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

/*
 *  drivers/net/veth.c
 *
 *  Copyright (C) 2007 OpenVZ http://openvz.org, SWsoft Inc
 *
 * Author: Pavel Emelianov <xemul@openvz.org>
 * Ethtool interface from: Eric W. Biederman <ebiederm@xmission.com>
 *
 */

#include <linux/netdevice.h>
#include <linux/slab.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) */

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 __percpu *stats;
};

/*
 * 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;
	ethtool_cmd_speed_set(cmd, 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 const struct ethtool_ops veth_ethtool_ops = {
	.get_settings		= veth_get_settings,
	.get_drvinfo		= veth_get_drvinfo,
	.get_link		= ethtool_op_get_link,
	.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;

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

	stats = this_cpu_ptr(priv->stats);
	rcv_stats = this_cpu_ptr(rcv_priv->stats);

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

	/* don't change ip_summed == CHECKSUM_PARTIAL, as that
	   will cause bad checksum on forwarded packets */
	if (skb->ip_summed == CHECKSUM_NONE &&
	    rcv->features & NETIF_F_RXCSUM)
		skb->ip_summed = CHECKSUM_UNNECESSARY;

	length = skb->len;
	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:
	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;
	int cpu;
	struct veth_net_stats *stats, total = {0};

	priv = netdev_priv(dev);

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

		total.rx_packets += stats->rx_packets;
		total.tx_packets += stats->tx_packets;
		total.rx_bytes   += stats->rx_bytes;
		total.tx_bytes   += stats->tx_bytes;
		total.tx_dropped += stats->tx_dropped;
		total.rx_dropped += stats->rx_dropped;
	}
	dev->stats.rx_packets = total.rx_packets;
	dev->stats.tx_packets = total.tx_packets;
	dev->stats.rx_bytes   = total.rx_bytes;
	dev->stats.tx_bytes   = total.tx_bytes;
	dev->stats.tx_dropped = total.tx_dropped;
	dev->stats.rx_dropped = total.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 __percpu *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;

	dev->hw_features = NETIF_F_NO_CSUM | NETIF_F_SG | NETIF_F_RXCSUM;
}

/*
 * 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 ifinfomsg *ifmp;
	struct net *net;

	/*
	 * create and register peer first
	 */
	if (data != NULL && data[VETH_INFO_PEER] != NULL) {
		struct nlattr *nla_peer;

		nla_peer = data[VETH_INFO_PEER];
		ifmp = nla_data(nla_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 {
		ifmp = NULL;
		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);

	err = rtnl_configure_link(peer, ifmp);
	if (err < 0)
		goto err_configure_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:
err_configure_peer:
	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 <xemul@openvz.org>
	7	* Ethtool interface from: Eric W. Biederman <ebiederm@xmission.com>
	8	*
	9	*/
	10
e314dbdc	11	#include <linux/netdevice.h>
5a0e3ad6	12	#include <linux/slab.h>
e314dbdc PE	13	#include <linux/ethtool.h>
	14	#include <linux/etherdevice.h>
	15
	16	#include <net/dst.h>
	17	#include <net/xfrm.h>
ecef969e	18	#include <linux/veth.h>
e314dbdc PE	19
	20	#define DRV_NAME "veth"
	21	#define DRV_VERSION "1.0"
	22
38d40815 EB	23	#define MIN_MTU 68 /* Min L3 MTU */
38d40815 EB	24	#define MAX_MTU 65535 /* Max L3 MTU (arbitrary) */
38d40815	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;
47d74275	37	struct veth_net_stats __percpu *stats;
e314dbdc PE	38	};
e314dbdc PE	39
e314dbdc PE	40	/*
	41	* ethtool interface
	42	*/
	43
	44	static struct {
	45	const char string[ETH_GSTRING_LEN];
	46	} ethtool_stats_keys[] = {
	47	{ "peer_ifindex" },
	48	};
	49
	50	static int veth_get_settings(struct net_device dev, struct ethtool_cmd cmd)
	51	{
	52	cmd->supported = 0;
	53	cmd->advertising = 0;
70739497	54	ethtool_cmd_speed_set(cmd, SPEED_10000);
e314dbdc PE	55	cmd->duplex = DUPLEX_FULL;
	56	cmd->port = PORT_TP;
	57	cmd->phy_address = 0;
	58	cmd->transceiver = XCVR_INTERNAL;
	59	cmd->autoneg = AUTONEG_DISABLE;
	60	cmd->maxtxpkt = 0;
	61	cmd->maxrxpkt = 0;
	62	return 0;
	63	}
	64
	65	static void veth_get_drvinfo(struct net_device dev, struct ethtool_drvinfo info)
	66	{
	67	strcpy(info->driver, DRV_NAME);
	68	strcpy(info->version, DRV_VERSION);
	69	strcpy(info->fw_version, "N/A");
	70	}
	71
	72	static void veth_get_strings(struct net_device dev, u32 stringset, u8 buf)
	73	{
	74	switch(stringset) {
	75	case ETH_SS_STATS:
	76	memcpy(buf, &ethtool_stats_keys, sizeof(ethtool_stats_keys));
	77	break;
	78	}
	79	}
	80
b9f2c044	81	static int veth_get_sset_count(struct net_device *dev, int sset)
e314dbdc	82	{
b9f2c044 JG	83	switch (sset) {
	84	case ETH_SS_STATS:
	85	return ARRAY_SIZE(ethtool_stats_keys);
	86	default:
	87	return -EOPNOTSUPP;
	88	}
e314dbdc PE	89	}
	90
	91	static void veth_get_ethtool_stats(struct net_device *dev,
	92	struct ethtool_stats stats, u64 data)
	93	{
	94	struct veth_priv *priv;
	95
	96	priv = netdev_priv(dev);
	97	data[0] = priv->peer->ifindex;
	98	}
	99
0fc0b732	100	static const struct ethtool_ops veth_ethtool_ops = {
e314dbdc PE	101	.get_settings = veth_get_settings,
	102	.get_drvinfo = veth_get_drvinfo,
	103	.get_link = ethtool_op_get_link,
e314dbdc	104	.get_strings = veth_get_strings,
b9f2c044	105	.get_sset_count = veth_get_sset_count,
e314dbdc PE	106	.get_ethtool_stats = veth_get_ethtool_stats,
	107	};
	108
	109	/*
	110	* xmit
	111	*/
	112
424efe9c	113	static netdev_tx_t veth_xmit(struct sk_buff skb, struct net_device dev)
e314dbdc PE	114	{
	115	struct net_device *rcv = NULL;
	116	struct veth_priv priv, rcv_priv;
38d40815	117	struct veth_net_stats stats, rcv_stats;
e7dcaa47	118	int length;
e314dbdc	119
e314dbdc PE	120	priv = netdev_priv(dev);
	121	rcv = priv->peer;
	122	rcv_priv = netdev_priv(rcv);
	123
e7dcaa47 CL	124	stats = this_cpu_ptr(priv->stats);
e7dcaa47 CL	125	rcv_stats = this_cpu_ptr(rcv_priv->stats);
e314dbdc PE	126
e314dbdc PE	127	if (!(rcv->flags & IFF_UP))
38d40815 EB	128	goto tx_drop;
38d40815 EB	129
0b796750 MM	130	/* don't change ip_summed == CHECKSUM_PARTIAL, as that
0b796750 MM	131	will cause bad checksum on forwarded packets */
a2c725fa MM	132	if (skb->ip_summed == CHECKSUM_NONE &&
	133	rcv->features & NETIF_F_RXCSUM)
	134	skb->ip_summed = CHECKSUM_UNNECESSARY;
e314dbdc	135
675071a2	136	length = skb->len;
44540960 AB	137	if (dev_forward_skb(rcv, skb) != NET_RX_SUCCESS)
44540960 AB	138	goto rx_drop;
e314dbdc PE	139
	140	stats->tx_bytes += length;
	141	stats->tx_packets++;
	142
38d40815 EB	143	rcv_stats->rx_bytes += length;
38d40815 EB	144	rcv_stats->rx_packets++;
e314dbdc	145
6ed10654	146	return NETDEV_TX_OK;
e314dbdc	147
38d40815	148	tx_drop:
e314dbdc PE	149	kfree_skb(skb);
e314dbdc PE	150	stats->tx_dropped++;
6ed10654	151	return NETDEV_TX_OK;
38d40815 EB	152
38d40815 EB	153	rx_drop:
38d40815	154	rcv_stats->rx_dropped++;
6ed10654	155	return NETDEV_TX_OK;
e314dbdc PE	156	}
	157
	158	/*
	159	* general routines
	160	*/
	161
	162	static struct net_device_stats veth_get_stats(struct net_device dev)
	163	{
11687a10	164	struct veth_priv *priv;
11687a10	165	int cpu;
2b1c8b0f	166	struct veth_net_stats *stats, total = {0};
e314dbdc	167
11687a10	168	priv = netdev_priv(dev);
e314dbdc	169
2b1c8b0f	170	for_each_possible_cpu(cpu) {
11687a10	171	stats = per_cpu_ptr(priv->stats, cpu);
e314dbdc	172
2b1c8b0f ED	173	total.rx_packets += stats->rx_packets;
	174	total.tx_packets += stats->tx_packets;
	175	total.rx_bytes += stats->rx_bytes;
	176	total.tx_bytes += stats->tx_bytes;
	177	total.tx_dropped += stats->tx_dropped;
	178	total.rx_dropped += stats->rx_dropped;
11687a10	179	}
2b1c8b0f ED	180	dev->stats.rx_packets = total.rx_packets;
	181	dev->stats.tx_packets = total.tx_packets;
	182	dev->stats.rx_bytes = total.rx_bytes;
	183	dev->stats.tx_bytes = total.tx_bytes;
	184	dev->stats.tx_dropped = total.tx_dropped;
	185	dev->stats.rx_dropped = total.rx_dropped;
	186
	187	return &dev->stats;
e314dbdc PE	188	}
	189
	190	static int veth_open(struct net_device *dev)
	191	{
	192	struct veth_priv *priv;
	193
	194	priv = netdev_priv(dev);
	195	if (priv->peer == NULL)
	196	return -ENOTCONN;
	197
	198	if (priv->peer->flags & IFF_UP) {
	199	netif_carrier_on(dev);
	200	netif_carrier_on(priv->peer);
	201	}
	202	return 0;
	203	}
	204
2cf48a10 EB	205	static int veth_close(struct net_device *dev)
	206	{
	207	struct veth_priv *priv = netdev_priv(dev);
	208
	209	netif_carrier_off(dev);
	210	netif_carrier_off(priv->peer);
	211
	212	return 0;
	213	}
	214
38d40815 EB	215	static int is_valid_veth_mtu(int new_mtu)
38d40815 EB	216	{
807540ba	217	return new_mtu >= MIN_MTU && new_mtu <= MAX_MTU;
38d40815 EB	218	}
	219
	220	static int veth_change_mtu(struct net_device *dev, int new_mtu)
	221	{
	222	if (!is_valid_veth_mtu(new_mtu))
	223	return -EINVAL;
	224	dev->mtu = new_mtu;
	225	return 0;
	226	}
	227
e314dbdc PE	228	static int veth_dev_init(struct net_device *dev)
e314dbdc PE	229	{
47d74275	230	struct veth_net_stats __percpu *stats;
e314dbdc PE	231	struct veth_priv *priv;
	232
	233	stats = alloc_percpu(struct veth_net_stats);
	234	if (stats == NULL)
	235	return -ENOMEM;
	236
	237	priv = netdev_priv(dev);
	238	priv->stats = stats;
	239	return 0;
	240	}
	241
11687a10 DM	242	static void veth_dev_free(struct net_device *dev)
	243	{
	244	struct veth_priv *priv;
	245
	246	priv = netdev_priv(dev);
	247	free_percpu(priv->stats);
	248	free_netdev(dev);
	249	}
	250
4456e7bd	251	static const struct net_device_ops veth_netdev_ops = {
ee923623 DL	252	.ndo_init = veth_dev_init,
ee923623 DL	253	.ndo_open = veth_open,
2cf48a10	254	.ndo_stop = veth_close,
ee923623	255	.ndo_start_xmit = veth_xmit,
38d40815	256	.ndo_change_mtu = veth_change_mtu,
ee923623 DL	257	.ndo_get_stats = veth_get_stats,
ee923623 DL	258	.ndo_set_mac_address = eth_mac_addr,
4456e7bd SH	259	};
4456e7bd SH	260
e314dbdc PE	261	static void veth_setup(struct net_device *dev)
	262	{
	263	ether_setup(dev);
	264
4456e7bd	265	dev->netdev_ops = &veth_netdev_ops;
e314dbdc PE	266	dev->ethtool_ops = &veth_ethtool_ops;
e314dbdc PE	267	dev->features \|= NETIF_F_LLTX;
11687a10	268	dev->destructor = veth_dev_free;
a2c725fa MM	269
a2c725fa MM	270	dev->hw_features = NETIF_F_NO_CSUM \| NETIF_F_SG \| NETIF_F_RXCSUM;
e314dbdc PE	271	}
	272
	273	/*
	274	* netlink interface
	275	*/
	276
	277	static int veth_validate(struct nlattr tb[], struct nlattr data[])
	278	{
	279	if (tb[IFLA_ADDRESS]) {
	280	if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN)
	281	return -EINVAL;
	282	if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS])))
	283	return -EADDRNOTAVAIL;
	284	}
38d40815 EB	285	if (tb[IFLA_MTU]) {
	286	if (!is_valid_veth_mtu(nla_get_u32(tb[IFLA_MTU])))
	287	return -EINVAL;
	288	}
e314dbdc PE	289	return 0;
	290	}
	291
	292	static struct rtnl_link_ops veth_link_ops;
	293
81adee47	294	static int veth_newlink(struct net src_net, struct net_device dev,
e314dbdc PE	295	struct nlattr tb[], struct nlattr data[])
	296	{
	297	int err;
	298	struct net_device *peer;
	299	struct veth_priv *priv;
	300	char ifname[IFNAMSIZ];
	301	struct nlattr peer_tb[IFLA_MAX + 1], *tbp;
3729d502	302	struct ifinfomsg *ifmp;
81adee47	303	struct net *net;
e314dbdc PE	304
	305	/*
	306	* create and register peer first
e314dbdc	307	*/
e314dbdc PE	308	if (data != NULL && data[VETH_INFO_PEER] != NULL) {
	309	struct nlattr *nla_peer;
	310
	311	nla_peer = data[VETH_INFO_PEER];
3729d502	312	ifmp = nla_data(nla_peer);
e314dbdc PE	313	err = nla_parse(peer_tb, IFLA_MAX,
	314	nla_data(nla_peer) + sizeof(struct ifinfomsg),
	315	nla_len(nla_peer) - sizeof(struct ifinfomsg),
	316	ifla_policy);
	317	if (err < 0)
	318	return err;
	319
	320	err = veth_validate(peer_tb, NULL);
	321	if (err < 0)
	322	return err;
	323
	324	tbp = peer_tb;
3729d502 PM	325	} else {
3729d502 PM	326	ifmp = NULL;
e314dbdc	327	tbp = tb;
3729d502	328	}
e314dbdc PE	329
	330	if (tbp[IFLA_IFNAME])
	331	nla_strlcpy(ifname, tbp[IFLA_IFNAME], IFNAMSIZ);
	332	else
	333	snprintf(ifname, IFNAMSIZ, DRV_NAME "%%d");
	334
81adee47 EB	335	net = rtnl_link_get_net(src_net, tbp);
	336	if (IS_ERR(net))
	337	return PTR_ERR(net);
	338
	339	peer = rtnl_create_link(src_net, net, ifname, &veth_link_ops, tbp);
	340	if (IS_ERR(peer)) {
	341	put_net(net);
e314dbdc	342	return PTR_ERR(peer);
81adee47	343	}
e314dbdc PE	344
	345	if (tbp[IFLA_ADDRESS] == NULL)
	346	random_ether_addr(peer->dev_addr);
	347
	348	err = register_netdevice(peer);
81adee47 EB	349	put_net(net);
81adee47 EB	350	net = NULL;
e314dbdc PE	351	if (err < 0)
	352	goto err_register_peer;
	353
	354	netif_carrier_off(peer);
	355
3729d502 PM	356	err = rtnl_configure_link(peer, ifmp);
	357	if (err < 0)
	358	goto err_configure_peer;
	359
e314dbdc PE	360	/*
	361	* register dev last
	362	*
	363	* note, that since we've registered new device the dev's name
	364	* should be re-allocated
	365	*/
	366
	367	if (tb[IFLA_ADDRESS] == NULL)
	368	random_ether_addr(dev->dev_addr);
	369
6c8c4446 JP	370	if (tb[IFLA_IFNAME])
	371	nla_strlcpy(dev->name, tb[IFLA_IFNAME], IFNAMSIZ);
	372	else
	373	snprintf(dev->name, IFNAMSIZ, DRV_NAME "%%d");
	374
	375	if (strchr(dev->name, '%')) {
	376	err = dev_alloc_name(dev, dev->name);
	377	if (err < 0)
	378	goto err_alloc_name;
	379	}
	380
e314dbdc PE	381	err = register_netdevice(dev);
	382	if (err < 0)
	383	goto err_register_dev;
	384
	385	netif_carrier_off(dev);
	386
	387	/*
	388	* tie the deviced together
	389	*/
	390
	391	priv = netdev_priv(dev);
e314dbdc	392	priv->peer = peer;
e314dbdc PE	393
e314dbdc PE	394	priv = netdev_priv(peer);
e314dbdc	395	priv->peer = dev;
e314dbdc PE	396	return 0;
	397
	398	err_register_dev:
	399	/* nothing to do */
6c8c4446	400	err_alloc_name:
3729d502	401	err_configure_peer:
e314dbdc PE	402	unregister_netdevice(peer);
	403	return err;
	404
	405	err_register_peer:
	406	free_netdev(peer);
	407	return err;
	408	}
	409
23289a37	410	static void veth_dellink(struct net_device dev, struct list_head head)
e314dbdc PE	411	{
	412	struct veth_priv *priv;
	413	struct net_device *peer;
	414
	415	priv = netdev_priv(dev);
	416	peer = priv->peer;
	417
24540535 ED	418	unregister_netdevice_queue(dev, head);
24540535 ED	419	unregister_netdevice_queue(peer, head);
e314dbdc PE	420	}
	421
	422	static const struct nla_policy veth_policy[VETH_INFO_MAX + 1];
	423
	424	static struct rtnl_link_ops veth_link_ops = {
	425	.kind = DRV_NAME,
	426	.priv_size = sizeof(struct veth_priv),
	427	.setup = veth_setup,
	428	.validate = veth_validate,
	429	.newlink = veth_newlink,
	430	.dellink = veth_dellink,
	431	.policy = veth_policy,
	432	.maxtype = VETH_INFO_MAX,
	433	};
	434
	435	/*
	436	* init/fini
	437	*/
	438
	439	static __init int veth_init(void)
	440	{
	441	return rtnl_link_register(&veth_link_ops);
	442	}
	443
	444	static __exit void veth_exit(void)
	445	{
68365458	446	rtnl_link_unregister(&veth_link_ops);
e314dbdc PE	447	}
	448
	449	module_init(veth_init);
	450	module_exit(veth_exit);
	451
	452	MODULE_DESCRIPTION("Virtual Ethernet Tunnel");
	453	MODULE_LICENSE("GPL v2");
	454	MODULE_ALIAS_RTNL_LINK(DRV_NAME);