[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 <linux/u64_stats_sync.h>

#include <net/dst.h>
#include <net/xfrm.h>
#include <linux/veth.h>
#include <linux/module.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 {
	u64			rx_packets;
	u64			tx_packets;
	u64			rx_bytes;
	u64			tx_bytes;
	u64			rx_dropped;
	struct u64_stats_sync	syncp;
};

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);

	/* 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;

	u64_stats_update_begin(&stats->syncp);
	stats->tx_bytes += length;
	stats->tx_packets++;
	u64_stats_update_end(&stats->syncp);

	u64_stats_update_begin(&rcv_stats->syncp);
	rcv_stats->rx_bytes += length;
	rcv_stats->rx_packets++;
	u64_stats_update_end(&rcv_stats->syncp);

	return NETDEV_TX_OK;

rx_drop:
	u64_stats_update_begin(&rcv_stats->syncp);
	rcv_stats->rx_dropped++;
	u64_stats_update_end(&rcv_stats->syncp);
	return NETDEV_TX_OK;
}

/*
 * general routines
 */

static struct rtnl_link_stats64 *veth_get_stats64(struct net_device *dev,
						  struct rtnl_link_stats64 *tot)
{
	struct veth_priv *priv = netdev_priv(dev);
	int cpu;

	for_each_possible_cpu(cpu) {
		struct veth_net_stats *stats = per_cpu_ptr(priv->stats, cpu);
		u64 rx_packets, rx_bytes, rx_dropped;
		u64 tx_packets, tx_bytes;
		unsigned int start;

		do {
			start = u64_stats_fetch_begin_bh(&stats->syncp);
			rx_packets = stats->rx_packets;
			tx_packets = stats->tx_packets;
			rx_bytes = stats->rx_bytes;
			tx_bytes = stats->tx_bytes;
			rx_dropped = stats->rx_dropped;
		} while (u64_stats_fetch_retry_bh(&stats->syncp, start));
		tot->rx_packets += rx_packets;
		tot->tx_packets += tx_packets;
		tot->rx_bytes   += rx_bytes;
		tot->tx_bytes   += tx_bytes;
		tot->rx_dropped += rx_dropped;
	}

	return tot;
}

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_stats64     = veth_get_stats64,
	.ndo_set_mac_address = eth_mac_addr,
};

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

	dev->priv_flags &= ~IFF_TX_SKB_SHARING;

	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>
e314dbdc PE	14	#include <linux/etherdevice.h>
cf05c700	15	#include <linux/u64_stats_sync.h>
e314dbdc PE	16
	17	#include <net/dst.h>
	18	#include <net/xfrm.h>
ecef969e	19	#include <linux/veth.h>
9d9779e7	20	#include <linux/module.h>
e314dbdc PE	21
	22	#define DRV_NAME "veth"
	23	#define DRV_VERSION "1.0"
	24
38d40815 EB	25	#define MIN_MTU 68 /* Min L3 MTU */
38d40815 EB	26	#define MAX_MTU 65535 /* Max L3 MTU (arbitrary) */
38d40815	27
e314dbdc	28	struct veth_net_stats {
cf05c700 ED	29	u64 rx_packets;
	30	u64 tx_packets;
	31	u64 rx_bytes;
	32	u64 tx_bytes;
	33	u64 rx_dropped;
cf05c700	34	struct u64_stats_sync syncp;
e314dbdc PE	35	};
	36
	37	struct veth_priv {
	38	struct net_device *peer;
47d74275	39	struct veth_net_stats __percpu *stats;
e314dbdc PE	40	};
e314dbdc PE	41
e314dbdc PE	42	/*
	43	* ethtool interface
	44	*/
	45
	46	static struct {
	47	const char string[ETH_GSTRING_LEN];
	48	} ethtool_stats_keys[] = {
	49	{ "peer_ifindex" },
	50	};
	51
	52	static int veth_get_settings(struct net_device dev, struct ethtool_cmd cmd)
	53	{
	54	cmd->supported = 0;
	55	cmd->advertising = 0;
70739497	56	ethtool_cmd_speed_set(cmd, SPEED_10000);
e314dbdc PE	57	cmd->duplex = DUPLEX_FULL;
	58	cmd->port = PORT_TP;
	59	cmd->phy_address = 0;
	60	cmd->transceiver = XCVR_INTERNAL;
	61	cmd->autoneg = AUTONEG_DISABLE;
	62	cmd->maxtxpkt = 0;
	63	cmd->maxrxpkt = 0;
	64	return 0;
	65	}
	66
	67	static void veth_get_drvinfo(struct net_device dev, struct ethtool_drvinfo info)
	68	{
	69	strcpy(info->driver, DRV_NAME);
	70	strcpy(info->version, DRV_VERSION);
	71	strcpy(info->fw_version, "N/A");
	72	}
	73
	74	static void veth_get_strings(struct net_device dev, u32 stringset, u8 buf)
	75	{
	76	switch(stringset) {
	77	case ETH_SS_STATS:
	78	memcpy(buf, &ethtool_stats_keys, sizeof(ethtool_stats_keys));
	79	break;
	80	}
	81	}
	82
b9f2c044	83	static int veth_get_sset_count(struct net_device *dev, int sset)
e314dbdc	84	{
b9f2c044 JG	85	switch (sset) {
	86	case ETH_SS_STATS:
	87	return ARRAY_SIZE(ethtool_stats_keys);
	88	default:
	89	return -EOPNOTSUPP;
	90	}
e314dbdc PE	91	}
	92
	93	static void veth_get_ethtool_stats(struct net_device *dev,
	94	struct ethtool_stats stats, u64 data)
	95	{
	96	struct veth_priv *priv;
	97
	98	priv = netdev_priv(dev);
	99	data[0] = priv->peer->ifindex;
	100	}
	101
0fc0b732	102	static const struct ethtool_ops veth_ethtool_ops = {
e314dbdc PE	103	.get_settings = veth_get_settings,
	104	.get_drvinfo = veth_get_drvinfo,
	105	.get_link = ethtool_op_get_link,
e314dbdc	106	.get_strings = veth_get_strings,
b9f2c044	107	.get_sset_count = veth_get_sset_count,
e314dbdc PE	108	.get_ethtool_stats = veth_get_ethtool_stats,
	109	};
	110
	111	/*
	112	* xmit
	113	*/
	114
424efe9c	115	static netdev_tx_t veth_xmit(struct sk_buff skb, struct net_device dev)
e314dbdc PE	116	{
	117	struct net_device *rcv = NULL;
	118	struct veth_priv priv, rcv_priv;
38d40815	119	struct veth_net_stats stats, rcv_stats;
e7dcaa47	120	int length;
e314dbdc	121
e314dbdc PE	122	priv = netdev_priv(dev);
	123	rcv = priv->peer;
	124	rcv_priv = netdev_priv(rcv);
	125
e7dcaa47 CL	126	stats = this_cpu_ptr(priv->stats);
e7dcaa47 CL	127	rcv_stats = this_cpu_ptr(rcv_priv->stats);
e314dbdc	128
0b796750 MM	129	/* don't change ip_summed == CHECKSUM_PARTIAL, as that
0b796750 MM	130	will cause bad checksum on forwarded packets */
a2c725fa MM	131	if (skb->ip_summed == CHECKSUM_NONE &&
	132	rcv->features & NETIF_F_RXCSUM)
	133	skb->ip_summed = CHECKSUM_UNNECESSARY;
e314dbdc	134
675071a2	135	length = skb->len;
44540960 AB	136	if (dev_forward_skb(rcv, skb) != NET_RX_SUCCESS)
44540960 AB	137	goto rx_drop;
e314dbdc	138
cf05c700	139	u64_stats_update_begin(&stats->syncp);
e314dbdc PE	140	stats->tx_bytes += length;
e314dbdc PE	141	stats->tx_packets++;
cf05c700	142	u64_stats_update_end(&stats->syncp);
e314dbdc	143
cf05c700	144	u64_stats_update_begin(&rcv_stats->syncp);
38d40815 EB	145	rcv_stats->rx_bytes += length;
38d40815 EB	146	rcv_stats->rx_packets++;
cf05c700	147	u64_stats_update_end(&rcv_stats->syncp);
e314dbdc	148
6ed10654	149	return NETDEV_TX_OK;
e314dbdc	150
38d40815	151	rx_drop:
cf05c700	152	u64_stats_update_begin(&rcv_stats->syncp);
38d40815	153	rcv_stats->rx_dropped++;
cf05c700	154	u64_stats_update_end(&rcv_stats->syncp);
6ed10654	155	return NETDEV_TX_OK;
e314dbdc PE	156	}
	157
	158	/*
	159	* general routines
	160	*/
	161
6311cc44	162	static struct rtnl_link_stats64 veth_get_stats64(struct net_device dev,
6311cc44	163	struct rtnl_link_stats64 *tot)
e314dbdc	164	{
cf05c700	165	struct veth_priv *priv = netdev_priv(dev);
11687a10	166	int cpu;
e314dbdc	167
2b1c8b0f	168	for_each_possible_cpu(cpu) {
cf05c700 ED	169	struct veth_net_stats *stats = per_cpu_ptr(priv->stats, cpu);
cf05c700 ED	170	u64 rx_packets, rx_bytes, rx_dropped;
81b16ba2	171	u64 tx_packets, tx_bytes;
cf05c700 ED	172	unsigned int start;
	173
	174	do {
	175	start = u64_stats_fetch_begin_bh(&stats->syncp);
	176	rx_packets = stats->rx_packets;
	177	tx_packets = stats->tx_packets;
	178	rx_bytes = stats->rx_bytes;
	179	tx_bytes = stats->tx_bytes;
	180	rx_dropped = stats->rx_dropped;
cf05c700 ED	181	} while (u64_stats_fetch_retry_bh(&stats->syncp, start));
	182	tot->rx_packets += rx_packets;
	183	tot->tx_packets += tx_packets;
	184	tot->rx_bytes += rx_bytes;
	185	tot->tx_bytes += tx_bytes;
	186	tot->rx_dropped += rx_dropped;
11687a10	187	}
6311cc44	188
6311cc44	189	return tot;
e314dbdc PE	190	}
	191
	192	static int veth_open(struct net_device *dev)
	193	{
	194	struct veth_priv *priv;
	195
	196	priv = netdev_priv(dev);
	197	if (priv->peer == NULL)
	198	return -ENOTCONN;
	199
	200	if (priv->peer->flags & IFF_UP) {
	201	netif_carrier_on(dev);
	202	netif_carrier_on(priv->peer);
	203	}
	204	return 0;
	205	}
	206
2cf48a10 EB	207	static int veth_close(struct net_device *dev)
	208	{
	209	struct veth_priv *priv = netdev_priv(dev);
	210
	211	netif_carrier_off(dev);
	212	netif_carrier_off(priv->peer);
	213
	214	return 0;
	215	}
	216
38d40815 EB	217	static int is_valid_veth_mtu(int new_mtu)
38d40815 EB	218	{
807540ba	219	return new_mtu >= MIN_MTU && new_mtu <= MAX_MTU;
38d40815 EB	220	}
	221
	222	static int veth_change_mtu(struct net_device *dev, int new_mtu)
	223	{
	224	if (!is_valid_veth_mtu(new_mtu))
	225	return -EINVAL;
	226	dev->mtu = new_mtu;
	227	return 0;
	228	}
	229
e314dbdc PE	230	static int veth_dev_init(struct net_device *dev)
e314dbdc PE	231	{
47d74275	232	struct veth_net_stats __percpu *stats;
e314dbdc PE	233	struct veth_priv *priv;
	234
	235	stats = alloc_percpu(struct veth_net_stats);
	236	if (stats == NULL)
	237	return -ENOMEM;
	238
	239	priv = netdev_priv(dev);
	240	priv->stats = stats;
	241	return 0;
	242	}
	243
11687a10 DM	244	static void veth_dev_free(struct net_device *dev)
	245	{
	246	struct veth_priv *priv;
	247
	248	priv = netdev_priv(dev);
	249	free_percpu(priv->stats);
	250	free_netdev(dev);
	251	}
	252
4456e7bd	253	static const struct net_device_ops veth_netdev_ops = {
ee923623 DL	254	.ndo_init = veth_dev_init,
ee923623 DL	255	.ndo_open = veth_open,
2cf48a10	256	.ndo_stop = veth_close,
ee923623	257	.ndo_start_xmit = veth_xmit,
38d40815	258	.ndo_change_mtu = veth_change_mtu,
6311cc44	259	.ndo_get_stats64 = veth_get_stats64,
ee923623	260	.ndo_set_mac_address = eth_mac_addr,
4456e7bd SH	261	};
4456e7bd SH	262
e314dbdc PE	263	static void veth_setup(struct net_device *dev)
	264	{
	265	ether_setup(dev);
	266
550fd08c NH	267	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
550fd08c NH	268
4456e7bd	269	dev->netdev_ops = &veth_netdev_ops;
e314dbdc PE	270	dev->ethtool_ops = &veth_ethtool_ops;
e314dbdc PE	271	dev->features \|= NETIF_F_LLTX;
11687a10	272	dev->destructor = veth_dev_free;
a2c725fa MM	273
a2c725fa MM	274	dev->hw_features = NETIF_F_NO_CSUM \| NETIF_F_SG \| NETIF_F_RXCSUM;
e314dbdc PE	275	}
	276
	277	/*
	278	* netlink interface
	279	*/
	280
	281	static int veth_validate(struct nlattr tb[], struct nlattr data[])
	282	{
	283	if (tb[IFLA_ADDRESS]) {
	284	if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN)
	285	return -EINVAL;
	286	if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS])))
	287	return -EADDRNOTAVAIL;
	288	}
38d40815 EB	289	if (tb[IFLA_MTU]) {
	290	if (!is_valid_veth_mtu(nla_get_u32(tb[IFLA_MTU])))
	291	return -EINVAL;
	292	}
e314dbdc PE	293	return 0;
	294	}
	295
	296	static struct rtnl_link_ops veth_link_ops;
	297
81adee47	298	static int veth_newlink(struct net src_net, struct net_device dev,
e314dbdc PE	299	struct nlattr tb[], struct nlattr data[])
	300	{
	301	int err;
	302	struct net_device *peer;
	303	struct veth_priv *priv;
	304	char ifname[IFNAMSIZ];
	305	struct nlattr peer_tb[IFLA_MAX + 1], *tbp;
3729d502	306	struct ifinfomsg *ifmp;
81adee47	307	struct net *net;
e314dbdc PE	308
	309	/*
	310	* create and register peer first
e314dbdc	311	*/
e314dbdc PE	312	if (data != NULL && data[VETH_INFO_PEER] != NULL) {
	313	struct nlattr *nla_peer;
	314
	315	nla_peer = data[VETH_INFO_PEER];
3729d502	316	ifmp = nla_data(nla_peer);
e314dbdc PE	317	err = nla_parse(peer_tb, IFLA_MAX,
	318	nla_data(nla_peer) + sizeof(struct ifinfomsg),
	319	nla_len(nla_peer) - sizeof(struct ifinfomsg),
	320	ifla_policy);
	321	if (err < 0)
	322	return err;
	323
	324	err = veth_validate(peer_tb, NULL);
	325	if (err < 0)
	326	return err;
	327
	328	tbp = peer_tb;
3729d502 PM	329	} else {
3729d502 PM	330	ifmp = NULL;
e314dbdc	331	tbp = tb;
3729d502	332	}
e314dbdc PE	333
	334	if (tbp[IFLA_IFNAME])
	335	nla_strlcpy(ifname, tbp[IFLA_IFNAME], IFNAMSIZ);
	336	else
	337	snprintf(ifname, IFNAMSIZ, DRV_NAME "%%d");
	338
81adee47 EB	339	net = rtnl_link_get_net(src_net, tbp);
	340	if (IS_ERR(net))
	341	return PTR_ERR(net);
	342
	343	peer = rtnl_create_link(src_net, net, ifname, &veth_link_ops, tbp);
	344	if (IS_ERR(peer)) {
	345	put_net(net);
e314dbdc	346	return PTR_ERR(peer);
81adee47	347	}
e314dbdc PE	348
	349	if (tbp[IFLA_ADDRESS] == NULL)
	350	random_ether_addr(peer->dev_addr);
	351
	352	err = register_netdevice(peer);
81adee47 EB	353	put_net(net);
81adee47 EB	354	net = NULL;
e314dbdc PE	355	if (err < 0)
	356	goto err_register_peer;
	357
	358	netif_carrier_off(peer);
	359
3729d502 PM	360	err = rtnl_configure_link(peer, ifmp);
	361	if (err < 0)
	362	goto err_configure_peer;
	363
e314dbdc PE	364	/*
	365	* register dev last
	366	*
	367	* note, that since we've registered new device the dev's name
	368	* should be re-allocated
	369	*/
	370
	371	if (tb[IFLA_ADDRESS] == NULL)
	372	random_ether_addr(dev->dev_addr);
	373
6c8c4446 JP	374	if (tb[IFLA_IFNAME])
	375	nla_strlcpy(dev->name, tb[IFLA_IFNAME], IFNAMSIZ);
	376	else
	377	snprintf(dev->name, IFNAMSIZ, DRV_NAME "%%d");
	378
	379	if (strchr(dev->name, '%')) {
	380	err = dev_alloc_name(dev, dev->name);
	381	if (err < 0)
	382	goto err_alloc_name;
	383	}
	384
e314dbdc PE	385	err = register_netdevice(dev);
	386	if (err < 0)
	387	goto err_register_dev;
	388
	389	netif_carrier_off(dev);
	390
	391	/*
	392	* tie the deviced together
	393	*/
	394
	395	priv = netdev_priv(dev);
e314dbdc	396	priv->peer = peer;
e314dbdc PE	397
e314dbdc PE	398	priv = netdev_priv(peer);
e314dbdc	399	priv->peer = dev;
e314dbdc PE	400	return 0;
	401
	402	err_register_dev:
	403	/* nothing to do */
6c8c4446	404	err_alloc_name:
3729d502	405	err_configure_peer:
e314dbdc PE	406	unregister_netdevice(peer);
	407	return err;
	408
	409	err_register_peer:
	410	free_netdev(peer);
	411	return err;
	412	}
	413
23289a37	414	static void veth_dellink(struct net_device dev, struct list_head head)
e314dbdc PE	415	{
	416	struct veth_priv *priv;
	417	struct net_device *peer;
	418
	419	priv = netdev_priv(dev);
	420	peer = priv->peer;
	421
24540535 ED	422	unregister_netdevice_queue(dev, head);
24540535 ED	423	unregister_netdevice_queue(peer, head);
e314dbdc PE	424	}
	425
	426	static const struct nla_policy veth_policy[VETH_INFO_MAX + 1];
	427
	428	static struct rtnl_link_ops veth_link_ops = {
	429	.kind = DRV_NAME,
	430	.priv_size = sizeof(struct veth_priv),
	431	.setup = veth_setup,
	432	.validate = veth_validate,
	433	.newlink = veth_newlink,
	434	.dellink = veth_dellink,
	435	.policy = veth_policy,
	436	.maxtype = VETH_INFO_MAX,
	437	};
	438
	439	/*
	440	* init/fini
	441	*/
	442
	443	static __init int veth_init(void)
	444	{
	445	return rtnl_link_register(&veth_link_ops);
	446	}
	447
	448	static __exit void veth_exit(void)
	449	{
68365458	450	rtnl_link_unregister(&veth_link_ops);
e314dbdc PE	451	}
	452
	453	module_init(veth_init);
	454	module_exit(veth_exit);
	455
	456	MODULE_DESCRIPTION("Virtual Ethernet Tunnel");
	457	MODULE_LICENSE("GPL v2");
	458	MODULE_ALIAS_RTNL_LINK(DRV_NAME);