[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;
}

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);

	netif_carrier_off(dev);
	netif_carrier_off(rtnl_dereference(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)
{
	dev->vstats = alloc_percpu(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_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_HW_VLAN_TX | NETIF_F_HW_VLAN_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->destructor = veth_dev_free;

	dev->hw_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");

	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);
	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_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 = 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);

	priv = netdev_priv(peer);
	RCU_INIT_POINTER(priv->peer, NULL);

	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 <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	}
	190
	191	static int veth_open(struct net_device *dev)
	192	{
d0e2c55e ED	193	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e ED	194	struct net_device *peer = rtnl_dereference(priv->peer);
e314dbdc	195
d0e2c55e	196	if (!peer)
e314dbdc PE	197	return -ENOTCONN;
e314dbdc PE	198
d0e2c55e	199	if (peer->flags & IFF_UP) {
e314dbdc	200	netif_carrier_on(dev);
d0e2c55e	201	netif_carrier_on(peer);
e314dbdc PE	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);
d0e2c55e	211	netif_carrier_off(rtnl_dereference(priv->peer));
2cf48a10 EB	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	{
2681128f ED	231	dev->vstats = alloc_percpu(struct pcpu_vstats);
2681128f ED	232	if (!dev->vstats)
e314dbdc PE	233	return -ENOMEM;
e314dbdc PE	234
e314dbdc PE	235	return 0;
	236	}
	237
11687a10 DM	238	static void veth_dev_free(struct net_device *dev)
11687a10 DM	239	{
2681128f	240	free_percpu(dev->vstats);
11687a10 DM	241	free_netdev(dev);
	242	}
	243
4456e7bd	244	static const struct net_device_ops veth_netdev_ops = {
ee923623 DL	245	.ndo_init = veth_dev_init,
ee923623 DL	246	.ndo_open = veth_open,
2cf48a10	247	.ndo_stop = veth_close,
ee923623	248	.ndo_start_xmit = veth_xmit,
38d40815	249	.ndo_change_mtu = veth_change_mtu,
6311cc44	250	.ndo_get_stats64 = veth_get_stats64,
ee923623	251	.ndo_set_mac_address = eth_mac_addr,
4456e7bd SH	252	};
4456e7bd SH	253
8093315a ED	254	#define VETH_FEATURES (NETIF_F_SG \| NETIF_F_FRAGLIST \| NETIF_F_ALL_TSO \| \
	255	NETIF_F_HW_CSUM \| NETIF_F_RXCSUM \| NETIF_F_HIGHDMA \| \
	256	NETIF_F_HW_VLAN_TX \| NETIF_F_HW_VLAN_RX)
	257
e314dbdc PE	258	static void veth_setup(struct net_device *dev)
	259	{
	260	ether_setup(dev);
	261
550fd08c	262	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
23ea5a96	263	dev->priv_flags \|= IFF_LIVE_ADDR_CHANGE;
550fd08c	264
4456e7bd	265	dev->netdev_ops = &veth_netdev_ops;
e314dbdc PE	266	dev->ethtool_ops = &veth_ethtool_ops;
e314dbdc PE	267	dev->features \|= NETIF_F_LLTX;
8093315a	268	dev->features \|= VETH_FEATURES;
11687a10	269	dev->destructor = veth_dev_free;
a2c725fa	270
8093315a	271	dev->hw_features = VETH_FEATURES;
e314dbdc PE	272	}
	273
	274	/*
	275	* netlink interface
	276	*/
	277
	278	static int veth_validate(struct nlattr tb[], struct nlattr data[])
	279	{
	280	if (tb[IFLA_ADDRESS]) {
	281	if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN)
	282	return -EINVAL;
	283	if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS])))
	284	return -EADDRNOTAVAIL;
	285	}
38d40815 EB	286	if (tb[IFLA_MTU]) {
	287	if (!is_valid_veth_mtu(nla_get_u32(tb[IFLA_MTU])))
	288	return -EINVAL;
	289	}
e314dbdc PE	290	return 0;
	291	}
	292
	293	static struct rtnl_link_ops veth_link_ops;
	294
81adee47	295	static int veth_newlink(struct net src_net, struct net_device dev,
e314dbdc PE	296	struct nlattr tb[], struct nlattr data[])
	297	{
	298	int err;
	299	struct net_device *peer;
	300	struct veth_priv *priv;
	301	char ifname[IFNAMSIZ];
	302	struct nlattr peer_tb[IFLA_MAX + 1], *tbp;
3729d502	303	struct ifinfomsg *ifmp;
81adee47	304	struct net *net;
e314dbdc PE	305
	306	/*
	307	* create and register peer first
e314dbdc	308	*/
e314dbdc PE	309	if (data != NULL && data[VETH_INFO_PEER] != NULL) {
	310	struct nlattr *nla_peer;
	311
	312	nla_peer = data[VETH_INFO_PEER];
3729d502	313	ifmp = nla_data(nla_peer);
e314dbdc PE	314	err = nla_parse(peer_tb, IFLA_MAX,
	315	nla_data(nla_peer) + sizeof(struct ifinfomsg),
	316	nla_len(nla_peer) - sizeof(struct ifinfomsg),
	317	ifla_policy);
	318	if (err < 0)
	319	return err;
	320
	321	err = veth_validate(peer_tb, NULL);
	322	if (err < 0)
	323	return err;
	324
	325	tbp = peer_tb;
3729d502 PM	326	} else {
3729d502 PM	327	ifmp = NULL;
e314dbdc	328	tbp = tb;
3729d502	329	}
e314dbdc PE	330
	331	if (tbp[IFLA_IFNAME])
	332	nla_strlcpy(ifname, tbp[IFLA_IFNAME], IFNAMSIZ);
	333	else
	334	snprintf(ifname, IFNAMSIZ, DRV_NAME "%%d");
	335
81adee47 EB	336	net = rtnl_link_get_net(src_net, tbp);
	337	if (IS_ERR(net))
	338	return PTR_ERR(net);
	339
c0713563	340	peer = rtnl_create_link(net, ifname, &veth_link_ops, tbp);
81adee47 EB	341	if (IS_ERR(peer)) {
81adee47 EB	342	put_net(net);
e314dbdc	343	return PTR_ERR(peer);
81adee47	344	}
e314dbdc PE	345
e314dbdc PE	346	if (tbp[IFLA_ADDRESS] == NULL)
f2cedb63	347	eth_hw_addr_random(peer);
e6f8f1a7 PE	348
	349	if (ifmp && (dev->ifindex != 0))
	350	peer->ifindex = ifmp->ifi_index;
e314dbdc PE	351
e314dbdc PE	352	err = register_netdevice(peer);
81adee47 EB	353	put_net(net);
81adee47 EB	354	net = NULL;
e314dbdc PE	355	if (err < 0)
	356	goto err_register_peer;
	357
	358	netif_carrier_off(peer);
	359
3729d502 PM	360	err = rtnl_configure_link(peer, ifmp);
	361	if (err < 0)
	362	goto err_configure_peer;
	363
e314dbdc PE	364	/*
	365	* register dev last
	366	*
	367	* note, that since we've registered new device the dev's name
	368	* should be re-allocated
	369	*/
	370
	371	if (tb[IFLA_ADDRESS] == NULL)
f2cedb63	372	eth_hw_addr_random(dev);
e314dbdc	373
6c8c4446 JP	374	if (tb[IFLA_IFNAME])
	375	nla_strlcpy(dev->name, tb[IFLA_IFNAME], IFNAMSIZ);
	376	else
	377	snprintf(dev->name, IFNAMSIZ, DRV_NAME "%%d");
	378
	379	if (strchr(dev->name, '%')) {
	380	err = dev_alloc_name(dev, dev->name);
	381	if (err < 0)
	382	goto err_alloc_name;
	383	}
	384
e314dbdc PE	385	err = register_netdevice(dev);
	386	if (err < 0)
	387	goto err_register_dev;
	388
	389	netif_carrier_off(dev);
	390
	391	/*
	392	* tie the deviced together
	393	*/
	394
	395	priv = netdev_priv(dev);
d0e2c55e	396	rcu_assign_pointer(priv->peer, peer);
e314dbdc PE	397
e314dbdc PE	398	priv = netdev_priv(peer);
d0e2c55e	399	rcu_assign_pointer(priv->peer, dev);
e314dbdc PE	400	return 0;
	401
	402	err_register_dev:
	403	/* nothing to do */
6c8c4446	404	err_alloc_name:
3729d502	405	err_configure_peer:
e314dbdc PE	406	unregister_netdevice(peer);
	407	return err;
	408
	409	err_register_peer:
	410	free_netdev(peer);
	411	return err;
	412	}
	413
23289a37	414	static void veth_dellink(struct net_device dev, struct list_head head)
e314dbdc PE	415	{
	416	struct veth_priv *priv;
	417	struct net_device *peer;
	418
	419	priv = netdev_priv(dev);
d0e2c55e ED	420	peer = rtnl_dereference(priv->peer);
	421
	422	/* Note : dellink() is called from default_device_exit_batch(),
	423	* before a rcu_synchronize() point. The devices are guaranteed
	424	* not being freed before one RCU grace period.
	425	*/
	426	RCU_INIT_POINTER(priv->peer, NULL);
	427
	428	priv = netdev_priv(peer);
	429	RCU_INIT_POINTER(priv->peer, NULL);
e314dbdc	430
24540535 ED	431	unregister_netdevice_queue(dev, head);
24540535 ED	432	unregister_netdevice_queue(peer, head);
e314dbdc PE	433	}
e314dbdc PE	434
23711438 TG	435	static const struct nla_policy veth_policy[VETH_INFO_MAX + 1] = {
	436	[VETH_INFO_PEER] = { .len = sizeof(struct ifinfomsg) },
	437	};
e314dbdc PE	438
	439	static struct rtnl_link_ops veth_link_ops = {
	440	.kind = DRV_NAME,
	441	.priv_size = sizeof(struct veth_priv),
	442	.setup = veth_setup,
	443	.validate = veth_validate,
	444	.newlink = veth_newlink,
	445	.dellink = veth_dellink,
	446	.policy = veth_policy,
	447	.maxtype = VETH_INFO_MAX,
	448	};
	449
	450	/*
	451	* init/fini
	452	*/
	453
	454	static __init int veth_init(void)
	455	{
	456	return rtnl_link_register(&veth_link_ops);
	457	}
	458
	459	static __exit void veth_exit(void)
	460	{
68365458	461	rtnl_link_unregister(&veth_link_ops);
e314dbdc PE	462	}
	463
	464	module_init(veth_init);
	465	module_exit(veth_exit);
	466
	467	MODULE_DESCRIPTION("Virtual Ethernet Tunnel");
	468	MODULE_LICENSE("GPL v2");
	469	MODULE_ALIAS_RTNL_LINK(DRV_NAME);