan");
	321	snprintf(drvinfo->version, 32, "0.1");
	322	}
	323
	324	static u32 macvlan_ethtool_get_rx_csum(struct net_device *dev)
	325	{
	326	const struct macvlan_dev *vlan = netdev_priv(dev);
	327	struct net_device *lowerdev = vlan->lowerdev;
	328
	329	if (lowerdev->ethtool_ops->get_rx_csum == NULL)
	330	return 0;
	331	return lowerdev->ethtool_ops->get_rx_csum(lowerdev);
	332	}
	333
9edb8bb6 SH	334	static int macvlan_ethtool_get_settings(struct net_device *dev,
	335	struct ethtool_cmd *cmd)
	336	{
	337	const struct macvlan_dev *vlan = netdev_priv(dev);
	338	struct net_device *lowerdev = vlan->lowerdev;
	339
	340	if (!lowerdev->ethtool_ops->get_settings)
	341	return -EOPNOTSUPP;
	342
	343	return lowerdev->ethtool_ops->get_settings(lowerdev, cmd);
	344	}
	345
	346	static u32 macvlan_ethtool_get_flags(struct net_device *dev)
	347	{
	348	const struct macvlan_dev *vlan = netdev_priv(dev);
	349	struct net_device *lowerdev = vlan->lowerdev;
	350
	351	if (!lowerdev->ethtool_ops->get_flags)
	352	return 0;
	353	return lowerdev->ethtool_ops->get_flags(lowerdev);
	354	}
	355
b863ceb7 PM	356	static const struct ethtool_ops macvlan_ethtool_ops = {
b863ceb7 PM	357	.get_link = ethtool_op_get_link,
9edb8bb6	358	.get_settings = macvlan_ethtool_get_settings,
b863ceb7	359	.get_rx_csum = macvlan_ethtool_get_rx_csum,
b863ceb7	360	.get_drvinfo = macvlan_ethtool_get_drvinfo,
9edb8bb6	361	.get_flags = macvlan_ethtool_get_flags,
b863ceb7 PM	362	};
b863ceb7 PM	363
54a30c97 SH	364	static const struct net_device_ops macvlan_netdev_ops = {
	365	.ndo_init = macvlan_init,
	366	.ndo_open = macvlan_open,
	367	.ndo_stop = macvlan_stop,
	368	.ndo_change_mtu = macvlan_change_mtu,
	369	.ndo_change_rx_flags = macvlan_change_rx_flags,
	370	.ndo_set_mac_address = macvlan_set_mac_address,
	371	.ndo_set_multicast_list = macvlan_set_multicast_list,
	372	.ndo_validate_addr = eth_validate_addr,
	373	};
	374
b863ceb7 PM	375	static void macvlan_setup(struct net_device *dev)
	376	{
	377	ether_setup(dev);
	378
54a30c97	379	dev->netdev_ops = &macvlan_netdev_ops;
b863ceb7 PM	380	dev->hard_start_xmit = macvlan_hard_start_xmit;
b863ceb7 PM	381	dev->destructor = free_netdev;
3b04ddde	382	dev->header_ops = &macvlan_hard_header_ops,
b863ceb7 PM	383	dev->ethtool_ops = &macvlan_ethtool_ops;
	384	dev->tx_queue_len = 0;
	385	}
	386
	387	static int macvlan_port_create(struct net_device *dev)
	388	{
	389	struct macvlan_port *port;
	390	unsigned int i;
	391
	392	if (dev->type != ARPHRD_ETHER \|\| dev->flags & IFF_LOOPBACK)
	393	return -EINVAL;
	394
	395	port = kzalloc(sizeof(*port), GFP_KERNEL);
	396	if (port == NULL)
	397	return -ENOMEM;
	398
	399	port->dev = dev;
	400	INIT_LIST_HEAD(&port->vlans);
	401	for (i = 0; i < MACVLAN_HASH_SIZE; i++)
	402	INIT_HLIST_HEAD(&port->vlan_hash[i]);
	403	rcu_assign_pointer(dev->macvlan_port, port);
	404	return 0;
	405	}
	406
	407	static void macvlan_port_destroy(struct net_device *dev)
	408	{
	409	struct macvlan_port *port = dev->macvlan_port;
	410
	411	rcu_assign_pointer(dev->macvlan_port, NULL);
	412	synchronize_rcu();
	413	kfree(port);
	414	}
	415
	416	static void macvlan_transfer_operstate(struct net_device *dev)
	417	{
	418	struct macvlan_dev *vlan = netdev_priv(dev);
	419	const struct net_device *lowerdev = vlan->lowerdev;
	420
	421	if (lowerdev->operstate == IF_OPER_DORMANT)
	422	netif_dormant_on(dev);
	423	else
	424	netif_dormant_off(dev);
	425
	426	if (netif_carrier_ok(lowerdev)) {
	427	if (!netif_carrier_ok(dev))
	428	netif_carrier_on(dev);
	429	} else {
f12ca5f9	430	if (netif_carrier_ok(dev))
b863ceb7 PM	431	netif_carrier_off(dev);
	432	}
	433	}
	434
	435	static int macvlan_validate(struct nlattr tb[], struct nlattr data[])
	436	{
	437	if (tb[IFLA_ADDRESS]) {
	438	if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN)
	439	return -EINVAL;
	440	if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS])))
	441	return -EADDRNOTAVAIL;
	442	}
	443	return 0;
	444	}
	445
	446	static int macvlan_newlink(struct net_device *dev,
	447	struct nlattr tb[], struct nlattr data[])
	448	{
	449	struct macvlan_dev *vlan = netdev_priv(dev);
	450	struct macvlan_port *port;
	451	struct net_device *lowerdev;
	452	int err;
	453
	454	if (!tb[IFLA_LINK])
	455	return -EINVAL;
	456
c346dca1	457	lowerdev = __dev_get_by_index(dev_net(dev), nla_get_u32(tb[IFLA_LINK]));
b863ceb7 PM	458	if (lowerdev == NULL)
	459	return -ENODEV;
	460
a6ca5f1d PM	461	/* Don't allow macvlans on top of other macvlans - its not really
	462	* wrong, but lockdep can't handle it and its not useful for anything
	463	* you couldn't do directly on top of the real device.
	464	*/
	465	if (lowerdev->rtnl_link_ops == dev->rtnl_link_ops)
	466	return -ENODEV;
	467
b863ceb7 PM	468	if (!tb[IFLA_MTU])
	469	dev->mtu = lowerdev->mtu;
	470	else if (dev->mtu > lowerdev->mtu)
	471	return -EINVAL;
	472
	473	if (!tb[IFLA_ADDRESS])
	474	random_ether_addr(dev->dev_addr);
	475
	476	if (lowerdev->macvlan_port == NULL) {
	477	err = macvlan_port_create(lowerdev);
	478	if (err < 0)
	479	return err;
	480	}
	481	port = lowerdev->macvlan_port;
	482
	483	vlan->lowerdev = lowerdev;
	484	vlan->dev = dev;
	485	vlan->port = port;
	486
	487	err = register_netdevice(dev);
	488	if (err < 0)
	489	return err;
	490
	491	list_add_tail(&vlan->list, &port->vlans);
	492	macvlan_transfer_operstate(dev);
	493	return 0;
	494	}
	495
	496	static void macvlan_dellink(struct net_device *dev)
	497	{
	498	struct macvlan_dev *vlan = netdev_priv(dev);
	499	struct macvlan_port *port = vlan->port;
	500
	501	list_del(&vlan->list);
	502	unregister_netdevice(dev);
	503
	504	if (list_empty(&port->vlans))
73120964	505	macvlan_port_destroy(port->dev);
b863ceb7 PM	506	}
	507
	508	static struct rtnl_link_ops macvlan_link_ops __read_mostly = {
	509	.kind = "macvlan",
	510	.priv_size = sizeof(struct macvlan_dev),
	511	.setup = macvlan_setup,
	512	.validate = macvlan_validate,
	513	.newlink = macvlan_newlink,
	514	.dellink = macvlan_dellink,
	515	};
	516
	517	static int macvlan_device_event(struct notifier_block *unused,
	518	unsigned long event, void *ptr)
	519	{
	520	struct net_device *dev = ptr;
	521	struct macvlan_dev vlan, next;
	522	struct macvlan_port *port;
	523
	524	port = dev->macvlan_port;
	525	if (port == NULL)
	526	return NOTIFY_DONE;
	527
	528	switch (event) {
	529	case NETDEV_CHANGE:
	530	list_for_each_entry(vlan, &port->vlans, list)
	531	macvlan_transfer_operstate(vlan->dev);
	532	break;
	533	case NETDEV_FEAT_CHANGE:
	534	list_for_each_entry(vlan, &port->vlans, list) {
	535	vlan->dev->features = dev->features & MACVLAN_FEATURES;
	536	netdev_features_change(vlan->dev);
	537	}
	538	break;
	539	case NETDEV_UNREGISTER:
	540	list_for_each_entry_safe(vlan, next, &port->vlans, list)
	541	macvlan_dellink(vlan->dev);
	542	break;
	543	}
	544	return NOTIFY_DONE;
	545	}
	546
	547	static struct notifier_block macvlan_notifier_block __read_mostly = {
	548	.notifier_call = macvlan_device_event,
	549	};
	550
	551	static int __init macvlan_init_module(void)
	552	{
	553	int err;
	554
	555	register_netdevice_notifier(&macvlan_notifier_block);
	556	macvlan_handle_frame_hook = macvlan_handle_frame;
	557
	558	err = rtnl_link_register(&macvlan_link_ops);
	559	if (err < 0)
	560	goto err1;
	561	return 0;
	562	err1:
52913246	563	macvlan_handle_frame_hook = NULL;
b863ceb7 PM	564	unregister_netdevice_notifier(&macvlan_notifier_block);
	565	return err;
	566	}
	567
	568	static void __exit macvlan_cleanup_module(void)
	569	{
	570	rtnl_link_unregister(&macvlan_link_ops);
	571	macvlan_handle_frame_hook = NULL;
	572	unregister_netdevice_notifier(&macvlan_notifier_block);
	573	}
	574
	575	module_init(macvlan_init_module);
	576	module_exit(macvlan_cleanup_module);
	577
	578	MODULE_LICENSE("GPL");
	579	MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>");
	580	MODULE_DESCRIPTION("Driver for MAC address based VLANs");
	581	MODULE_ALIAS_RTNL_LINK("macvlan");