[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/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			rx_bytes;
	u64			tx_packets;
	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)
{
	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
}

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_HW_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)
		eth_hw_addr_random(peer);

	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)
		eth_hw_addr_random(dev);

	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] = {
	[VETH_INFO_PEER]	= { .len = sizeof(struct ifinfomsg) },
};

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	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	29	u64 rx_packets;
cf05c700	30	u64 rx_bytes;
8ce120f1	31	u64 tx_packets;
cf05c700 ED	32	u64 tx_bytes;
cf05c700 ED	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	{
33a5ba14 RJ	69	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
33a5ba14 RJ	70	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
e314dbdc PE	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
0fc0b732	101	static const struct ethtool_ops veth_ethtool_ops = {
e314dbdc PE	102	.get_settings = veth_get_settings,
	103	.get_drvinfo = veth_get_drvinfo,
	104	.get_link = ethtool_op_get_link,
e314dbdc	105	.get_strings = veth_get_strings,
b9f2c044	106	.get_sset_count = veth_get_sset_count,
e314dbdc PE	107	.get_ethtool_stats = veth_get_ethtool_stats,
	108	};
	109
	110	/*
	111	* xmit
	112	*/
	113
424efe9c	114	static netdev_tx_t veth_xmit(struct sk_buff skb, struct net_device dev)
e314dbdc PE	115	{
	116	struct net_device *rcv = NULL;
	117	struct veth_priv priv, rcv_priv;
38d40815	118	struct veth_net_stats stats, rcv_stats;
e7dcaa47	119	int length;
e314dbdc	120
e314dbdc PE	121	priv = netdev_priv(dev);
	122	rcv = priv->peer;
	123	rcv_priv = netdev_priv(rcv);
	124
e7dcaa47 CL	125	stats = this_cpu_ptr(priv->stats);
e7dcaa47 CL	126	rcv_stats = this_cpu_ptr(rcv_priv->stats);
e314dbdc	127
0b796750 MM	128	/* don't change ip_summed == CHECKSUM_PARTIAL, as that
0b796750 MM	129	will cause bad checksum on forwarded packets */
a2c725fa MM	130	if (skb->ip_summed == CHECKSUM_NONE &&
	131	rcv->features & NETIF_F_RXCSUM)
	132	skb->ip_summed = CHECKSUM_UNNECESSARY;
e314dbdc	133
675071a2	134	length = skb->len;
44540960 AB	135	if (dev_forward_skb(rcv, skb) != NET_RX_SUCCESS)
44540960 AB	136	goto rx_drop;
e314dbdc	137
cf05c700	138	u64_stats_update_begin(&stats->syncp);
e314dbdc PE	139	stats->tx_bytes += length;
e314dbdc PE	140	stats->tx_packets++;
cf05c700	141	u64_stats_update_end(&stats->syncp);
e314dbdc	142
cf05c700	143	u64_stats_update_begin(&rcv_stats->syncp);
38d40815 EB	144	rcv_stats->rx_bytes += length;
38d40815 EB	145	rcv_stats->rx_packets++;
cf05c700	146	u64_stats_update_end(&rcv_stats->syncp);
e314dbdc	147
6ed10654	148	return NETDEV_TX_OK;
e314dbdc	149
38d40815	150	rx_drop:
cf05c700	151	u64_stats_update_begin(&rcv_stats->syncp);
38d40815	152	rcv_stats->rx_dropped++;
cf05c700	153	u64_stats_update_end(&rcv_stats->syncp);
6ed10654	154	return NETDEV_TX_OK;
e314dbdc PE	155	}
	156
	157	/*
	158	* general routines
	159	*/
	160
6311cc44	161	static struct rtnl_link_stats64 veth_get_stats64(struct net_device dev,
6311cc44	162	struct rtnl_link_stats64 *tot)
e314dbdc	163	{
cf05c700	164	struct veth_priv *priv = netdev_priv(dev);
11687a10	165	int cpu;
e314dbdc	166
2b1c8b0f	167	for_each_possible_cpu(cpu) {
cf05c700 ED	168	struct veth_net_stats *stats = per_cpu_ptr(priv->stats, cpu);
cf05c700 ED	169	u64 rx_packets, rx_bytes, rx_dropped;
81b16ba2	170	u64 tx_packets, tx_bytes;
cf05c700 ED	171	unsigned int start;
	172
	173	do {
	174	start = u64_stats_fetch_begin_bh(&stats->syncp);
	175	rx_packets = stats->rx_packets;
	176	tx_packets = stats->tx_packets;
	177	rx_bytes = stats->rx_bytes;
	178	tx_bytes = stats->tx_bytes;
	179	rx_dropped = stats->rx_dropped;
cf05c700 ED	180	} while (u64_stats_fetch_retry_bh(&stats->syncp, start));
	181	tot->rx_packets += rx_packets;
	182	tot->tx_packets += tx_packets;
	183	tot->rx_bytes += rx_bytes;
	184	tot->tx_bytes += tx_bytes;
	185	tot->rx_dropped += rx_dropped;
11687a10	186	}
6311cc44	187
6311cc44	188	return tot;
e314dbdc PE	189	}
	190
	191	static int veth_open(struct net_device *dev)
	192	{
	193	struct veth_priv *priv;
	194
	195	priv = netdev_priv(dev);
	196	if (priv->peer == NULL)
	197	return -ENOTCONN;
	198
	199	if (priv->peer->flags & IFF_UP) {
	200	netif_carrier_on(dev);
	201	netif_carrier_on(priv->peer);
	202	}
	203	return 0;
	204	}
	205
2cf48a10 EB	206	static int veth_close(struct net_device *dev)
	207	{
	208	struct veth_priv *priv = netdev_priv(dev);
	209
	210	netif_carrier_off(dev);
	211	netif_carrier_off(priv->peer);
	212
	213	return 0;
	214	}
	215
38d40815 EB	216	static int is_valid_veth_mtu(int new_mtu)
38d40815 EB	217	{
807540ba	218	return new_mtu >= MIN_MTU && new_mtu <= MAX_MTU;
38d40815 EB	219	}
	220
	221	static int veth_change_mtu(struct net_device *dev, int new_mtu)
	222	{
	223	if (!is_valid_veth_mtu(new_mtu))
	224	return -EINVAL;
	225	dev->mtu = new_mtu;
	226	return 0;
	227	}
	228
e314dbdc PE	229	static int veth_dev_init(struct net_device *dev)
e314dbdc PE	230	{
47d74275	231	struct veth_net_stats __percpu *stats;
e314dbdc PE	232	struct veth_priv *priv;
	233
	234	stats = alloc_percpu(struct veth_net_stats);
	235	if (stats == NULL)
	236	return -ENOMEM;
	237
	238	priv = netdev_priv(dev);
	239	priv->stats = stats;
	240	return 0;
	241	}
	242
11687a10 DM	243	static void veth_dev_free(struct net_device *dev)
	244	{
	245	struct veth_priv *priv;
	246
	247	priv = netdev_priv(dev);
	248	free_percpu(priv->stats);
	249	free_netdev(dev);
	250	}
	251
4456e7bd	252	static const struct net_device_ops veth_netdev_ops = {
ee923623 DL	253	.ndo_init = veth_dev_init,
ee923623 DL	254	.ndo_open = veth_open,
2cf48a10	255	.ndo_stop = veth_close,
ee923623	256	.ndo_start_xmit = veth_xmit,
38d40815	257	.ndo_change_mtu = veth_change_mtu,
6311cc44	258	.ndo_get_stats64 = veth_get_stats64,
ee923623	259	.ndo_set_mac_address = eth_mac_addr,
4456e7bd SH	260	};
4456e7bd SH	261
e314dbdc PE	262	static void veth_setup(struct net_device *dev)
	263	{
	264	ether_setup(dev);
	265
550fd08c NH	266	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
550fd08c NH	267
4456e7bd	268	dev->netdev_ops = &veth_netdev_ops;
e314dbdc PE	269	dev->ethtool_ops = &veth_ethtool_ops;
e314dbdc PE	270	dev->features \|= NETIF_F_LLTX;
11687a10	271	dev->destructor = veth_dev_free;
a2c725fa	272
34324dc2	273	dev->hw_features = NETIF_F_HW_CSUM \| NETIF_F_SG \| NETIF_F_RXCSUM;
e314dbdc PE	274	}
	275
	276	/*
	277	* netlink interface
	278	*/
	279
	280	static int veth_validate(struct nlattr tb[], struct nlattr data[])
	281	{
	282	if (tb[IFLA_ADDRESS]) {
	283	if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN)
	284	return -EINVAL;
	285	if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS])))
	286	return -EADDRNOTAVAIL;
	287	}
38d40815 EB	288	if (tb[IFLA_MTU]) {
	289	if (!is_valid_veth_mtu(nla_get_u32(tb[IFLA_MTU])))
	290	return -EINVAL;
	291	}
e314dbdc PE	292	return 0;
	293	}
	294
	295	static struct rtnl_link_ops veth_link_ops;
	296
81adee47	297	static int veth_newlink(struct net src_net, struct net_device dev,
e314dbdc PE	298	struct nlattr tb[], struct nlattr data[])
	299	{
	300	int err;
	301	struct net_device *peer;
	302	struct veth_priv *priv;
	303	char ifname[IFNAMSIZ];
	304	struct nlattr peer_tb[IFLA_MAX + 1], *tbp;
3729d502	305	struct ifinfomsg *ifmp;
81adee47	306	struct net *net;
e314dbdc PE	307
	308	/*
	309	* create and register peer first
e314dbdc	310	*/
e314dbdc PE	311	if (data != NULL && data[VETH_INFO_PEER] != NULL) {
	312	struct nlattr *nla_peer;
	313
	314	nla_peer = data[VETH_INFO_PEER];
3729d502	315	ifmp = nla_data(nla_peer);
e314dbdc PE	316	err = nla_parse(peer_tb, IFLA_MAX,
	317	nla_data(nla_peer) + sizeof(struct ifinfomsg),
	318	nla_len(nla_peer) - sizeof(struct ifinfomsg),
	319	ifla_policy);
	320	if (err < 0)
	321	return err;
	322
	323	err = veth_validate(peer_tb, NULL);
	324	if (err < 0)
	325	return err;
	326
	327	tbp = peer_tb;
3729d502 PM	328	} else {
3729d502 PM	329	ifmp = NULL;
e314dbdc	330	tbp = tb;
3729d502	331	}
e314dbdc PE	332
	333	if (tbp[IFLA_IFNAME])
	334	nla_strlcpy(ifname, tbp[IFLA_IFNAME], IFNAMSIZ);
	335	else
	336	snprintf(ifname, IFNAMSIZ, DRV_NAME "%%d");
	337
81adee47 EB	338	net = rtnl_link_get_net(src_net, tbp);
	339	if (IS_ERR(net))
	340	return PTR_ERR(net);
	341
	342	peer = rtnl_create_link(src_net, net, ifname, &veth_link_ops, tbp);
	343	if (IS_ERR(peer)) {
	344	put_net(net);
e314dbdc	345	return PTR_ERR(peer);
81adee47	346	}
e314dbdc PE	347
e314dbdc PE	348	if (tbp[IFLA_ADDRESS] == NULL)
f2cedb63	349	eth_hw_addr_random(peer);
e314dbdc PE	350
e314dbdc PE	351	err = register_netdevice(peer);
81adee47 EB	352	put_net(net);
81adee47 EB	353	net = NULL;
e314dbdc PE	354	if (err < 0)
	355	goto err_register_peer;
	356
	357	netif_carrier_off(peer);
	358
3729d502 PM	359	err = rtnl_configure_link(peer, ifmp);
	360	if (err < 0)
	361	goto err_configure_peer;
	362
e314dbdc PE	363	/*
	364	* register dev last
	365	*
	366	* note, that since we've registered new device the dev's name
	367	* should be re-allocated
	368	*/
	369
	370	if (tb[IFLA_ADDRESS] == NULL)
f2cedb63	371	eth_hw_addr_random(dev);
e314dbdc	372
6c8c4446 JP	373	if (tb[IFLA_IFNAME])
	374	nla_strlcpy(dev->name, tb[IFLA_IFNAME], IFNAMSIZ);
	375	else
	376	snprintf(dev->name, IFNAMSIZ, DRV_NAME "%%d");
	377
	378	if (strchr(dev->name, '%')) {
	379	err = dev_alloc_name(dev, dev->name);
	380	if (err < 0)
	381	goto err_alloc_name;
	382	}
	383
e314dbdc PE	384	err = register_netdevice(dev);
	385	if (err < 0)
	386	goto err_register_dev;
	387
	388	netif_carrier_off(dev);
	389
	390	/*
	391	* tie the deviced together
	392	*/
	393
	394	priv = netdev_priv(dev);
e314dbdc	395	priv->peer = peer;
e314dbdc PE	396
e314dbdc PE	397	priv = netdev_priv(peer);
e314dbdc	398	priv->peer = dev;
e314dbdc PE	399	return 0;
	400
	401	err_register_dev:
	402	/* nothing to do */
6c8c4446	403	err_alloc_name:
3729d502	404	err_configure_peer:
e314dbdc PE	405	unregister_netdevice(peer);
	406	return err;
	407
	408	err_register_peer:
	409	free_netdev(peer);
	410	return err;
	411	}
	412
23289a37	413	static void veth_dellink(struct net_device dev, struct list_head head)
e314dbdc PE	414	{
	415	struct veth_priv *priv;
	416	struct net_device *peer;
	417
	418	priv = netdev_priv(dev);
	419	peer = priv->peer;
	420
24540535 ED	421	unregister_netdevice_queue(dev, head);
24540535 ED	422	unregister_netdevice_queue(peer, head);
e314dbdc PE	423	}
e314dbdc PE	424
23711438 TG	425	static const struct nla_policy veth_policy[VETH_INFO_MAX + 1] = {
	426	[VETH_INFO_PEER] = { .len = sizeof(struct ifinfomsg) },
	427	};
e314dbdc PE	428
	429	static struct rtnl_link_ops veth_link_ops = {
	430	.kind = DRV_NAME,
	431	.priv_size = sizeof(struct veth_priv),
	432	.setup = veth_setup,
	433	.validate = veth_validate,
	434	.newlink = veth_newlink,
	435	.dellink = veth_dellink,
	436	.policy = veth_policy,
	437	.maxtype = VETH_INFO_MAX,
	438	};
	439
	440	/*
	441	* init/fini
	442	*/
	443
	444	static __init int veth_init(void)
	445	{
	446	return rtnl_link_register(&veth_link_ops);
	447	}
	448
	449	static __exit void veth_exit(void)
	450	{
68365458	451	rtnl_link_unregister(&veth_link_ops);
e314dbdc PE	452	}
	453
	454	module_init(veth_init);
	455	module_exit(veth_exit);
	456
	457	MODULE_DESCRIPTION("Virtual Ethernet Tunnel");
	458	MODULE_LICENSE("GPL v2");
	459	MODULE_ALIAS_RTNL_LINK(DRV_NAME);