[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/rtnetlink.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;
	unsigned		requested_headroom;
};

/*
 * 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;
	}

	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_irq(&stats->syncp);
			packets = stats->packets;
			bytes = stats->bytes;
		} while (u64_stats_fetch_retry_irq(&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);
}

#ifdef CONFIG_NET_POLL_CONTROLLER
static void veth_poll_controller(struct net_device *dev)
{
	/* veth only receives frames when its peer sends one
	 * Since it's a synchronous operation, we are guaranteed
	 * never to have pending data when we poll for it so
	 * there is nothing to do here.
	 *
	 * We need this though so netpoll recognizes us as an interface that
	 * supports polling, which enables bridge devices in virt setups to
	 * still use netconsole
	 */
}
#endif	/* CONFIG_NET_POLL_CONTROLLER */

static int veth_get_iflink(const struct net_device *dev)
{
	struct veth_priv *priv = netdev_priv(dev);
	struct net_device *peer;
	int iflink;

	rcu_read_lock();
	peer = rcu_dereference(priv->peer);
	iflink = peer ? peer->ifindex : 0;
	rcu_read_unlock();

	return iflink;
}

static void veth_set_rx_headroom(struct net_device *dev, int new_hr)
{
	struct veth_priv *peer_priv, *priv = netdev_priv(dev);
	struct net_device *peer;

	if (new_hr < 0)
		new_hr = 0;

	rcu_read_lock();
	peer = rcu_dereference(priv->peer);
	if (unlikely(!peer))
		goto out;

	peer_priv = netdev_priv(peer);
	priv->requested_headroom = new_hr;
	new_hr = max(priv->requested_headroom, peer_priv->requested_headroom);
	dev->needed_headroom = new_hr;
	peer->needed_headroom = new_hr;

out:
	rcu_read_unlock();
}

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,
#ifdef CONFIG_NET_POLL_CONTROLLER
	.ndo_poll_controller	= veth_poll_controller,
#endif
	.ndo_get_iflink		= veth_get_iflink,
	.ndo_features_check	= passthru_features_check,
	.ndo_set_rx_headroom	= veth_set_rx_headroom,
};

#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->priv_flags |= IFF_NO_QUEUE;
	dev->priv_flags |= IFF_PHONY_HEADROOM;

	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 &
			     ~(NETIF_F_HW_VLAN_CTAG_TX |
			       NETIF_F_HW_VLAN_STAG_TX |
			       NETIF_F_HW_VLAN_CTAG_RX |
			       NETIF_F_HW_VLAN_STAG_RX);
	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;
	unsigned char name_assign_type;
	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 = rtnl_nla_parse_ifla(peer_tb,
					  nla_data(nla_peer) + sizeof(struct ifinfomsg),
					  nla_len(nla_peer) - sizeof(struct ifinfomsg));
		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);
		name_assign_type = NET_NAME_USER;
	} else {
		snprintf(ifname, IFNAMSIZ, DRV_NAME "%%d");
		name_assign_type = NET_NAME_ENUM;
	}

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

	peer = rtnl_create_link(net, ifname, name_assign_type,
				&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 net *veth_get_link_net(const struct net_device *dev)
{
	struct veth_priv *priv = netdev_priv(dev);
	struct net_device *peer = rtnl_dereference(priv->peer);

	return peer ? dev_net(peer) : dev_net(dev);
}

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,
	.get_link_net	= veth_get_link_net,
};

/*
 * 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	16
f7b12606	17	#include <net/rtnetlink.h>
e314dbdc PE	18	#include <net/dst.h>
e314dbdc PE	19	#include <net/xfrm.h>
ecef969e	20	#include <linux/veth.h>
9d9779e7	21	#include <linux/module.h>
e314dbdc PE	22
	23	#define DRV_NAME "veth"
	24	#define DRV_VERSION "1.0"
	25
38d40815 EB	26	#define MIN_MTU 68 /* Min L3 MTU */
38d40815 EB	27	#define MAX_MTU 65535 /* Max L3 MTU (arbitrary) */
38d40815	28
2681128f ED	29	struct pcpu_vstats {
	30	u64 packets;
	31	u64 bytes;
cf05c700	32	struct u64_stats_sync syncp;
e314dbdc PE	33	};
	34
	35	struct veth_priv {
d0e2c55e	36	struct net_device __rcu *peer;
2681128f	37	atomic64_t dropped;
163e5292	38	unsigned requested_headroom;
e314dbdc PE	39	};
e314dbdc PE	40
e314dbdc PE	41	/*
	42	* ethtool interface
	43	*/
	44
	45	static struct {
	46	const char string[ETH_GSTRING_LEN];
	47	} ethtool_stats_keys[] = {
	48	{ "peer_ifindex" },
	49	};
	50
	51	static int veth_get_settings(struct net_device dev, struct ethtool_cmd cmd)
	52	{
	53	cmd->supported = 0;
	54	cmd->advertising = 0;
70739497	55	ethtool_cmd_speed_set(cmd, SPEED_10000);
e314dbdc PE	56	cmd->duplex = DUPLEX_FULL;
	57	cmd->port = PORT_TP;
	58	cmd->phy_address = 0;
	59	cmd->transceiver = XCVR_INTERNAL;
	60	cmd->autoneg = AUTONEG_DISABLE;
	61	cmd->maxtxpkt = 0;
	62	cmd->maxrxpkt = 0;
	63	return 0;
	64	}
	65
	66	static void veth_get_drvinfo(struct net_device dev, struct ethtool_drvinfo info)
	67	{
33a5ba14 RJ	68	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
33a5ba14 RJ	69	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
e314dbdc PE	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	{
d0e2c55e ED	94	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e ED	95	struct net_device *peer = rtnl_dereference(priv->peer);
e314dbdc	96
d0e2c55e	97	data[0] = peer ? peer->ifindex : 0;
e314dbdc PE	98	}
e314dbdc PE	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
424efe9c	109	static netdev_tx_t veth_xmit(struct sk_buff skb, struct net_device dev)
e314dbdc	110	{
2681128f	111	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e	112	struct net_device *rcv;
2681128f	113	int length = skb->len;
e314dbdc	114
d0e2c55e ED	115	rcu_read_lock();
	116	rcv = rcu_dereference(priv->peer);
	117	if (unlikely(!rcv)) {
	118	kfree_skb(skb);
	119	goto drop;
	120	}
e314dbdc	121
2681128f ED	122	if (likely(dev_forward_skb(rcv, skb) == NET_RX_SUCCESS)) {
2681128f ED	123	struct pcpu_vstats *stats = this_cpu_ptr(dev->vstats);
e314dbdc	124
2681128f ED	125	u64_stats_update_begin(&stats->syncp);
	126	stats->bytes += length;
	127	stats->packets++;
	128	u64_stats_update_end(&stats->syncp);
	129	} else {
d0e2c55e	130	drop:
2681128f ED	131	atomic64_inc(&priv->dropped);
2681128f ED	132	}
d0e2c55e	133	rcu_read_unlock();
6ed10654	134	return NETDEV_TX_OK;
e314dbdc PE	135	}
	136
	137	/*
	138	* general routines
	139	*/
	140
2681128f	141	static u64 veth_stats_one(struct pcpu_vstats result, struct net_device dev)
e314dbdc	142	{
cf05c700	143	struct veth_priv *priv = netdev_priv(dev);
11687a10	144	int cpu;
e314dbdc	145
2681128f ED	146	result->packets = 0;
2681128f ED	147	result->bytes = 0;
2b1c8b0f	148	for_each_possible_cpu(cpu) {
2681128f ED	149	struct pcpu_vstats *stats = per_cpu_ptr(dev->vstats, cpu);
2681128f ED	150	u64 packets, bytes;
cf05c700 ED	151	unsigned int start;
	152
	153	do {
57a7744e	154	start = u64_stats_fetch_begin_irq(&stats->syncp);
2681128f ED	155	packets = stats->packets;
2681128f ED	156	bytes = stats->bytes;
57a7744e	157	} while (u64_stats_fetch_retry_irq(&stats->syncp, start));
2681128f ED	158	result->packets += packets;
2681128f ED	159	result->bytes += bytes;
11687a10	160	}
2681128f ED	161	return atomic64_read(&priv->dropped);
	162	}
	163
	164	static struct rtnl_link_stats64 veth_get_stats64(struct net_device dev,
	165	struct rtnl_link_stats64 *tot)
	166	{
	167	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e	168	struct net_device *peer;
2681128f ED	169	struct pcpu_vstats one;
	170
	171	tot->tx_dropped = veth_stats_one(&one, dev);
	172	tot->tx_bytes = one.bytes;
	173	tot->tx_packets = one.packets;
	174
d0e2c55e ED	175	rcu_read_lock();
	176	peer = rcu_dereference(priv->peer);
	177	if (peer) {
	178	tot->rx_dropped = veth_stats_one(&one, peer);
	179	tot->rx_bytes = one.bytes;
	180	tot->rx_packets = one.packets;
	181	}
	182	rcu_read_unlock();
6311cc44	183
6311cc44	184	return tot;
e314dbdc PE	185	}
e314dbdc PE	186
5c70ef85 G	187	/* fake multicast ability */
	188	static void veth_set_multicast_list(struct net_device *dev)
	189	{
	190	}
	191
e314dbdc PE	192	static int veth_open(struct net_device *dev)
e314dbdc PE	193	{
d0e2c55e ED	194	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e ED	195	struct net_device *peer = rtnl_dereference(priv->peer);
e314dbdc	196
d0e2c55e	197	if (!peer)
e314dbdc PE	198	return -ENOTCONN;
e314dbdc PE	199
d0e2c55e	200	if (peer->flags & IFF_UP) {
e314dbdc	201	netif_carrier_on(dev);
d0e2c55e	202	netif_carrier_on(peer);
e314dbdc PE	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);
2efd32ee	210	struct net_device *peer = rtnl_dereference(priv->peer);
2cf48a10 EB	211
2cf48a10 EB	212	netif_carrier_off(dev);
2efd32ee ED	213	if (peer)
2efd32ee ED	214	netif_carrier_off(peer);
2cf48a10 EB	215
	216	return 0;
	217	}
	218
38d40815 EB	219	static int is_valid_veth_mtu(int new_mtu)
38d40815 EB	220	{
807540ba	221	return new_mtu >= MIN_MTU && new_mtu <= MAX_MTU;
38d40815 EB	222	}
	223
	224	static int veth_change_mtu(struct net_device *dev, int new_mtu)
	225	{
	226	if (!is_valid_veth_mtu(new_mtu))
	227	return -EINVAL;
	228	dev->mtu = new_mtu;
	229	return 0;
	230	}
	231
e314dbdc PE	232	static int veth_dev_init(struct net_device *dev)
e314dbdc PE	233	{
1c213bd2	234	dev->vstats = netdev_alloc_pcpu_stats(struct pcpu_vstats);
2681128f	235	if (!dev->vstats)
e314dbdc	236	return -ENOMEM;
e314dbdc PE	237	return 0;
	238	}
	239
11687a10 DM	240	static void veth_dev_free(struct net_device *dev)
11687a10 DM	241	{
2681128f	242	free_percpu(dev->vstats);
11687a10 DM	243	free_netdev(dev);
	244	}
	245
bb446c19 WC	246	#ifdef CONFIG_NET_POLL_CONTROLLER
	247	static void veth_poll_controller(struct net_device *dev)
	248	{
	249	/* veth only receives frames when its peer sends one
	250	* Since it's a synchronous operation, we are guaranteed
	251	* never to have pending data when we poll for it so
	252	* there is nothing to do here.
	253	*
	254	* We need this though so netpoll recognizes us as an interface that
	255	* supports polling, which enables bridge devices in virt setups to
	256	* still use netconsole
	257	*/
	258	}
	259	#endif /* CONFIG_NET_POLL_CONTROLLER */
	260
a45253bf ND	261	static int veth_get_iflink(const struct net_device *dev)
	262	{
	263	struct veth_priv *priv = netdev_priv(dev);
	264	struct net_device *peer;
	265	int iflink;
	266
	267	rcu_read_lock();
	268	peer = rcu_dereference(priv->peer);
	269	iflink = peer ? peer->ifindex : 0;
	270	rcu_read_unlock();
	271
	272	return iflink;
	273	}
	274
163e5292 PA	275	static void veth_set_rx_headroom(struct net_device *dev, int new_hr)
	276	{
	277	struct veth_priv peer_priv, priv = netdev_priv(dev);
	278	struct net_device *peer;
	279
	280	if (new_hr < 0)
	281	new_hr = 0;
	282
	283	rcu_read_lock();
	284	peer = rcu_dereference(priv->peer);
	285	if (unlikely(!peer))
	286	goto out;
	287
	288	peer_priv = netdev_priv(peer);
	289	priv->requested_headroom = new_hr;
	290	new_hr = max(priv->requested_headroom, peer_priv->requested_headroom);
	291	dev->needed_headroom = new_hr;
	292	peer->needed_headroom = new_hr;
	293
	294	out:
	295	rcu_read_unlock();
	296	}
	297
4456e7bd	298	static const struct net_device_ops veth_netdev_ops = {
ee923623 DL	299	.ndo_init = veth_dev_init,
ee923623 DL	300	.ndo_open = veth_open,
2cf48a10	301	.ndo_stop = veth_close,
ee923623	302	.ndo_start_xmit = veth_xmit,
38d40815	303	.ndo_change_mtu = veth_change_mtu,
6311cc44	304	.ndo_get_stats64 = veth_get_stats64,
5c70ef85	305	.ndo_set_rx_mode = veth_set_multicast_list,
ee923623	306	.ndo_set_mac_address = eth_mac_addr,
bb446c19 WC	307	#ifdef CONFIG_NET_POLL_CONTROLLER
	308	.ndo_poll_controller = veth_poll_controller,
	309	#endif
a45253bf	310	.ndo_get_iflink = veth_get_iflink,
1a04a821	311	.ndo_features_check = passthru_features_check,
163e5292	312	.ndo_set_rx_headroom = veth_set_rx_headroom,
4456e7bd SH	313	};
4456e7bd SH	314
8093315a ED	315	#define VETH_FEATURES (NETIF_F_SG \| NETIF_F_FRAGLIST \| NETIF_F_ALL_TSO \| \
8093315a ED	316	NETIF_F_HW_CSUM \| NETIF_F_RXCSUM \| NETIF_F_HIGHDMA \| \
82d81898 ED	317	NETIF_F_GSO_GRE \| NETIF_F_GSO_UDP_TUNNEL \| \
82d81898 ED	318	NETIF_F_GSO_IPIP \| NETIF_F_GSO_SIT \| NETIF_F_UFO \| \
28d2b136 PM	319	NETIF_F_HW_VLAN_CTAG_TX \| NETIF_F_HW_VLAN_CTAG_RX \| \
28d2b136 PM	320	NETIF_F_HW_VLAN_STAG_TX \| NETIF_F_HW_VLAN_STAG_RX )
8093315a	321
e314dbdc PE	322	static void veth_setup(struct net_device *dev)
	323	{
	324	ether_setup(dev);
	325
550fd08c	326	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
23ea5a96	327	dev->priv_flags \|= IFF_LIVE_ADDR_CHANGE;
02f01ec1	328	dev->priv_flags \|= IFF_NO_QUEUE;
163e5292	329	dev->priv_flags \|= IFF_PHONY_HEADROOM;
550fd08c	330
4456e7bd	331	dev->netdev_ops = &veth_netdev_ops;
e314dbdc PE	332	dev->ethtool_ops = &veth_ethtool_ops;
e314dbdc PE	333	dev->features \|= NETIF_F_LLTX;
8093315a	334	dev->features \|= VETH_FEATURES;
8d0d21f4	335	dev->vlan_features = dev->features &
3f8c707b VY	336	~(NETIF_F_HW_VLAN_CTAG_TX \|
	337	NETIF_F_HW_VLAN_STAG_TX \|
	338	NETIF_F_HW_VLAN_CTAG_RX \|
	339	NETIF_F_HW_VLAN_STAG_RX);
11687a10	340	dev->destructor = veth_dev_free;
a2c725fa	341
8093315a	342	dev->hw_features = VETH_FEATURES;
82d81898	343	dev->hw_enc_features = VETH_FEATURES;
e314dbdc PE	344	}
	345
	346	/*
	347	* netlink interface
	348	*/
	349
	350	static int veth_validate(struct nlattr tb[], struct nlattr data[])
	351	{
	352	if (tb[IFLA_ADDRESS]) {
	353	if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN)
	354	return -EINVAL;
	355	if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS])))
	356	return -EADDRNOTAVAIL;
	357	}
38d40815 EB	358	if (tb[IFLA_MTU]) {
	359	if (!is_valid_veth_mtu(nla_get_u32(tb[IFLA_MTU])))
	360	return -EINVAL;
	361	}
e314dbdc PE	362	return 0;
	363	}
	364
	365	static struct rtnl_link_ops veth_link_ops;
	366
81adee47	367	static int veth_newlink(struct net src_net, struct net_device dev,
e314dbdc PE	368	struct nlattr tb[], struct nlattr data[])
	369	{
	370	int err;
	371	struct net_device *peer;
	372	struct veth_priv *priv;
	373	char ifname[IFNAMSIZ];
	374	struct nlattr peer_tb[IFLA_MAX + 1], *tbp;
5517750f	375	unsigned char name_assign_type;
3729d502	376	struct ifinfomsg *ifmp;
81adee47	377	struct net *net;
e314dbdc PE	378
	379	/*
	380	* create and register peer first
e314dbdc	381	*/
e314dbdc PE	382	if (data != NULL && data[VETH_INFO_PEER] != NULL) {
	383	struct nlattr *nla_peer;
	384
	385	nla_peer = data[VETH_INFO_PEER];
3729d502	386	ifmp = nla_data(nla_peer);
f7b12606 JP	387	err = rtnl_nla_parse_ifla(peer_tb,
	388	nla_data(nla_peer) + sizeof(struct ifinfomsg),
	389	nla_len(nla_peer) - sizeof(struct ifinfomsg));
e314dbdc PE	390	if (err < 0)
	391	return err;
	392
	393	err = veth_validate(peer_tb, NULL);
	394	if (err < 0)
	395	return err;
	396
	397	tbp = peer_tb;
3729d502 PM	398	} else {
3729d502 PM	399	ifmp = NULL;
e314dbdc	400	tbp = tb;
3729d502	401	}
e314dbdc	402
5517750f	403	if (tbp[IFLA_IFNAME]) {
e314dbdc	404	nla_strlcpy(ifname, tbp[IFLA_IFNAME], IFNAMSIZ);
5517750f TG	405	name_assign_type = NET_NAME_USER;
5517750f TG	406	} else {
e314dbdc	407	snprintf(ifname, IFNAMSIZ, DRV_NAME "%%d");
5517750f TG	408	name_assign_type = NET_NAME_ENUM;
5517750f TG	409	}
e314dbdc	410
81adee47 EB	411	net = rtnl_link_get_net(src_net, tbp);
	412	if (IS_ERR(net))
	413	return PTR_ERR(net);
	414
5517750f TG	415	peer = rtnl_create_link(net, ifname, name_assign_type,
5517750f TG	416	&veth_link_ops, tbp);
81adee47 EB	417	if (IS_ERR(peer)) {
81adee47 EB	418	put_net(net);
e314dbdc	419	return PTR_ERR(peer);
81adee47	420	}
e314dbdc PE	421
e314dbdc PE	422	if (tbp[IFLA_ADDRESS] == NULL)
f2cedb63	423	eth_hw_addr_random(peer);
e6f8f1a7 PE	424
	425	if (ifmp && (dev->ifindex != 0))
	426	peer->ifindex = ifmp->ifi_index;
e314dbdc PE	427
e314dbdc PE	428	err = register_netdevice(peer);
81adee47 EB	429	put_net(net);
81adee47 EB	430	net = NULL;
e314dbdc PE	431	if (err < 0)
	432	goto err_register_peer;
	433
	434	netif_carrier_off(peer);
	435
3729d502 PM	436	err = rtnl_configure_link(peer, ifmp);
	437	if (err < 0)
	438	goto err_configure_peer;
	439
e314dbdc PE	440	/*
	441	* register dev last
	442	*
	443	* note, that since we've registered new device the dev's name
	444	* should be re-allocated
	445	*/
	446
	447	if (tb[IFLA_ADDRESS] == NULL)
f2cedb63	448	eth_hw_addr_random(dev);
e314dbdc	449
6c8c4446 JP	450	if (tb[IFLA_IFNAME])
	451	nla_strlcpy(dev->name, tb[IFLA_IFNAME], IFNAMSIZ);
	452	else
	453	snprintf(dev->name, IFNAMSIZ, DRV_NAME "%%d");
	454
e314dbdc PE	455	err = register_netdevice(dev);
	456	if (err < 0)
	457	goto err_register_dev;
	458
	459	netif_carrier_off(dev);
	460
	461	/*
	462	* tie the deviced together
	463	*/
	464
	465	priv = netdev_priv(dev);
d0e2c55e	466	rcu_assign_pointer(priv->peer, peer);
e314dbdc PE	467
e314dbdc PE	468	priv = netdev_priv(peer);
d0e2c55e	469	rcu_assign_pointer(priv->peer, dev);
e314dbdc PE	470	return 0;
	471
	472	err_register_dev:
	473	/* nothing to do */
3729d502	474	err_configure_peer:
e314dbdc PE	475	unregister_netdevice(peer);
	476	return err;
	477
	478	err_register_peer:
	479	free_netdev(peer);
	480	return err;
	481	}
	482
23289a37	483	static void veth_dellink(struct net_device dev, struct list_head head)
e314dbdc PE	484	{
	485	struct veth_priv *priv;
	486	struct net_device *peer;
	487
	488	priv = netdev_priv(dev);
d0e2c55e ED	489	peer = rtnl_dereference(priv->peer);
	490
	491	/* Note : dellink() is called from default_device_exit_batch(),
	492	* before a rcu_synchronize() point. The devices are guaranteed
	493	* not being freed before one RCU grace period.
	494	*/
	495	RCU_INIT_POINTER(priv->peer, NULL);
24540535	496	unregister_netdevice_queue(dev, head);
f45a5c26 ED	497
	498	if (peer) {
	499	priv = netdev_priv(peer);
	500	RCU_INIT_POINTER(priv->peer, NULL);
	501	unregister_netdevice_queue(peer, head);
	502	}
e314dbdc PE	503	}
e314dbdc PE	504
23711438 TG	505	static const struct nla_policy veth_policy[VETH_INFO_MAX + 1] = {
	506	[VETH_INFO_PEER] = { .len = sizeof(struct ifinfomsg) },
	507	};
e314dbdc	508
e5f4e7b9 ND	509	static struct net veth_get_link_net(const struct net_device dev)
	510	{
	511	struct veth_priv *priv = netdev_priv(dev);
	512	struct net_device *peer = rtnl_dereference(priv->peer);
	513
	514	return peer ? dev_net(peer) : dev_net(dev);
	515	}
	516
e314dbdc PE	517	static struct rtnl_link_ops veth_link_ops = {
	518	.kind = DRV_NAME,
	519	.priv_size = sizeof(struct veth_priv),
	520	.setup = veth_setup,
	521	.validate = veth_validate,
	522	.newlink = veth_newlink,
	523	.dellink = veth_dellink,
	524	.policy = veth_policy,
	525	.maxtype = VETH_INFO_MAX,
e5f4e7b9	526	.get_link_net = veth_get_link_net,
e314dbdc PE	527	};
	528
	529	/*
	530	* init/fini
	531	*/
	532
	533	static __init int veth_init(void)
	534	{
	535	return rtnl_link_register(&veth_link_ops);
	536	}
	537
	538	static __exit void veth_exit(void)
	539	{
68365458	540	rtnl_link_unregister(&veth_link_ops);
e314dbdc PE	541	}
	542
	543	module_init(veth_init);
	544	module_exit(veth_exit);
	545
	546	MODULE_DESCRIPTION("Virtual Ethernet Tunnel");
	547	MODULE_LICENSE("GPL v2");
	548	MODULE_ALIAS_RTNL_LINK(DRV_NAME);