[mirror_ubuntu-artful-kernel.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/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"

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 void 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();
}

/* 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 mtu)
{
	return mtu >= ETH_MIN_MTU && mtu <= ETH_MAX_MTU;
}

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_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_HW_CSUM | \
		       NETIF_F_RXCSUM | NETIF_F_SCTP_CRC | NETIF_F_HIGHDMA | \
		       NETIF_F_GSO_SOFTWARE | NETIF_F_GSO_ENCAP_ALL | \
		       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->max_mtu = ETH_MAX_MTU;

	dev->hw_features = VETH_FEATURES;
	dev->hw_enc_features = VETH_FEATURES;
	dev->mpls_features = NETIF_F_HW_CSUM | NETIF_F_GSO_SOFTWARE;
}

/*
 * 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 <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	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
2681128f ED	26	struct pcpu_vstats {
	27	u64 packets;
	28	u64 bytes;
cf05c700	29	struct u64_stats_sync syncp;
e314dbdc PE	30	};
	31
	32	struct veth_priv {
d0e2c55e	33	struct net_device __rcu *peer;
2681128f	34	atomic64_t dropped;
163e5292	35	unsigned requested_headroom;
e314dbdc PE	36	};
e314dbdc PE	37
e314dbdc PE	38	/*
	39	* ethtool interface
	40	*/
	41
	42	static struct {
	43	const char string[ETH_GSTRING_LEN];
	44	} ethtool_stats_keys[] = {
	45	{ "peer_ifindex" },
	46	};
	47
	48	static int veth_get_settings(struct net_device dev, struct ethtool_cmd cmd)
	49	{
	50	cmd->supported = 0;
	51	cmd->advertising = 0;
70739497	52	ethtool_cmd_speed_set(cmd, SPEED_10000);
e314dbdc PE	53	cmd->duplex = DUPLEX_FULL;
	54	cmd->port = PORT_TP;
	55	cmd->phy_address = 0;
	56	cmd->transceiver = XCVR_INTERNAL;
	57	cmd->autoneg = AUTONEG_DISABLE;
	58	cmd->maxtxpkt = 0;
	59	cmd->maxrxpkt = 0;
	60	return 0;
	61	}
	62
	63	static void veth_get_drvinfo(struct net_device dev, struct ethtool_drvinfo info)
	64	{
33a5ba14 RJ	65	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
33a5ba14 RJ	66	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
e314dbdc PE	67	}
	68
	69	static void veth_get_strings(struct net_device dev, u32 stringset, u8 buf)
	70	{
	71	switch(stringset) {
	72	case ETH_SS_STATS:
	73	memcpy(buf, &ethtool_stats_keys, sizeof(ethtool_stats_keys));
	74	break;
	75	}
	76	}
	77
b9f2c044	78	static int veth_get_sset_count(struct net_device *dev, int sset)
e314dbdc	79	{
b9f2c044 JG	80	switch (sset) {
	81	case ETH_SS_STATS:
	82	return ARRAY_SIZE(ethtool_stats_keys);
	83	default:
	84	return -EOPNOTSUPP;
	85	}
e314dbdc PE	86	}
	87
	88	static void veth_get_ethtool_stats(struct net_device *dev,
	89	struct ethtool_stats stats, u64 data)
	90	{
d0e2c55e ED	91	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e ED	92	struct net_device *peer = rtnl_dereference(priv->peer);
e314dbdc	93
d0e2c55e	94	data[0] = peer ? peer->ifindex : 0;
e314dbdc PE	95	}
e314dbdc PE	96
0fc0b732	97	static const struct ethtool_ops veth_ethtool_ops = {
e314dbdc PE	98	.get_settings = veth_get_settings,
	99	.get_drvinfo = veth_get_drvinfo,
	100	.get_link = ethtool_op_get_link,
e314dbdc	101	.get_strings = veth_get_strings,
b9f2c044	102	.get_sset_count = veth_get_sset_count,
e314dbdc PE	103	.get_ethtool_stats = veth_get_ethtool_stats,
	104	};
	105
424efe9c	106	static netdev_tx_t veth_xmit(struct sk_buff skb, struct net_device dev)
e314dbdc	107	{
2681128f	108	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e	109	struct net_device *rcv;
2681128f	110	int length = skb->len;
e314dbdc	111
d0e2c55e ED	112	rcu_read_lock();
	113	rcv = rcu_dereference(priv->peer);
	114	if (unlikely(!rcv)) {
	115	kfree_skb(skb);
	116	goto drop;
	117	}
e314dbdc	118
2681128f ED	119	if (likely(dev_forward_skb(rcv, skb) == NET_RX_SUCCESS)) {
2681128f ED	120	struct pcpu_vstats *stats = this_cpu_ptr(dev->vstats);
e314dbdc	121
2681128f ED	122	u64_stats_update_begin(&stats->syncp);
	123	stats->bytes += length;
	124	stats->packets++;
	125	u64_stats_update_end(&stats->syncp);
	126	} else {
d0e2c55e	127	drop:
2681128f ED	128	atomic64_inc(&priv->dropped);
2681128f ED	129	}
d0e2c55e	130	rcu_read_unlock();
6ed10654	131	return NETDEV_TX_OK;
e314dbdc PE	132	}
	133
	134	/*
	135	* general routines
	136	*/
	137
2681128f	138	static u64 veth_stats_one(struct pcpu_vstats result, struct net_device dev)
e314dbdc	139	{
cf05c700	140	struct veth_priv *priv = netdev_priv(dev);
11687a10	141	int cpu;
e314dbdc	142
2681128f ED	143	result->packets = 0;
2681128f ED	144	result->bytes = 0;
2b1c8b0f	145	for_each_possible_cpu(cpu) {
2681128f ED	146	struct pcpu_vstats *stats = per_cpu_ptr(dev->vstats, cpu);
2681128f ED	147	u64 packets, bytes;
cf05c700 ED	148	unsigned int start;
	149
	150	do {
57a7744e	151	start = u64_stats_fetch_begin_irq(&stats->syncp);
2681128f ED	152	packets = stats->packets;
2681128f ED	153	bytes = stats->bytes;
57a7744e	154	} while (u64_stats_fetch_retry_irq(&stats->syncp, start));
2681128f ED	155	result->packets += packets;
2681128f ED	156	result->bytes += bytes;
11687a10	157	}
2681128f ED	158	return atomic64_read(&priv->dropped);
	159	}
	160
bc1f4470	161	static void veth_get_stats64(struct net_device *dev,
bc1f4470	162	struct rtnl_link_stats64 *tot)
2681128f ED	163	{
2681128f ED	164	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e	165	struct net_device *peer;
2681128f ED	166	struct pcpu_vstats one;
	167
	168	tot->tx_dropped = veth_stats_one(&one, dev);
	169	tot->tx_bytes = one.bytes;
	170	tot->tx_packets = one.packets;
	171
d0e2c55e ED	172	rcu_read_lock();
	173	peer = rcu_dereference(priv->peer);
	174	if (peer) {
	175	tot->rx_dropped = veth_stats_one(&one, peer);
	176	tot->rx_bytes = one.bytes;
	177	tot->rx_packets = one.packets;
	178	}
	179	rcu_read_unlock();
e314dbdc PE	180	}
e314dbdc PE	181
5c70ef85 G	182	/* fake multicast ability */
	183	static void veth_set_multicast_list(struct net_device *dev)
	184	{
	185	}
	186
e314dbdc PE	187	static int veth_open(struct net_device *dev)
e314dbdc PE	188	{
d0e2c55e ED	189	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e ED	190	struct net_device *peer = rtnl_dereference(priv->peer);
e314dbdc	191
d0e2c55e	192	if (!peer)
e314dbdc PE	193	return -ENOTCONN;
e314dbdc PE	194
d0e2c55e	195	if (peer->flags & IFF_UP) {
e314dbdc	196	netif_carrier_on(dev);
d0e2c55e	197	netif_carrier_on(peer);
e314dbdc PE	198	}
	199	return 0;
	200	}
	201
2cf48a10 EB	202	static int veth_close(struct net_device *dev)
	203	{
	204	struct veth_priv *priv = netdev_priv(dev);
2efd32ee	205	struct net_device *peer = rtnl_dereference(priv->peer);
2cf48a10 EB	206
2cf48a10 EB	207	netif_carrier_off(dev);
2efd32ee ED	208	if (peer)
2efd32ee ED	209	netif_carrier_off(peer);
2cf48a10 EB	210
	211	return 0;
	212	}
	213
91572088	214	static int is_valid_veth_mtu(int mtu)
38d40815	215	{
91572088	216	return mtu >= ETH_MIN_MTU && mtu <= ETH_MAX_MTU;
38d40815 EB	217	}
38d40815 EB	218
e314dbdc PE	219	static int veth_dev_init(struct net_device *dev)
e314dbdc PE	220	{
1c213bd2	221	dev->vstats = netdev_alloc_pcpu_stats(struct pcpu_vstats);
2681128f	222	if (!dev->vstats)
e314dbdc	223	return -ENOMEM;
e314dbdc PE	224	return 0;
	225	}
	226
11687a10 DM	227	static void veth_dev_free(struct net_device *dev)
11687a10 DM	228	{
2681128f	229	free_percpu(dev->vstats);
11687a10 DM	230	free_netdev(dev);
	231	}
	232
bb446c19 WC	233	#ifdef CONFIG_NET_POLL_CONTROLLER
	234	static void veth_poll_controller(struct net_device *dev)
	235	{
	236	/* veth only receives frames when its peer sends one
	237	* Since it's a synchronous operation, we are guaranteed
	238	* never to have pending data when we poll for it so
	239	* there is nothing to do here.
	240	*
	241	* We need this though so netpoll recognizes us as an interface that
	242	* supports polling, which enables bridge devices in virt setups to
	243	* still use netconsole
	244	*/
	245	}
	246	#endif /* CONFIG_NET_POLL_CONTROLLER */
	247
a45253bf ND	248	static int veth_get_iflink(const struct net_device *dev)
	249	{
	250	struct veth_priv *priv = netdev_priv(dev);
	251	struct net_device *peer;
	252	int iflink;
	253
	254	rcu_read_lock();
	255	peer = rcu_dereference(priv->peer);
	256	iflink = peer ? peer->ifindex : 0;
	257	rcu_read_unlock();
	258
	259	return iflink;
	260	}
	261
163e5292 PA	262	static void veth_set_rx_headroom(struct net_device *dev, int new_hr)
	263	{
	264	struct veth_priv peer_priv, priv = netdev_priv(dev);
	265	struct net_device *peer;
	266
	267	if (new_hr < 0)
	268	new_hr = 0;
	269
	270	rcu_read_lock();
	271	peer = rcu_dereference(priv->peer);
	272	if (unlikely(!peer))
	273	goto out;
	274
	275	peer_priv = netdev_priv(peer);
	276	priv->requested_headroom = new_hr;
	277	new_hr = max(priv->requested_headroom, peer_priv->requested_headroom);
	278	dev->needed_headroom = new_hr;
	279	peer->needed_headroom = new_hr;
	280
	281	out:
	282	rcu_read_unlock();
	283	}
	284
4456e7bd	285	static const struct net_device_ops veth_netdev_ops = {
ee923623 DL	286	.ndo_init = veth_dev_init,
ee923623 DL	287	.ndo_open = veth_open,
2cf48a10	288	.ndo_stop = veth_close,
ee923623	289	.ndo_start_xmit = veth_xmit,
6311cc44	290	.ndo_get_stats64 = veth_get_stats64,
5c70ef85	291	.ndo_set_rx_mode = veth_set_multicast_list,
ee923623	292	.ndo_set_mac_address = eth_mac_addr,
bb446c19 WC	293	#ifdef CONFIG_NET_POLL_CONTROLLER
	294	.ndo_poll_controller = veth_poll_controller,
	295	#endif
a45253bf	296	.ndo_get_iflink = veth_get_iflink,
1a04a821	297	.ndo_features_check = passthru_features_check,
163e5292	298	.ndo_set_rx_headroom = veth_set_rx_headroom,
4456e7bd SH	299	};
4456e7bd SH	300
732912d7	301	#define VETH_FEATURES (NETIF_F_SG \| NETIF_F_FRAGLIST \| NETIF_F_HW_CSUM \| \
c80fafbb	302	NETIF_F_RXCSUM \| NETIF_F_SCTP_CRC \| NETIF_F_HIGHDMA \| \
732912d7	303	NETIF_F_GSO_SOFTWARE \| NETIF_F_GSO_ENCAP_ALL \| \
28d2b136 PM	304	NETIF_F_HW_VLAN_CTAG_TX \| NETIF_F_HW_VLAN_CTAG_RX \| \
28d2b136 PM	305	NETIF_F_HW_VLAN_STAG_TX \| NETIF_F_HW_VLAN_STAG_RX )
8093315a	306
e314dbdc PE	307	static void veth_setup(struct net_device *dev)
	308	{
	309	ether_setup(dev);
	310
550fd08c	311	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
23ea5a96	312	dev->priv_flags \|= IFF_LIVE_ADDR_CHANGE;
02f01ec1	313	dev->priv_flags \|= IFF_NO_QUEUE;
163e5292	314	dev->priv_flags \|= IFF_PHONY_HEADROOM;
550fd08c	315
4456e7bd	316	dev->netdev_ops = &veth_netdev_ops;
e314dbdc PE	317	dev->ethtool_ops = &veth_ethtool_ops;
e314dbdc PE	318	dev->features \|= NETIF_F_LLTX;
8093315a	319	dev->features \|= VETH_FEATURES;
8d0d21f4	320	dev->vlan_features = dev->features &
3f8c707b VY	321	~(NETIF_F_HW_VLAN_CTAG_TX \|
	322	NETIF_F_HW_VLAN_STAG_TX \|
	323	NETIF_F_HW_VLAN_CTAG_RX \|
	324	NETIF_F_HW_VLAN_STAG_RX);
11687a10	325	dev->destructor = veth_dev_free;
91572088	326	dev->max_mtu = ETH_MAX_MTU;
a2c725fa	327
8093315a	328	dev->hw_features = VETH_FEATURES;
82d81898	329	dev->hw_enc_features = VETH_FEATURES;
607fca9a	330	dev->mpls_features = NETIF_F_HW_CSUM \| NETIF_F_GSO_SOFTWARE;
e314dbdc PE	331	}
	332
	333	/*
	334	* netlink interface
	335	*/
	336
	337	static int veth_validate(struct nlattr tb[], struct nlattr data[])
	338	{
	339	if (tb[IFLA_ADDRESS]) {
	340	if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN)
	341	return -EINVAL;
	342	if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS])))
	343	return -EADDRNOTAVAIL;
	344	}
38d40815 EB	345	if (tb[IFLA_MTU]) {
	346	if (!is_valid_veth_mtu(nla_get_u32(tb[IFLA_MTU])))
	347	return -EINVAL;
	348	}
e314dbdc PE	349	return 0;
	350	}
	351
	352	static struct rtnl_link_ops veth_link_ops;
	353
81adee47	354	static int veth_newlink(struct net src_net, struct net_device dev,
e314dbdc PE	355	struct nlattr tb[], struct nlattr data[])
	356	{
	357	int err;
	358	struct net_device *peer;
	359	struct veth_priv *priv;
	360	char ifname[IFNAMSIZ];
	361	struct nlattr peer_tb[IFLA_MAX + 1], *tbp;
5517750f	362	unsigned char name_assign_type;
3729d502	363	struct ifinfomsg *ifmp;
81adee47	364	struct net *net;
e314dbdc PE	365
	366	/*
	367	* create and register peer first
e314dbdc	368	*/
e314dbdc PE	369	if (data != NULL && data[VETH_INFO_PEER] != NULL) {
	370	struct nlattr *nla_peer;
	371
	372	nla_peer = data[VETH_INFO_PEER];
3729d502	373	ifmp = nla_data(nla_peer);
f7b12606 JP	374	err = rtnl_nla_parse_ifla(peer_tb,
	375	nla_data(nla_peer) + sizeof(struct ifinfomsg),
	376	nla_len(nla_peer) - sizeof(struct ifinfomsg));
e314dbdc PE	377	if (err < 0)
	378	return err;
	379
	380	err = veth_validate(peer_tb, NULL);
	381	if (err < 0)
	382	return err;
	383
	384	tbp = peer_tb;
3729d502 PM	385	} else {
3729d502 PM	386	ifmp = NULL;
e314dbdc	387	tbp = tb;
3729d502	388	}
e314dbdc	389
5517750f	390	if (tbp[IFLA_IFNAME]) {
e314dbdc	391	nla_strlcpy(ifname, tbp[IFLA_IFNAME], IFNAMSIZ);
5517750f TG	392	name_assign_type = NET_NAME_USER;
5517750f TG	393	} else {
e314dbdc	394	snprintf(ifname, IFNAMSIZ, DRV_NAME "%%d");
5517750f TG	395	name_assign_type = NET_NAME_ENUM;
5517750f TG	396	}
e314dbdc	397
81adee47 EB	398	net = rtnl_link_get_net(src_net, tbp);
	399	if (IS_ERR(net))
	400	return PTR_ERR(net);
	401
5517750f TG	402	peer = rtnl_create_link(net, ifname, name_assign_type,
5517750f TG	403	&veth_link_ops, tbp);
81adee47 EB	404	if (IS_ERR(peer)) {
81adee47 EB	405	put_net(net);
e314dbdc	406	return PTR_ERR(peer);
81adee47	407	}
e314dbdc PE	408
e314dbdc PE	409	if (tbp[IFLA_ADDRESS] == NULL)
f2cedb63	410	eth_hw_addr_random(peer);
e6f8f1a7 PE	411
	412	if (ifmp && (dev->ifindex != 0))
	413	peer->ifindex = ifmp->ifi_index;
e314dbdc PE	414
e314dbdc PE	415	err = register_netdevice(peer);
81adee47 EB	416	put_net(net);
81adee47 EB	417	net = NULL;
e314dbdc PE	418	if (err < 0)
	419	goto err_register_peer;
	420
	421	netif_carrier_off(peer);
	422
3729d502 PM	423	err = rtnl_configure_link(peer, ifmp);
	424	if (err < 0)
	425	goto err_configure_peer;
	426
e314dbdc PE	427	/*
	428	* register dev last
	429	*
	430	* note, that since we've registered new device the dev's name
	431	* should be re-allocated
	432	*/
	433
	434	if (tb[IFLA_ADDRESS] == NULL)
f2cedb63	435	eth_hw_addr_random(dev);
e314dbdc	436
6c8c4446 JP	437	if (tb[IFLA_IFNAME])
	438	nla_strlcpy(dev->name, tb[IFLA_IFNAME], IFNAMSIZ);
	439	else
	440	snprintf(dev->name, IFNAMSIZ, DRV_NAME "%%d");
	441
e314dbdc PE	442	err = register_netdevice(dev);
	443	if (err < 0)
	444	goto err_register_dev;
	445
	446	netif_carrier_off(dev);
	447
	448	/*
	449	* tie the deviced together
	450	*/
	451
	452	priv = netdev_priv(dev);
d0e2c55e	453	rcu_assign_pointer(priv->peer, peer);
e314dbdc PE	454
e314dbdc PE	455	priv = netdev_priv(peer);
d0e2c55e	456	rcu_assign_pointer(priv->peer, dev);
e314dbdc PE	457	return 0;
	458
	459	err_register_dev:
	460	/* nothing to do */
3729d502	461	err_configure_peer:
e314dbdc PE	462	unregister_netdevice(peer);
	463	return err;
	464
	465	err_register_peer:
	466	free_netdev(peer);
	467	return err;
	468	}
	469
23289a37	470	static void veth_dellink(struct net_device dev, struct list_head head)
e314dbdc PE	471	{
	472	struct veth_priv *priv;
	473	struct net_device *peer;
	474
	475	priv = netdev_priv(dev);
d0e2c55e ED	476	peer = rtnl_dereference(priv->peer);
	477
	478	/* Note : dellink() is called from default_device_exit_batch(),
	479	* before a rcu_synchronize() point. The devices are guaranteed
	480	* not being freed before one RCU grace period.
	481	*/
	482	RCU_INIT_POINTER(priv->peer, NULL);
24540535	483	unregister_netdevice_queue(dev, head);
f45a5c26 ED	484
	485	if (peer) {
	486	priv = netdev_priv(peer);
	487	RCU_INIT_POINTER(priv->peer, NULL);
	488	unregister_netdevice_queue(peer, head);
	489	}
e314dbdc PE	490	}
e314dbdc PE	491
23711438 TG	492	static const struct nla_policy veth_policy[VETH_INFO_MAX + 1] = {
	493	[VETH_INFO_PEER] = { .len = sizeof(struct ifinfomsg) },
	494	};
e314dbdc	495
e5f4e7b9 ND	496	static struct net veth_get_link_net(const struct net_device dev)
	497	{
	498	struct veth_priv *priv = netdev_priv(dev);
	499	struct net_device *peer = rtnl_dereference(priv->peer);
	500
	501	return peer ? dev_net(peer) : dev_net(dev);
	502	}
	503
e314dbdc PE	504	static struct rtnl_link_ops veth_link_ops = {
	505	.kind = DRV_NAME,
	506	.priv_size = sizeof(struct veth_priv),
	507	.setup = veth_setup,
	508	.validate = veth_validate,
	509	.newlink = veth_newlink,
	510	.dellink = veth_dellink,
	511	.policy = veth_policy,
	512	.maxtype = VETH_INFO_MAX,
e5f4e7b9	513	.get_link_net = veth_get_link_net,
e314dbdc PE	514	};
	515
	516	/*
	517	* init/fini
	518	*/
	519
	520	static __init int veth_init(void)
	521	{
	522	return rtnl_link_register(&veth_link_ops);
	523	}
	524
	525	static __exit void veth_exit(void)
	526	{
68365458	527	rtnl_link_unregister(&veth_link_ops);
e314dbdc PE	528	}
	529
	530	module_init(veth_init);
	531	module_exit(veth_exit);
	532
	533	MODULE_DESCRIPTION("Virtual Ethernet Tunnel");
	534	MODULE_LICENSE("GPL v2");
	535	MODULE_ALIAS_RTNL_LINK(DRV_NAME);