[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 pcpu_vstats {
	u64			packets;
	u64			bytes;
	struct u64_stats_sync	syncp;
};

struct veth_priv {
	struct net_device __rcu	*peer;
	atomic64_t		dropped;
};

/*
 * 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 = netdev_priv(dev);
	struct net_device *peer = rtnl_dereference(priv->peer);

	data[0] = peer ? peer->ifindex : 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_strings		= veth_get_strings,
	.get_sset_count		= veth_get_sset_count,
	.get_ethtool_stats	= veth_get_ethtool_stats,
};

static netdev_tx_t veth_xmit(struct sk_buff *skb, struct net_device *dev)
{
	struct veth_priv *priv = netdev_priv(dev);
	struct net_device *rcv;
	int length = skb->len;

	rcu_read_lock();
	rcv = rcu_dereference(priv->peer);
	if (unlikely(!rcv)) {
		kfree_skb(skb);
		goto 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;

	if (likely(dev_forward_skb(rcv, skb) == NET_RX_SUCCESS)) {
		struct pcpu_vstats *stats = this_cpu_ptr(dev->vstats);

		u64_stats_update_begin(&stats->syncp);
		stats->bytes += length;
		stats->packets++;
		u64_stats_update_end(&stats->syncp);
	} else {
drop:
		atomic64_inc(&priv->dropped);
	}
	rcu_read_unlock();
	return NETDEV_TX_OK;
}

/*
 * general routines
 */

static u64 veth_stats_one(struct pcpu_vstats *result, struct net_device *dev)
{
	struct veth_priv *priv = netdev_priv(dev);
	int cpu;

	result->packets = 0;
	result->bytes = 0;
	for_each_possible_cpu(cpu) {
		struct pcpu_vstats *stats = per_cpu_ptr(dev->vstats, cpu);
		u64 packets, bytes;
		unsigned int start;

		do {
			start = u64_stats_fetch_begin_bh(&stats->syncp);
			packets = stats->packets;
			bytes = stats->bytes;
		} while (u64_stats_fetch_retry_bh(&stats->syncp, start));
		result->packets += packets;
		result->bytes += bytes;
	}
	return atomic64_read(&priv->dropped);
}

static struct rtnl_link_stats64 *veth_get_stats64(struct net_device *dev,
						  struct rtnl_link_stats64 *tot)
{
	struct veth_priv *priv = netdev_priv(dev);
	struct net_device *peer;
	struct pcpu_vstats one;

	tot->tx_dropped = veth_stats_one(&one, dev);
	tot->tx_bytes = one.bytes;
	tot->tx_packets = one.packets;

	rcu_read_lock();
	peer = rcu_dereference(priv->peer);
	if (peer) {
		tot->rx_dropped = veth_stats_one(&one, peer);
		tot->rx_bytes = one.bytes;
		tot->rx_packets = one.packets;
	}
	rcu_read_unlock();

	return tot;
}

/* fake multicast ability */
static void veth_set_multicast_list(struct net_device *dev)
{
}

static int veth_open(struct net_device *dev)
{
	struct veth_priv *priv = netdev_priv(dev);
	struct net_device *peer = rtnl_dereference(priv->peer);

	if (!peer)
		return -ENOTCONN;

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

static int veth_close(struct net_device *dev)
{
	struct veth_priv *priv = netdev_priv(dev);
	struct net_device *peer = rtnl_dereference(priv->peer);

	netif_carrier_off(dev);
	if (peer)
		netif_carrier_off(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)
{
	dev->vstats = netdev_alloc_pcpu_stats(struct pcpu_vstats);
	if (!dev->vstats)
		return -ENOMEM;
	return 0;
}

static void veth_dev_free(struct net_device *dev)
{
	free_percpu(dev->vstats);
	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_rx_mode     = veth_set_multicast_list,
	.ndo_set_mac_address = eth_mac_addr,
};

#define VETH_FEATURES (NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_ALL_TSO |    \
		       NETIF_F_HW_CSUM | NETIF_F_RXCSUM | NETIF_F_HIGHDMA | \
		       NETIF_F_GSO_GRE | NETIF_F_GSO_UDP_TUNNEL |	    \
		       NETIF_F_GSO_IPIP | NETIF_F_GSO_SIT | NETIF_F_UFO	|   \
		       NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX | \
		       NETIF_F_HW_VLAN_STAG_TX | NETIF_F_HW_VLAN_STAG_RX )

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

	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;

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

	dev->hw_features = VETH_FEATURES;
	dev->hw_enc_features = VETH_FEATURES;
}

/*
 * 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(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);

	if (ifmp && (dev->ifindex != 0))
		peer->ifindex = ifmp->ifi_index;

	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");

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

	netif_carrier_off(dev);

	/*
	 * tie the deviced together
	 */

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

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

err_register_dev:
	/* nothing to do */
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 = rtnl_dereference(priv->peer);

	/* Note : dellink() is called from default_device_exit_batch(),
	 * before a rcu_synchronize() point. The devices are guaranteed
	 * not being freed before one RCU grace period.
	 */
	RCU_INIT_POINTER(priv->peer, NULL);
	unregister_netdevice_queue(dev, head);

	if (peer) {
		priv = netdev_priv(peer);
		RCU_INIT_POINTER(priv->peer, NULL);
		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 <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
2681128f ED	28	struct pcpu_vstats {
	29	u64 packets;
	30	u64 bytes;
cf05c700	31	struct u64_stats_sync syncp;
e314dbdc PE	32	};
	33
	34	struct veth_priv {
d0e2c55e	35	struct net_device __rcu *peer;
2681128f	36	atomic64_t dropped;
e314dbdc PE	37	};
e314dbdc PE	38
e314dbdc PE	39	/*
	40	* ethtool interface
	41	*/
	42
	43	static struct {
	44	const char string[ETH_GSTRING_LEN];
	45	} ethtool_stats_keys[] = {
	46	{ "peer_ifindex" },
	47	};
	48
	49	static int veth_get_settings(struct net_device dev, struct ethtool_cmd cmd)
	50	{
	51	cmd->supported = 0;
	52	cmd->advertising = 0;
70739497	53	ethtool_cmd_speed_set(cmd, SPEED_10000);
e314dbdc PE	54	cmd->duplex = DUPLEX_FULL;
	55	cmd->port = PORT_TP;
	56	cmd->phy_address = 0;
	57	cmd->transceiver = XCVR_INTERNAL;
	58	cmd->autoneg = AUTONEG_DISABLE;
	59	cmd->maxtxpkt = 0;
	60	cmd->maxrxpkt = 0;
	61	return 0;
	62	}
	63
	64	static void veth_get_drvinfo(struct net_device dev, struct ethtool_drvinfo info)
	65	{
33a5ba14 RJ	66	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
33a5ba14 RJ	67	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
e314dbdc PE	68	}
	69
	70	static void veth_get_strings(struct net_device dev, u32 stringset, u8 buf)
	71	{
	72	switch(stringset) {
	73	case ETH_SS_STATS:
	74	memcpy(buf, &ethtool_stats_keys, sizeof(ethtool_stats_keys));
	75	break;
	76	}
	77	}
	78
b9f2c044	79	static int veth_get_sset_count(struct net_device *dev, int sset)
e314dbdc	80	{
b9f2c044 JG	81	switch (sset) {
	82	case ETH_SS_STATS:
	83	return ARRAY_SIZE(ethtool_stats_keys);
	84	default:
	85	return -EOPNOTSUPP;
	86	}
e314dbdc PE	87	}
	88
	89	static void veth_get_ethtool_stats(struct net_device *dev,
	90	struct ethtool_stats stats, u64 data)
	91	{
d0e2c55e ED	92	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e ED	93	struct net_device *peer = rtnl_dereference(priv->peer);
e314dbdc	94
d0e2c55e	95	data[0] = peer ? peer->ifindex : 0;
e314dbdc PE	96	}
e314dbdc PE	97
0fc0b732	98	static const struct ethtool_ops veth_ethtool_ops = {
e314dbdc PE	99	.get_settings = veth_get_settings,
	100	.get_drvinfo = veth_get_drvinfo,
	101	.get_link = ethtool_op_get_link,
e314dbdc	102	.get_strings = veth_get_strings,
b9f2c044	103	.get_sset_count = veth_get_sset_count,
e314dbdc PE	104	.get_ethtool_stats = veth_get_ethtool_stats,
	105	};
	106
424efe9c	107	static netdev_tx_t veth_xmit(struct sk_buff skb, struct net_device dev)
e314dbdc	108	{
2681128f	109	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e	110	struct net_device *rcv;
2681128f	111	int length = skb->len;
e314dbdc	112
d0e2c55e ED	113	rcu_read_lock();
	114	rcv = rcu_dereference(priv->peer);
	115	if (unlikely(!rcv)) {
	116	kfree_skb(skb);
	117	goto drop;
	118	}
0b796750	119	/* don't change ip_summed == CHECKSUM_PARTIAL, as that
2681128f ED	120	* will cause bad checksum on forwarded packets
2681128f ED	121	*/
a2c725fa MM	122	if (skb->ip_summed == CHECKSUM_NONE &&
	123	rcv->features & NETIF_F_RXCSUM)
	124	skb->ip_summed = CHECKSUM_UNNECESSARY;
e314dbdc	125
2681128f ED	126	if (likely(dev_forward_skb(rcv, skb) == NET_RX_SUCCESS)) {
2681128f ED	127	struct pcpu_vstats *stats = this_cpu_ptr(dev->vstats);
e314dbdc	128
2681128f ED	129	u64_stats_update_begin(&stats->syncp);
	130	stats->bytes += length;
	131	stats->packets++;
	132	u64_stats_update_end(&stats->syncp);
	133	} else {
d0e2c55e	134	drop:
2681128f ED	135	atomic64_inc(&priv->dropped);
2681128f ED	136	}
d0e2c55e	137	rcu_read_unlock();
6ed10654	138	return NETDEV_TX_OK;
e314dbdc PE	139	}
	140
	141	/*
	142	* general routines
	143	*/
	144
2681128f	145	static u64 veth_stats_one(struct pcpu_vstats result, struct net_device dev)
e314dbdc	146	{
cf05c700	147	struct veth_priv *priv = netdev_priv(dev);
11687a10	148	int cpu;
e314dbdc	149
2681128f ED	150	result->packets = 0;
2681128f ED	151	result->bytes = 0;
2b1c8b0f	152	for_each_possible_cpu(cpu) {
2681128f ED	153	struct pcpu_vstats *stats = per_cpu_ptr(dev->vstats, cpu);
2681128f ED	154	u64 packets, bytes;
cf05c700 ED	155	unsigned int start;
	156
	157	do {
	158	start = u64_stats_fetch_begin_bh(&stats->syncp);
2681128f ED	159	packets = stats->packets;
2681128f ED	160	bytes = stats->bytes;
cf05c700	161	} while (u64_stats_fetch_retry_bh(&stats->syncp, start));
2681128f ED	162	result->packets += packets;
2681128f ED	163	result->bytes += bytes;
11687a10	164	}
2681128f ED	165	return atomic64_read(&priv->dropped);
	166	}
	167
	168	static struct rtnl_link_stats64 veth_get_stats64(struct net_device dev,
	169	struct rtnl_link_stats64 *tot)
	170	{
	171	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e	172	struct net_device *peer;
2681128f ED	173	struct pcpu_vstats one;
	174
	175	tot->tx_dropped = veth_stats_one(&one, dev);
	176	tot->tx_bytes = one.bytes;
	177	tot->tx_packets = one.packets;
	178
d0e2c55e ED	179	rcu_read_lock();
	180	peer = rcu_dereference(priv->peer);
	181	if (peer) {
	182	tot->rx_dropped = veth_stats_one(&one, peer);
	183	tot->rx_bytes = one.bytes;
	184	tot->rx_packets = one.packets;
	185	}
	186	rcu_read_unlock();
6311cc44	187
6311cc44	188	return tot;
e314dbdc PE	189	}
e314dbdc PE	190
5c70ef85 G	191	/* fake multicast ability */
	192	static void veth_set_multicast_list(struct net_device *dev)
	193	{
	194	}
	195
e314dbdc PE	196	static int veth_open(struct net_device *dev)
e314dbdc PE	197	{
d0e2c55e ED	198	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e ED	199	struct net_device *peer = rtnl_dereference(priv->peer);
e314dbdc	200
d0e2c55e	201	if (!peer)
e314dbdc PE	202	return -ENOTCONN;
e314dbdc PE	203
d0e2c55e	204	if (peer->flags & IFF_UP) {
e314dbdc	205	netif_carrier_on(dev);
d0e2c55e	206	netif_carrier_on(peer);
e314dbdc PE	207	}
	208	return 0;
	209	}
	210
2cf48a10 EB	211	static int veth_close(struct net_device *dev)
	212	{
	213	struct veth_priv *priv = netdev_priv(dev);
2efd32ee	214	struct net_device *peer = rtnl_dereference(priv->peer);
2cf48a10 EB	215
2cf48a10 EB	216	netif_carrier_off(dev);
2efd32ee ED	217	if (peer)
2efd32ee ED	218	netif_carrier_off(peer);
2cf48a10 EB	219
	220	return 0;
	221	}
	222
38d40815 EB	223	static int is_valid_veth_mtu(int new_mtu)
38d40815 EB	224	{
807540ba	225	return new_mtu >= MIN_MTU && new_mtu <= MAX_MTU;
38d40815 EB	226	}
	227
	228	static int veth_change_mtu(struct net_device *dev, int new_mtu)
	229	{
	230	if (!is_valid_veth_mtu(new_mtu))
	231	return -EINVAL;
	232	dev->mtu = new_mtu;
	233	return 0;
	234	}
	235
e314dbdc PE	236	static int veth_dev_init(struct net_device *dev)
e314dbdc PE	237	{
1c213bd2	238	dev->vstats = netdev_alloc_pcpu_stats(struct pcpu_vstats);
2681128f	239	if (!dev->vstats)
e314dbdc	240	return -ENOMEM;
e314dbdc PE	241	return 0;
	242	}
	243
11687a10 DM	244	static void veth_dev_free(struct net_device *dev)
11687a10 DM	245	{
2681128f	246	free_percpu(dev->vstats);
11687a10 DM	247	free_netdev(dev);
	248	}
	249
4456e7bd	250	static const struct net_device_ops veth_netdev_ops = {
ee923623 DL	251	.ndo_init = veth_dev_init,
ee923623 DL	252	.ndo_open = veth_open,
2cf48a10	253	.ndo_stop = veth_close,
ee923623	254	.ndo_start_xmit = veth_xmit,
38d40815	255	.ndo_change_mtu = veth_change_mtu,
6311cc44	256	.ndo_get_stats64 = veth_get_stats64,
5c70ef85	257	.ndo_set_rx_mode = veth_set_multicast_list,
ee923623	258	.ndo_set_mac_address = eth_mac_addr,
4456e7bd SH	259	};
4456e7bd SH	260
8093315a ED	261	#define VETH_FEATURES (NETIF_F_SG \| NETIF_F_FRAGLIST \| NETIF_F_ALL_TSO \| \
8093315a ED	262	NETIF_F_HW_CSUM \| NETIF_F_RXCSUM \| NETIF_F_HIGHDMA \| \
82d81898 ED	263	NETIF_F_GSO_GRE \| NETIF_F_GSO_UDP_TUNNEL \| \
82d81898 ED	264	NETIF_F_GSO_IPIP \| NETIF_F_GSO_SIT \| NETIF_F_UFO \| \
28d2b136 PM	265	NETIF_F_HW_VLAN_CTAG_TX \| NETIF_F_HW_VLAN_CTAG_RX \| \
28d2b136 PM	266	NETIF_F_HW_VLAN_STAG_TX \| NETIF_F_HW_VLAN_STAG_RX )
8093315a	267
e314dbdc PE	268	static void veth_setup(struct net_device *dev)
	269	{
	270	ether_setup(dev);
	271
550fd08c	272	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
23ea5a96	273	dev->priv_flags \|= IFF_LIVE_ADDR_CHANGE;
550fd08c	274
4456e7bd	275	dev->netdev_ops = &veth_netdev_ops;
e314dbdc PE	276	dev->ethtool_ops = &veth_ethtool_ops;
e314dbdc PE	277	dev->features \|= NETIF_F_LLTX;
8093315a	278	dev->features \|= VETH_FEATURES;
b69bbddf	279	dev->vlan_features = dev->features;
11687a10	280	dev->destructor = veth_dev_free;
a2c725fa	281
8093315a	282	dev->hw_features = VETH_FEATURES;
82d81898	283	dev->hw_enc_features = VETH_FEATURES;
e314dbdc PE	284	}
	285
	286	/*
	287	* netlink interface
	288	*/
	289
	290	static int veth_validate(struct nlattr tb[], struct nlattr data[])
	291	{
	292	if (tb[IFLA_ADDRESS]) {
	293	if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN)
	294	return -EINVAL;
	295	if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS])))
	296	return -EADDRNOTAVAIL;
	297	}
38d40815 EB	298	if (tb[IFLA_MTU]) {
	299	if (!is_valid_veth_mtu(nla_get_u32(tb[IFLA_MTU])))
	300	return -EINVAL;
	301	}
e314dbdc PE	302	return 0;
	303	}
	304
	305	static struct rtnl_link_ops veth_link_ops;
	306
81adee47	307	static int veth_newlink(struct net src_net, struct net_device dev,
e314dbdc PE	308	struct nlattr tb[], struct nlattr data[])
	309	{
	310	int err;
	311	struct net_device *peer;
	312	struct veth_priv *priv;
	313	char ifname[IFNAMSIZ];
	314	struct nlattr peer_tb[IFLA_MAX + 1], *tbp;
3729d502	315	struct ifinfomsg *ifmp;
81adee47	316	struct net *net;
e314dbdc PE	317
	318	/*
	319	* create and register peer first
e314dbdc	320	*/
e314dbdc PE	321	if (data != NULL && data[VETH_INFO_PEER] != NULL) {
	322	struct nlattr *nla_peer;
	323
	324	nla_peer = data[VETH_INFO_PEER];
3729d502	325	ifmp = nla_data(nla_peer);
e314dbdc PE	326	err = nla_parse(peer_tb, IFLA_MAX,
	327	nla_data(nla_peer) + sizeof(struct ifinfomsg),
	328	nla_len(nla_peer) - sizeof(struct ifinfomsg),
	329	ifla_policy);
	330	if (err < 0)
	331	return err;
	332
	333	err = veth_validate(peer_tb, NULL);
	334	if (err < 0)
	335	return err;
	336
	337	tbp = peer_tb;
3729d502 PM	338	} else {
3729d502 PM	339	ifmp = NULL;
e314dbdc	340	tbp = tb;
3729d502	341	}
e314dbdc PE	342
	343	if (tbp[IFLA_IFNAME])
	344	nla_strlcpy(ifname, tbp[IFLA_IFNAME], IFNAMSIZ);
	345	else
	346	snprintf(ifname, IFNAMSIZ, DRV_NAME "%%d");
	347
81adee47 EB	348	net = rtnl_link_get_net(src_net, tbp);
	349	if (IS_ERR(net))
	350	return PTR_ERR(net);
	351
c0713563	352	peer = rtnl_create_link(net, ifname, &veth_link_ops, tbp);
81adee47 EB	353	if (IS_ERR(peer)) {
81adee47 EB	354	put_net(net);
e314dbdc	355	return PTR_ERR(peer);
81adee47	356	}
e314dbdc PE	357
e314dbdc PE	358	if (tbp[IFLA_ADDRESS] == NULL)
f2cedb63	359	eth_hw_addr_random(peer);
e6f8f1a7 PE	360
	361	if (ifmp && (dev->ifindex != 0))
	362	peer->ifindex = ifmp->ifi_index;
e314dbdc PE	363
e314dbdc PE	364	err = register_netdevice(peer);
81adee47 EB	365	put_net(net);
81adee47 EB	366	net = NULL;
e314dbdc PE	367	if (err < 0)
	368	goto err_register_peer;
	369
	370	netif_carrier_off(peer);
	371
3729d502 PM	372	err = rtnl_configure_link(peer, ifmp);
	373	if (err < 0)
	374	goto err_configure_peer;
	375
e314dbdc PE	376	/*
	377	* register dev last
	378	*
	379	* note, that since we've registered new device the dev's name
	380	* should be re-allocated
	381	*/
	382
	383	if (tb[IFLA_ADDRESS] == NULL)
f2cedb63	384	eth_hw_addr_random(dev);
e314dbdc	385
6c8c4446 JP	386	if (tb[IFLA_IFNAME])
	387	nla_strlcpy(dev->name, tb[IFLA_IFNAME], IFNAMSIZ);
	388	else
	389	snprintf(dev->name, IFNAMSIZ, DRV_NAME "%%d");
	390
e314dbdc PE	391	err = register_netdevice(dev);
	392	if (err < 0)
	393	goto err_register_dev;
	394
	395	netif_carrier_off(dev);
	396
	397	/*
	398	* tie the deviced together
	399	*/
	400
	401	priv = netdev_priv(dev);
d0e2c55e	402	rcu_assign_pointer(priv->peer, peer);
e314dbdc PE	403
e314dbdc PE	404	priv = netdev_priv(peer);
d0e2c55e	405	rcu_assign_pointer(priv->peer, dev);
e314dbdc PE	406	return 0;
	407
	408	err_register_dev:
	409	/* nothing to do */
3729d502	410	err_configure_peer:
e314dbdc PE	411	unregister_netdevice(peer);
	412	return err;
	413
	414	err_register_peer:
	415	free_netdev(peer);
	416	return err;
	417	}
	418
23289a37	419	static void veth_dellink(struct net_device dev, struct list_head head)
e314dbdc PE	420	{
	421	struct veth_priv *priv;
	422	struct net_device *peer;
	423
	424	priv = netdev_priv(dev);
d0e2c55e ED	425	peer = rtnl_dereference(priv->peer);
	426
	427	/* Note : dellink() is called from default_device_exit_batch(),
	428	* before a rcu_synchronize() point. The devices are guaranteed
	429	* not being freed before one RCU grace period.
	430	*/
	431	RCU_INIT_POINTER(priv->peer, NULL);
24540535	432	unregister_netdevice_queue(dev, head);
f45a5c26 ED	433
	434	if (peer) {
	435	priv = netdev_priv(peer);
	436	RCU_INIT_POINTER(priv->peer, NULL);
	437	unregister_netdevice_queue(peer, head);
	438	}
e314dbdc PE	439	}
e314dbdc PE	440
23711438 TG	441	static const struct nla_policy veth_policy[VETH_INFO_MAX + 1] = {
	442	[VETH_INFO_PEER] = { .len = sizeof(struct ifinfomsg) },
	443	};
e314dbdc PE	444
	445	static struct rtnl_link_ops veth_link_ops = {
	446	.kind = DRV_NAME,
	447	.priv_size = sizeof(struct veth_priv),
	448	.setup = veth_setup,
	449	.validate = veth_validate,
	450	.newlink = veth_newlink,
	451	.dellink = veth_dellink,
	452	.policy = veth_policy,
	453	.maxtype = VETH_INFO_MAX,
	454	};
	455
	456	/*
	457	* init/fini
	458	*/
	459
	460	static __init int veth_init(void)
	461	{
	462	return rtnl_link_register(&veth_link_ops);
	463	}
	464
	465	static __exit void veth_exit(void)
	466	{
68365458	467	rtnl_link_unregister(&veth_link_ops);
e314dbdc PE	468	}
	469
	470	module_init(veth_init);
	471	module_exit(veth_exit);
	472
	473	MODULE_DESCRIPTION("Virtual Ethernet Tunnel");
	474	MODULE_LICENSE("GPL v2");
	475	MODULE_ALIAS_RTNL_LINK(DRV_NAME);