[linux.git] / drivers / net / veth.c

/*
 *  drivers/net/veth.c
 *
 *  Copyright (C) 2007 OpenVZ http://openvz.org, SWsoft Inc
 *
 * Author: Pavel Emelianov <[email protected]>
 * Ethtool interface from: Eric W. Biederman <[email protected]>
 *
 */

#include <linux/netdevice.h>
#include <linux/ethtool.h>
#include <linux/etherdevice.h>

#include <net/dst.h>
#include <net/xfrm.h>
#include <linux/veth.h>

#define DRV_NAME	"veth"
#define DRV_VERSION	"1.0"

#define MIN_MTU 68		/* Min L3 MTU */
#define MAX_MTU 65535		/* Max L3 MTU (arbitrary) */
#define MTU_PAD (ETH_HLEN + 4)  /* Max difference between L2 and L3 size MTU */

struct veth_net_stats {
	unsigned long	rx_packets;
	unsigned long	tx_packets;
	unsigned long	rx_bytes;
	unsigned long	tx_bytes;
	unsigned long	tx_dropped;
	unsigned long	rx_dropped;
};

struct veth_priv {
	struct net_device *peer;
	struct veth_net_stats *stats;
	unsigned ip_summed;
};

/*
 * ethtool interface
 */

static struct {
	const char string[ETH_GSTRING_LEN];
} ethtool_stats_keys[] = {
	{ "peer_ifindex" },
};

static int veth_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
	cmd->supported		= 0;
	cmd->advertising	= 0;
	cmd->speed		= SPEED_10000;
	cmd->duplex		= DUPLEX_FULL;
	cmd->port		= PORT_TP;
	cmd->phy_address	= 0;
	cmd->transceiver	= XCVR_INTERNAL;
	cmd->autoneg		= AUTONEG_DISABLE;
	cmd->maxtxpkt		= 0;
	cmd->maxrxpkt		= 0;
	return 0;
}

static void veth_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
{
	strcpy(info->driver, DRV_NAME);
	strcpy(info->version, DRV_VERSION);
	strcpy(info->fw_version, "N/A");
}

static void veth_get_strings(struct net_device *dev, u32 stringset, u8 *buf)
{
	switch(stringset) {
	case ETH_SS_STATS:
		memcpy(buf, &ethtool_stats_keys, sizeof(ethtool_stats_keys));
		break;
	}
}

static int veth_get_sset_count(struct net_device *dev, int sset)
{
	switch (sset) {
	case ETH_SS_STATS:
		return ARRAY_SIZE(ethtool_stats_keys);
	default:
		return -EOPNOTSUPP;
	}
}

static void veth_get_ethtool_stats(struct net_device *dev,
		struct ethtool_stats *stats, u64 *data)
{
	struct veth_priv *priv;

	priv = netdev_priv(dev);
	data[0] = priv->peer->ifindex;
}

static u32 veth_get_rx_csum(struct net_device *dev)
{
	struct veth_priv *priv;

	priv = netdev_priv(dev);
	return priv->ip_summed == CHECKSUM_UNNECESSARY;
}

static int veth_set_rx_csum(struct net_device *dev, u32 data)
{
	struct veth_priv *priv;

	priv = netdev_priv(dev);
	priv->ip_summed = data ? CHECKSUM_UNNECESSARY : CHECKSUM_NONE;
	return 0;
}

static u32 veth_get_tx_csum(struct net_device *dev)
{
	return (dev->features & NETIF_F_NO_CSUM) != 0;
}

static int veth_set_tx_csum(struct net_device *dev, u32 data)
{
	if (data)
		dev->features |= NETIF_F_NO_CSUM;
	else
		dev->features &= ~NETIF_F_NO_CSUM;
	return 0;
}

static const struct ethtool_ops veth_ethtool_ops = {
	.get_settings		= veth_get_settings,
	.get_drvinfo		= veth_get_drvinfo,
	.get_link		= ethtool_op_get_link,
	.get_rx_csum		= veth_get_rx_csum,
	.set_rx_csum		= veth_set_rx_csum,
	.get_tx_csum		= veth_get_tx_csum,
	.set_tx_csum		= veth_set_tx_csum,
	.get_sg			= ethtool_op_get_sg,
	.set_sg			= ethtool_op_set_sg,
	.get_strings		= veth_get_strings,
	.get_sset_count		= veth_get_sset_count,
	.get_ethtool_stats	= veth_get_ethtool_stats,
};

/*
 * xmit
 */

static netdev_tx_t veth_xmit(struct sk_buff *skb, struct net_device *dev)
{
	struct net_device *rcv = NULL;
	struct veth_priv *priv, *rcv_priv;
	struct veth_net_stats *stats, *rcv_stats;
	int length, cpu;

	priv = netdev_priv(dev);
	rcv = priv->peer;
	rcv_priv = netdev_priv(rcv);

	cpu = smp_processor_id();
	stats = per_cpu_ptr(priv->stats, cpu);
	rcv_stats = per_cpu_ptr(rcv_priv->stats, cpu);

	if (!(rcv->flags & IFF_UP))
		goto tx_drop;

	if (dev->features & NETIF_F_NO_CSUM)
		skb->ip_summed = rcv_priv->ip_summed;

	length = skb->len + ETH_HLEN;
	if (dev_forward_skb(rcv, skb) != NET_RX_SUCCESS)
		goto rx_drop;

	stats->tx_bytes += length;
	stats->tx_packets++;

	rcv_stats->rx_bytes += length;
	rcv_stats->rx_packets++;

	return NETDEV_TX_OK;

tx_drop:
	kfree_skb(skb);
	stats->tx_dropped++;
	return NETDEV_TX_OK;

rx_drop:
	kfree_skb(skb);
	rcv_stats->rx_dropped++;
	return NETDEV_TX_OK;
}

/*
 * general routines
 */

static struct net_device_stats *veth_get_stats(struct net_device *dev)
{
	struct veth_priv *priv;
	struct net_device_stats *dev_stats;
	int cpu;
	struct veth_net_stats *stats;

	priv = netdev_priv(dev);
	dev_stats = &dev->stats;

	dev_stats->rx_packets = 0;
	dev_stats->tx_packets = 0;
	dev_stats->rx_bytes = 0;
	dev_stats->tx_bytes = 0;
	dev_stats->tx_dropped = 0;
	dev_stats->rx_dropped = 0;

	for_each_online_cpu(cpu) {
		stats = per_cpu_ptr(priv->stats, cpu);

		dev_stats->rx_packets += stats->rx_packets;
		dev_stats->tx_packets += stats->tx_packets;
		dev_stats->rx_bytes += stats->rx_bytes;
		dev_stats->tx_bytes += stats->tx_bytes;
		dev_stats->tx_dropped += stats->tx_dropped;
		dev_stats->rx_dropped += stats->rx_dropped;
	}

	return dev_stats;
}

static int veth_open(struct net_device *dev)
{
	struct veth_priv *priv;

	priv = netdev_priv(dev);
	if (priv->peer == NULL)
		return -ENOTCONN;

	if (priv->peer->flags & IFF_UP) {
		netif_carrier_on(dev);
		netif_carrier_on(priv->peer);
	}
	return 0;
}

static int veth_close(struct net_device *dev)
{
	struct veth_priv *priv = netdev_priv(dev);

	netif_carrier_off(dev);
	netif_carrier_off(priv->peer);

	return 0;
}

static int is_valid_veth_mtu(int new_mtu)
{
	return (new_mtu >= MIN_MTU && new_mtu <= MAX_MTU);
}

static int veth_change_mtu(struct net_device *dev, int new_mtu)
{
	if (!is_valid_veth_mtu(new_mtu))
		return -EINVAL;
	dev->mtu = new_mtu;
	return 0;
}

static int veth_dev_init(struct net_device *dev)
{
	struct veth_net_stats *stats;
	struct veth_priv *priv;

	stats = alloc_percpu(struct veth_net_stats);
	if (stats == NULL)
		return -ENOMEM;

	priv = netdev_priv(dev);
	priv->stats = stats;
	return 0;
}

static void veth_dev_free(struct net_device *dev)
{
	struct veth_priv *priv;

	priv = netdev_priv(dev);
	free_percpu(priv->stats);
	free_netdev(dev);
}

static const struct net_device_ops veth_netdev_ops = {
	.ndo_init            = veth_dev_init,
	.ndo_open            = veth_open,
	.ndo_stop            = veth_close,
	.ndo_start_xmit      = veth_xmit,
	.ndo_change_mtu      = veth_change_mtu,
	.ndo_get_stats       = veth_get_stats,
	.ndo_set_mac_address = eth_mac_addr,
};

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

	dev->netdev_ops = &veth_netdev_ops;
	dev->ethtool_ops = &veth_ethtool_ops;
	dev->features |= NETIF_F_LLTX;
	dev->destructor = veth_dev_free;
}

/*
 * netlink interface
 */

static int veth_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;
	}
	if (tb[IFLA_MTU]) {
		if (!is_valid_veth_mtu(nla_get_u32(tb[IFLA_MTU])))
			return -EINVAL;
	}
	return 0;
}

static struct rtnl_link_ops veth_link_ops;

static int veth_newlink(struct net *src_net, struct net_device *dev,
			 struct nlattr *tb[], struct nlattr *data[])
{
	int err;
	struct net_device *peer;
	struct veth_priv *priv;
	char ifname[IFNAMSIZ];
	struct nlattr *peer_tb[IFLA_MAX + 1], **tbp;
	struct net *net;

	/*
	 * create and register peer first
	 *
	 * struct ifinfomsg is at the head of VETH_INFO_PEER, but we
	 * skip it since no info from it is useful yet
	 */

	if (data != NULL && data[VETH_INFO_PEER] != NULL) {
		struct nlattr *nla_peer;

		nla_peer = data[VETH_INFO_PEER];
		err = nla_parse(peer_tb, IFLA_MAX,
				nla_data(nla_peer) + sizeof(struct ifinfomsg),
				nla_len(nla_peer) - sizeof(struct ifinfomsg),
				ifla_policy);
		if (err < 0)
			return err;

		err = veth_validate(peer_tb, NULL);
		if (err < 0)
			return err;

		tbp = peer_tb;
	} else
		tbp = tb;

	if (tbp[IFLA_IFNAME])
		nla_strlcpy(ifname, tbp[IFLA_IFNAME], IFNAMSIZ);
	else
		snprintf(ifname, IFNAMSIZ, DRV_NAME "%%d");

	net = rtnl_link_get_net(src_net, tbp);
	if (IS_ERR(net))
		return PTR_ERR(net);

	peer = rtnl_create_link(src_net, net, ifname, &veth_link_ops, tbp);
	if (IS_ERR(peer)) {
		put_net(net);
		return PTR_ERR(peer);
	}

	if (tbp[IFLA_ADDRESS] == NULL)
		random_ether_addr(peer->dev_addr);

	err = register_netdevice(peer);
	put_net(net);
	net = NULL;
	if (err < 0)
		goto err_register_peer;

	netif_carrier_off(peer);

	/*
	 * register dev last
	 *
	 * note, that since we've registered new device the dev's name
	 * should be re-allocated
	 */

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

	if (tb[IFLA_IFNAME])
		nla_strlcpy(dev->name, tb[IFLA_IFNAME], IFNAMSIZ);
	else
		snprintf(dev->name, IFNAMSIZ, DRV_NAME "%%d");

	if (strchr(dev->name, '%')) {
		err = dev_alloc_name(dev, dev->name);
		if (err < 0)
			goto err_alloc_name;
	}

	err = register_netdevice(dev);
	if (err < 0)
		goto err_register_dev;

	netif_carrier_off(dev);

	/*
	 * tie the deviced together
	 */

	priv = netdev_priv(dev);
	priv->peer = peer;

	priv = netdev_priv(peer);
	priv->peer = dev;
	return 0;

err_register_dev:
	/* nothing to do */
err_alloc_name:
	unregister_netdevice(peer);
	return err;

err_register_peer:
	free_netdev(peer);
	return err;
}

static void veth_dellink(struct net_device *dev, struct list_head *head)
{
	struct veth_priv *priv;
	struct net_device *peer;

	priv = netdev_priv(dev);
	peer = priv->peer;

	unregister_netdevice_queue(dev, head);
	unregister_netdevice_queue(peer, head);
}

static const struct nla_policy veth_policy[VETH_INFO_MAX + 1];

static struct rtnl_link_ops veth_link_ops = {
	.kind		= DRV_NAME,
	.priv_size	= sizeof(struct veth_priv),
	.setup		= veth_setup,
	.validate	= veth_validate,
	.newlink	= veth_newlink,
	.dellink	= veth_dellink,
	.policy		= veth_policy,
	.maxtype	= VETH_INFO_MAX,
};

/*
 * init/fini
 */

static __init int veth_init(void)
{
	return rtnl_link_register(&veth_link_ops);
}

static __exit void veth_exit(void)
{
	rtnl_link_unregister(&veth_link_ops);
}

module_init(veth_init);
module_exit(veth_exit);

MODULE_DESCRIPTION("Virtual Ethernet Tunnel");
MODULE_LICENSE("GPL v2");
MODULE_ALIAS_RTNL_LINK(DRV_NAME);
Commit	Line	Data
e314dbdc PE	1	/*
	2	* drivers/net/veth.c
	3	*
	4	* Copyright (C) 2007 OpenVZ http://openvz.org, SWsoft Inc
	5	*
	6	* Author: Pavel Emelianov <[email protected]>
	7	* Ethtool interface from: Eric W. Biederman <[email protected]>
	8	*
	9	*/
	10
e314dbdc PE	11	#include <linux/netdevice.h>
	12	#include <linux/ethtool.h>
	13	#include <linux/etherdevice.h>
	14
	15	#include <net/dst.h>
	16	#include <net/xfrm.h>
ecef969e	17	#include <linux/veth.h>
e314dbdc PE	18
	19	#define DRV_NAME "veth"
	20	#define DRV_VERSION "1.0"
	21
38d40815 EB	22	#define MIN_MTU 68 /* Min L3 MTU */
	23	#define MAX_MTU 65535 /* Max L3 MTU (arbitrary) */
	24	#define MTU_PAD (ETH_HLEN + 4) /* Max difference between L2 and L3 size MTU */
	25
e314dbdc PE	26	struct veth_net_stats {
	27	unsigned long rx_packets;
	28	unsigned long tx_packets;
	29	unsigned long rx_bytes;
	30	unsigned long tx_bytes;
	31	unsigned long tx_dropped;
38d40815	32	unsigned long rx_dropped;
e314dbdc PE	33	};
	34
	35	struct veth_priv {
	36	struct net_device *peer;
e314dbdc PE	37	struct veth_net_stats *stats;
	38	unsigned ip_summed;
	39	};
	40
e314dbdc PE	41	/*
	42	* ethtool interface
	43	*/
	44
	45	static struct {
	46	const char string[ETH_GSTRING_LEN];
	47	} ethtool_stats_keys[] = {
	48	{ "peer_ifindex" },
	49	};
	50
	51	static int veth_get_settings(struct net_device dev, struct ethtool_cmd cmd)
	52	{
	53	cmd->supported = 0;
	54	cmd->advertising = 0;
	55	cmd->speed = SPEED_10000;
	56	cmd->duplex = DUPLEX_FULL;
	57	cmd->port = PORT_TP;
	58	cmd->phy_address = 0;
	59	cmd->transceiver = XCVR_INTERNAL;
	60	cmd->autoneg = AUTONEG_DISABLE;
	61	cmd->maxtxpkt = 0;
	62	cmd->maxrxpkt = 0;
	63	return 0;
	64	}
	65
	66	static void veth_get_drvinfo(struct net_device dev, struct ethtool_drvinfo info)
	67	{
	68	strcpy(info->driver, DRV_NAME);
	69	strcpy(info->version, DRV_VERSION);
	70	strcpy(info->fw_version, "N/A");
	71	}
	72
	73	static void veth_get_strings(struct net_device dev, u32 stringset, u8 buf)
	74	{
	75	switch(stringset) {
	76	case ETH_SS_STATS:
	77	memcpy(buf, &ethtool_stats_keys, sizeof(ethtool_stats_keys));
	78	break;
	79	}
	80	}
	81
b9f2c044	82	static int veth_get_sset_count(struct net_device *dev, int sset)
e314dbdc	83	{
b9f2c044 JG	84	switch (sset) {
	85	case ETH_SS_STATS:
	86	return ARRAY_SIZE(ethtool_stats_keys);
	87	default:
	88	return -EOPNOTSUPP;
	89	}
e314dbdc PE	90	}
	91
	92	static void veth_get_ethtool_stats(struct net_device *dev,
	93	struct ethtool_stats stats, u64 data)
	94	{
	95	struct veth_priv *priv;
	96
	97	priv = netdev_priv(dev);
	98	data[0] = priv->peer->ifindex;
	99	}
	100
	101	static u32 veth_get_rx_csum(struct net_device *dev)
	102	{
	103	struct veth_priv *priv;
	104
	105	priv = netdev_priv(dev);
	106	return priv->ip_summed == CHECKSUM_UNNECESSARY;
	107	}
	108
	109	static int veth_set_rx_csum(struct net_device *dev, u32 data)
	110	{
	111	struct veth_priv *priv;
	112
	113	priv = netdev_priv(dev);
	114	priv->ip_summed = data ? CHECKSUM_UNNECESSARY : CHECKSUM_NONE;
	115	return 0;
	116	}
	117
	118	static u32 veth_get_tx_csum(struct net_device *dev)
	119	{
	120	return (dev->features & NETIF_F_NO_CSUM) != 0;
	121	}
	122
	123	static int veth_set_tx_csum(struct net_device *dev, u32 data)
	124	{
	125	if (data)
	126	dev->features \|= NETIF_F_NO_CSUM;
	127	else
	128	dev->features &= ~NETIF_F_NO_CSUM;
	129	return 0;
	130	}
	131
0fc0b732	132	static const struct ethtool_ops veth_ethtool_ops = {
e314dbdc PE	133	.get_settings = veth_get_settings,
	134	.get_drvinfo = veth_get_drvinfo,
	135	.get_link = ethtool_op_get_link,
	136	.get_rx_csum = veth_get_rx_csum,
	137	.set_rx_csum = veth_set_rx_csum,
	138	.get_tx_csum = veth_get_tx_csum,
	139	.set_tx_csum = veth_set_tx_csum,
	140	.get_sg = ethtool_op_get_sg,
	141	.set_sg = ethtool_op_set_sg,
	142	.get_strings = veth_get_strings,
b9f2c044	143	.get_sset_count = veth_get_sset_count,
e314dbdc PE	144	.get_ethtool_stats = veth_get_ethtool_stats,
	145	};
	146
	147	/*
	148	* xmit
	149	*/
	150
424efe9c	151	static netdev_tx_t veth_xmit(struct sk_buff skb, struct net_device dev)
e314dbdc PE	152	{
	153	struct net_device *rcv = NULL;
	154	struct veth_priv priv, rcv_priv;
38d40815	155	struct veth_net_stats stats, rcv_stats;
e314dbdc PE	156	int length, cpu;
e314dbdc PE	157
e314dbdc PE	158	priv = netdev_priv(dev);
	159	rcv = priv->peer;
	160	rcv_priv = netdev_priv(rcv);
	161
	162	cpu = smp_processor_id();
	163	stats = per_cpu_ptr(priv->stats, cpu);
38d40815	164	rcv_stats = per_cpu_ptr(rcv_priv->stats, cpu);
e314dbdc PE	165
e314dbdc PE	166	if (!(rcv->flags & IFF_UP))
38d40815 EB	167	goto tx_drop;
38d40815 EB	168
e314dbdc PE	169	if (dev->features & NETIF_F_NO_CSUM)
	170	skb->ip_summed = rcv_priv->ip_summed;
	171
44540960 AB	172	length = skb->len + ETH_HLEN;
	173	if (dev_forward_skb(rcv, skb) != NET_RX_SUCCESS)
	174	goto rx_drop;
e314dbdc PE	175
	176	stats->tx_bytes += length;
	177	stats->tx_packets++;
	178
38d40815 EB	179	rcv_stats->rx_bytes += length;
38d40815 EB	180	rcv_stats->rx_packets++;
e314dbdc	181
6ed10654	182	return NETDEV_TX_OK;
e314dbdc	183
38d40815	184	tx_drop:
e314dbdc PE	185	kfree_skb(skb);
e314dbdc PE	186	stats->tx_dropped++;
6ed10654	187	return NETDEV_TX_OK;
38d40815 EB	188
	189	rx_drop:
	190	kfree_skb(skb);
	191	rcv_stats->rx_dropped++;
6ed10654	192	return NETDEV_TX_OK;
e314dbdc PE	193	}
	194
	195	/*
	196	* general routines
	197	*/
	198
	199	static struct net_device_stats veth_get_stats(struct net_device dev)
	200	{
11687a10 DM	201	struct veth_priv *priv;
	202	struct net_device_stats *dev_stats;
	203	int cpu;
e314dbdc PE	204	struct veth_net_stats *stats;
e314dbdc PE	205
11687a10 DM	206	priv = netdev_priv(dev);
	207	dev_stats = &dev->stats;
	208
e314dbdc PE	209	dev_stats->rx_packets = 0;
	210	dev_stats->tx_packets = 0;
	211	dev_stats->rx_bytes = 0;
	212	dev_stats->tx_bytes = 0;
	213	dev_stats->tx_dropped = 0;
38d40815	214	dev_stats->rx_dropped = 0;
e314dbdc	215
11687a10 DM	216	for_each_online_cpu(cpu) {
11687a10 DM	217	stats = per_cpu_ptr(priv->stats, cpu);
e314dbdc	218
11687a10 DM	219	dev_stats->rx_packets += stats->rx_packets;
	220	dev_stats->tx_packets += stats->tx_packets;
	221	dev_stats->rx_bytes += stats->rx_bytes;
	222	dev_stats->tx_bytes += stats->tx_bytes;
	223	dev_stats->tx_dropped += stats->tx_dropped;
	224	dev_stats->rx_dropped += stats->rx_dropped;
	225	}
e314dbdc PE	226
	227	return dev_stats;
	228	}
	229
	230	static int veth_open(struct net_device *dev)
	231	{
	232	struct veth_priv *priv;
	233
	234	priv = netdev_priv(dev);
	235	if (priv->peer == NULL)
	236	return -ENOTCONN;
	237
	238	if (priv->peer->flags & IFF_UP) {
	239	netif_carrier_on(dev);
	240	netif_carrier_on(priv->peer);
	241	}
	242	return 0;
	243	}
	244
2cf48a10 EB	245	static int veth_close(struct net_device *dev)
	246	{
	247	struct veth_priv *priv = netdev_priv(dev);
	248
	249	netif_carrier_off(dev);
	250	netif_carrier_off(priv->peer);
	251
	252	return 0;
	253	}
	254
38d40815 EB	255	static int is_valid_veth_mtu(int new_mtu)
	256	{
	257	return (new_mtu >= MIN_MTU && new_mtu <= MAX_MTU);
	258	}
	259
	260	static int veth_change_mtu(struct net_device *dev, int new_mtu)
	261	{
	262	if (!is_valid_veth_mtu(new_mtu))
	263	return -EINVAL;
	264	dev->mtu = new_mtu;
	265	return 0;
	266	}
	267
e314dbdc PE	268	static int veth_dev_init(struct net_device *dev)
	269	{
	270	struct veth_net_stats *stats;
	271	struct veth_priv *priv;
	272
	273	stats = alloc_percpu(struct veth_net_stats);
	274	if (stats == NULL)
	275	return -ENOMEM;
	276
	277	priv = netdev_priv(dev);
	278	priv->stats = stats;
	279	return 0;
	280	}
	281
11687a10 DM	282	static void veth_dev_free(struct net_device *dev)
	283	{
	284	struct veth_priv *priv;
	285
	286	priv = netdev_priv(dev);
	287	free_percpu(priv->stats);
	288	free_netdev(dev);
	289	}
	290
4456e7bd	291	static const struct net_device_ops veth_netdev_ops = {
ee923623 DL	292	.ndo_init = veth_dev_init,
ee923623 DL	293	.ndo_open = veth_open,
2cf48a10	294	.ndo_stop = veth_close,
ee923623	295	.ndo_start_xmit = veth_xmit,
38d40815	296	.ndo_change_mtu = veth_change_mtu,
ee923623 DL	297	.ndo_get_stats = veth_get_stats,
ee923623 DL	298	.ndo_set_mac_address = eth_mac_addr,
4456e7bd SH	299	};
4456e7bd SH	300
e314dbdc PE	301	static void veth_setup(struct net_device *dev)
	302	{
	303	ether_setup(dev);
	304
4456e7bd	305	dev->netdev_ops = &veth_netdev_ops;
e314dbdc PE	306	dev->ethtool_ops = &veth_ethtool_ops;
e314dbdc PE	307	dev->features \|= NETIF_F_LLTX;
11687a10	308	dev->destructor = veth_dev_free;
e314dbdc PE	309	}
	310
	311	/*
	312	* netlink interface
	313	*/
	314
	315	static int veth_validate(struct nlattr tb[], struct nlattr data[])
	316	{
	317	if (tb[IFLA_ADDRESS]) {
	318	if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN)
	319	return -EINVAL;
	320	if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS])))
	321	return -EADDRNOTAVAIL;
	322	}
38d40815 EB	323	if (tb[IFLA_MTU]) {
	324	if (!is_valid_veth_mtu(nla_get_u32(tb[IFLA_MTU])))
	325	return -EINVAL;
	326	}
e314dbdc PE	327	return 0;
	328	}
	329
	330	static struct rtnl_link_ops veth_link_ops;
	331
81adee47	332	static int veth_newlink(struct net src_net, struct net_device dev,
e314dbdc PE	333	struct nlattr tb[], struct nlattr data[])
	334	{
	335	int err;
	336	struct net_device *peer;
	337	struct veth_priv *priv;
	338	char ifname[IFNAMSIZ];
	339	struct nlattr peer_tb[IFLA_MAX + 1], *tbp;
81adee47	340	struct net *net;
e314dbdc PE	341
	342	/*
	343	* create and register peer first
	344	*
	345	* struct ifinfomsg is at the head of VETH_INFO_PEER, but we
	346	* skip it since no info from it is useful yet
	347	*/
	348
	349	if (data != NULL && data[VETH_INFO_PEER] != NULL) {
	350	struct nlattr *nla_peer;
	351
	352	nla_peer = data[VETH_INFO_PEER];
	353	err = nla_parse(peer_tb, IFLA_MAX,
	354	nla_data(nla_peer) + sizeof(struct ifinfomsg),
	355	nla_len(nla_peer) - sizeof(struct ifinfomsg),
	356	ifla_policy);
	357	if (err < 0)
	358	return err;
	359
	360	err = veth_validate(peer_tb, NULL);
	361	if (err < 0)
	362	return err;
	363
	364	tbp = peer_tb;
	365	} else
	366	tbp = tb;
	367
	368	if (tbp[IFLA_IFNAME])
	369	nla_strlcpy(ifname, tbp[IFLA_IFNAME], IFNAMSIZ);
	370	else
	371	snprintf(ifname, IFNAMSIZ, DRV_NAME "%%d");
	372
81adee47 EB	373	net = rtnl_link_get_net(src_net, tbp);
	374	if (IS_ERR(net))
	375	return PTR_ERR(net);
	376
	377	peer = rtnl_create_link(src_net, net, ifname, &veth_link_ops, tbp);
	378	if (IS_ERR(peer)) {
	379	put_net(net);
e314dbdc	380	return PTR_ERR(peer);
81adee47	381	}
e314dbdc PE	382
	383	if (tbp[IFLA_ADDRESS] == NULL)
	384	random_ether_addr(peer->dev_addr);
	385
	386	err = register_netdevice(peer);
81adee47 EB	387	put_net(net);
81adee47 EB	388	net = NULL;
e314dbdc PE	389	if (err < 0)
	390	goto err_register_peer;
	391
	392	netif_carrier_off(peer);
	393
	394	/*
	395	* register dev last
	396	*
	397	* note, that since we've registered new device the dev's name
	398	* should be re-allocated
	399	*/
	400
	401	if (tb[IFLA_ADDRESS] == NULL)
	402	random_ether_addr(dev->dev_addr);
	403
	404	if (tb[IFLA_IFNAME])
	405	nla_strlcpy(dev->name, tb[IFLA_IFNAME], IFNAMSIZ);
	406	else
	407	snprintf(dev->name, IFNAMSIZ, DRV_NAME "%%d");
	408
	409	if (strchr(dev->name, '%')) {
	410	err = dev_alloc_name(dev, dev->name);
	411	if (err < 0)
	412	goto err_alloc_name;
	413	}
	414
	415	err = register_netdevice(dev);
	416	if (err < 0)
	417	goto err_register_dev;
	418
	419	netif_carrier_off(dev);
	420
	421	/*
	422	* tie the deviced together
	423	*/
	424
	425	priv = netdev_priv(dev);
e314dbdc	426	priv->peer = peer;
e314dbdc PE	427
e314dbdc PE	428	priv = netdev_priv(peer);
e314dbdc	429	priv->peer = dev;
e314dbdc PE	430	return 0;
	431
	432	err_register_dev:
	433	/* nothing to do */
	434	err_alloc_name:
	435	unregister_netdevice(peer);
	436	return err;
	437
	438	err_register_peer:
	439	free_netdev(peer);
	440	return err;
	441	}
	442
23289a37	443	static void veth_dellink(struct net_device dev, struct list_head head)
e314dbdc PE	444	{
	445	struct veth_priv *priv;
	446	struct net_device *peer;
	447
	448	priv = netdev_priv(dev);
	449	peer = priv->peer;
	450
24540535 ED	451	unregister_netdevice_queue(dev, head);
24540535 ED	452	unregister_netdevice_queue(peer, head);
e314dbdc PE	453	}
	454
	455	static const struct nla_policy veth_policy[VETH_INFO_MAX + 1];
	456
	457	static struct rtnl_link_ops veth_link_ops = {
	458	.kind = DRV_NAME,
	459	.priv_size = sizeof(struct veth_priv),
	460	.setup = veth_setup,
	461	.validate = veth_validate,
	462	.newlink = veth_newlink,
	463	.dellink = veth_dellink,
	464	.policy = veth_policy,
	465	.maxtype = VETH_INFO_MAX,
	466	};
	467
	468	/*
	469	* init/fini
	470	*/
	471
	472	static __init int veth_init(void)
	473	{
	474	return rtnl_link_register(&veth_link_ops);
	475	}
	476
	477	static __exit void veth_exit(void)
	478	{
68365458	479	rtnl_link_unregister(&veth_link_ops);
e314dbdc PE	480	}
	481
	482	module_init(veth_init);
	483	module_exit(veth_exit);
	484
	485	MODULE_DESCRIPTION("Virtual Ethernet Tunnel");
	486	MODULE_LICENSE("GPL v2");
	487	MODULE_ALIAS_RTNL_LINK(DRV_NAME);