[mirror_ubuntu-bionic-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_link_ksettings(struct net_device *dev,
				   struct ethtool_link_ksettings *cmd)
{
	cmd->base.speed		= SPEED_10000;
	cmd->base.duplex	= DUPLEX_FULL;
	cmd->base.port		= PORT_TP;
	cmd->base.autoneg	= AUTONEG_DISABLE;
	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_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,
	.get_link_ksettings	= veth_get_link_ksettings,
};

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);
}

#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->needs_free_netdev = true;
	dev->priv_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[],
			 struct netlink_ext_ack *extack)
{
	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[],
			struct netlink_ext_ack *extack)
{
	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),
					  NULL);
		if (err < 0)
			return err;

		err = veth_validate(peer_tb, NULL, extack);
		if (err < 0)
			return err;

		tbp = peer_tb;
	} else {
		ifmp = NULL;
		tbp = tb;
	}

	if (ifmp && 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 (!ifmp || !tbp[IFLA_ADDRESS])
		eth_hw_addr_random(peer);

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

	peer->gso_max_size = dev->gso_max_size;
	peer->gso_max_segs = dev->gso_max_segs;

	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
56607b98 PR	48	static int veth_get_link_ksettings(struct net_device *dev,
56607b98 PR	49	struct ethtool_link_ksettings *cmd)
e314dbdc	50	{
56607b98 PR	51	cmd->base.speed = SPEED_10000;
	52	cmd->base.duplex = DUPLEX_FULL;
	53	cmd->base.port = PORT_TP;
	54	cmd->base.autoneg = AUTONEG_DISABLE;
e314dbdc PE	55	return 0;
	56	}
	57
	58	static void veth_get_drvinfo(struct net_device dev, struct ethtool_drvinfo info)
	59	{
33a5ba14 RJ	60	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
33a5ba14 RJ	61	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
e314dbdc PE	62	}
	63
	64	static void veth_get_strings(struct net_device dev, u32 stringset, u8 buf)
	65	{
	66	switch(stringset) {
	67	case ETH_SS_STATS:
	68	memcpy(buf, &ethtool_stats_keys, sizeof(ethtool_stats_keys));
	69	break;
	70	}
	71	}
	72
b9f2c044	73	static int veth_get_sset_count(struct net_device *dev, int sset)
e314dbdc	74	{
b9f2c044 JG	75	switch (sset) {
	76	case ETH_SS_STATS:
	77	return ARRAY_SIZE(ethtool_stats_keys);
	78	default:
	79	return -EOPNOTSUPP;
	80	}
e314dbdc PE	81	}
	82
	83	static void veth_get_ethtool_stats(struct net_device *dev,
	84	struct ethtool_stats stats, u64 data)
	85	{
d0e2c55e ED	86	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e ED	87	struct net_device *peer = rtnl_dereference(priv->peer);
e314dbdc	88
d0e2c55e	89	data[0] = peer ? peer->ifindex : 0;
e314dbdc PE	90	}
e314dbdc PE	91
0fc0b732	92	static const struct ethtool_ops veth_ethtool_ops = {
e314dbdc PE	93	.get_drvinfo = veth_get_drvinfo,
e314dbdc PE	94	.get_link = ethtool_op_get_link,
e314dbdc	95	.get_strings = veth_get_strings,
b9f2c044	96	.get_sset_count = veth_get_sset_count,
e314dbdc	97	.get_ethtool_stats = veth_get_ethtool_stats,
56607b98	98	.get_link_ksettings = veth_get_link_ksettings,
e314dbdc PE	99	};
e314dbdc PE	100
424efe9c	101	static netdev_tx_t veth_xmit(struct sk_buff skb, struct net_device dev)
e314dbdc	102	{
2681128f	103	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e	104	struct net_device *rcv;
2681128f	105	int length = skb->len;
e314dbdc	106
d0e2c55e ED	107	rcu_read_lock();
	108	rcv = rcu_dereference(priv->peer);
	109	if (unlikely(!rcv)) {
	110	kfree_skb(skb);
	111	goto drop;
	112	}
e314dbdc	113
2681128f ED	114	if (likely(dev_forward_skb(rcv, skb) == NET_RX_SUCCESS)) {
2681128f ED	115	struct pcpu_vstats *stats = this_cpu_ptr(dev->vstats);
e314dbdc	116
2681128f ED	117	u64_stats_update_begin(&stats->syncp);
	118	stats->bytes += length;
	119	stats->packets++;
	120	u64_stats_update_end(&stats->syncp);
	121	} else {
d0e2c55e	122	drop:
2681128f ED	123	atomic64_inc(&priv->dropped);
2681128f ED	124	}
d0e2c55e	125	rcu_read_unlock();
6ed10654	126	return NETDEV_TX_OK;
e314dbdc PE	127	}
	128
	129	/*
	130	* general routines
	131	*/
	132
2681128f	133	static u64 veth_stats_one(struct pcpu_vstats result, struct net_device dev)
e314dbdc	134	{
cf05c700	135	struct veth_priv *priv = netdev_priv(dev);
11687a10	136	int cpu;
e314dbdc	137
2681128f ED	138	result->packets = 0;
2681128f ED	139	result->bytes = 0;
2b1c8b0f	140	for_each_possible_cpu(cpu) {
2681128f ED	141	struct pcpu_vstats *stats = per_cpu_ptr(dev->vstats, cpu);
2681128f ED	142	u64 packets, bytes;
cf05c700 ED	143	unsigned int start;
	144
	145	do {
57a7744e	146	start = u64_stats_fetch_begin_irq(&stats->syncp);
2681128f ED	147	packets = stats->packets;
2681128f ED	148	bytes = stats->bytes;
57a7744e	149	} while (u64_stats_fetch_retry_irq(&stats->syncp, start));
2681128f ED	150	result->packets += packets;
2681128f ED	151	result->bytes += bytes;
11687a10	152	}
2681128f ED	153	return atomic64_read(&priv->dropped);
	154	}
	155
bc1f4470	156	static void veth_get_stats64(struct net_device *dev,
bc1f4470	157	struct rtnl_link_stats64 *tot)
2681128f ED	158	{
2681128f ED	159	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e	160	struct net_device *peer;
2681128f ED	161	struct pcpu_vstats one;
	162
	163	tot->tx_dropped = veth_stats_one(&one, dev);
	164	tot->tx_bytes = one.bytes;
	165	tot->tx_packets = one.packets;
	166
d0e2c55e ED	167	rcu_read_lock();
	168	peer = rcu_dereference(priv->peer);
	169	if (peer) {
	170	tot->rx_dropped = veth_stats_one(&one, peer);
	171	tot->rx_bytes = one.bytes;
	172	tot->rx_packets = one.packets;
	173	}
	174	rcu_read_unlock();
e314dbdc PE	175	}
e314dbdc PE	176
5c70ef85 G	177	/* fake multicast ability */
	178	static void veth_set_multicast_list(struct net_device *dev)
	179	{
	180	}
	181
e314dbdc PE	182	static int veth_open(struct net_device *dev)
e314dbdc PE	183	{
d0e2c55e ED	184	struct veth_priv *priv = netdev_priv(dev);
d0e2c55e ED	185	struct net_device *peer = rtnl_dereference(priv->peer);
e314dbdc	186
d0e2c55e	187	if (!peer)
e314dbdc PE	188	return -ENOTCONN;
e314dbdc PE	189
d0e2c55e	190	if (peer->flags & IFF_UP) {
e314dbdc	191	netif_carrier_on(dev);
d0e2c55e	192	netif_carrier_on(peer);
e314dbdc PE	193	}
	194	return 0;
	195	}
	196
2cf48a10 EB	197	static int veth_close(struct net_device *dev)
	198	{
	199	struct veth_priv *priv = netdev_priv(dev);
2efd32ee	200	struct net_device *peer = rtnl_dereference(priv->peer);
2cf48a10 EB	201
2cf48a10 EB	202	netif_carrier_off(dev);
2efd32ee ED	203	if (peer)
2efd32ee ED	204	netif_carrier_off(peer);
2cf48a10 EB	205
	206	return 0;
	207	}
	208
91572088	209	static int is_valid_veth_mtu(int mtu)
38d40815	210	{
91572088	211	return mtu >= ETH_MIN_MTU && mtu <= ETH_MAX_MTU;
38d40815 EB	212	}
38d40815 EB	213
e314dbdc PE	214	static int veth_dev_init(struct net_device *dev)
e314dbdc PE	215	{
1c213bd2	216	dev->vstats = netdev_alloc_pcpu_stats(struct pcpu_vstats);
2681128f	217	if (!dev->vstats)
e314dbdc	218	return -ENOMEM;
e314dbdc PE	219	return 0;
	220	}
	221
11687a10 DM	222	static void veth_dev_free(struct net_device *dev)
11687a10 DM	223	{
2681128f	224	free_percpu(dev->vstats);
11687a10 DM	225	}
11687a10 DM	226
bb446c19 WC	227	#ifdef CONFIG_NET_POLL_CONTROLLER
	228	static void veth_poll_controller(struct net_device *dev)
	229	{
	230	/* veth only receives frames when its peer sends one
	231	* Since it's a synchronous operation, we are guaranteed
	232	* never to have pending data when we poll for it so
	233	* there is nothing to do here.
	234	*
	235	* We need this though so netpoll recognizes us as an interface that
	236	* supports polling, which enables bridge devices in virt setups to
	237	* still use netconsole
	238	*/
	239	}
	240	#endif /* CONFIG_NET_POLL_CONTROLLER */
	241
a45253bf ND	242	static int veth_get_iflink(const struct net_device *dev)
	243	{
	244	struct veth_priv *priv = netdev_priv(dev);
	245	struct net_device *peer;
	246	int iflink;
	247
	248	rcu_read_lock();
	249	peer = rcu_dereference(priv->peer);
	250	iflink = peer ? peer->ifindex : 0;
	251	rcu_read_unlock();
	252
	253	return iflink;
	254	}
	255
163e5292 PA	256	static void veth_set_rx_headroom(struct net_device *dev, int new_hr)
	257	{
	258	struct veth_priv peer_priv, priv = netdev_priv(dev);
	259	struct net_device *peer;
	260
	261	if (new_hr < 0)
	262	new_hr = 0;
	263
	264	rcu_read_lock();
	265	peer = rcu_dereference(priv->peer);
	266	if (unlikely(!peer))
	267	goto out;
	268
	269	peer_priv = netdev_priv(peer);
	270	priv->requested_headroom = new_hr;
	271	new_hr = max(priv->requested_headroom, peer_priv->requested_headroom);
	272	dev->needed_headroom = new_hr;
	273	peer->needed_headroom = new_hr;
	274
	275	out:
	276	rcu_read_unlock();
	277	}
	278
4456e7bd	279	static const struct net_device_ops veth_netdev_ops = {
ee923623 DL	280	.ndo_init = veth_dev_init,
ee923623 DL	281	.ndo_open = veth_open,
2cf48a10	282	.ndo_stop = veth_close,
ee923623	283	.ndo_start_xmit = veth_xmit,
6311cc44	284	.ndo_get_stats64 = veth_get_stats64,
5c70ef85	285	.ndo_set_rx_mode = veth_set_multicast_list,
ee923623	286	.ndo_set_mac_address = eth_mac_addr,
bb446c19 WC	287	#ifdef CONFIG_NET_POLL_CONTROLLER
	288	.ndo_poll_controller = veth_poll_controller,
	289	#endif
a45253bf	290	.ndo_get_iflink = veth_get_iflink,
1a04a821	291	.ndo_features_check = passthru_features_check,
163e5292	292	.ndo_set_rx_headroom = veth_set_rx_headroom,
4456e7bd SH	293	};
4456e7bd SH	294
732912d7	295	#define VETH_FEATURES (NETIF_F_SG \| NETIF_F_FRAGLIST \| NETIF_F_HW_CSUM \| \
c80fafbb	296	NETIF_F_RXCSUM \| NETIF_F_SCTP_CRC \| NETIF_F_HIGHDMA \| \
732912d7	297	NETIF_F_GSO_SOFTWARE \| NETIF_F_GSO_ENCAP_ALL \| \
28d2b136 PM	298	NETIF_F_HW_VLAN_CTAG_TX \| NETIF_F_HW_VLAN_CTAG_RX \| \
28d2b136 PM	299	NETIF_F_HW_VLAN_STAG_TX \| NETIF_F_HW_VLAN_STAG_RX )
8093315a	300
e314dbdc PE	301	static void veth_setup(struct net_device *dev)
	302	{
	303	ether_setup(dev);
	304
550fd08c	305	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
23ea5a96	306	dev->priv_flags \|= IFF_LIVE_ADDR_CHANGE;
02f01ec1	307	dev->priv_flags \|= IFF_NO_QUEUE;
163e5292	308	dev->priv_flags \|= IFF_PHONY_HEADROOM;
550fd08c	309
4456e7bd	310	dev->netdev_ops = &veth_netdev_ops;
e314dbdc PE	311	dev->ethtool_ops = &veth_ethtool_ops;
e314dbdc PE	312	dev->features \|= NETIF_F_LLTX;
8093315a	313	dev->features \|= VETH_FEATURES;
8d0d21f4	314	dev->vlan_features = dev->features &
3f8c707b VY	315	~(NETIF_F_HW_VLAN_CTAG_TX \|
	316	NETIF_F_HW_VLAN_STAG_TX \|
	317	NETIF_F_HW_VLAN_CTAG_RX \|
	318	NETIF_F_HW_VLAN_STAG_RX);
cf124db5 DM	319	dev->needs_free_netdev = true;
cf124db5 DM	320	dev->priv_destructor = veth_dev_free;
91572088	321	dev->max_mtu = ETH_MAX_MTU;
a2c725fa	322
8093315a	323	dev->hw_features = VETH_FEATURES;
82d81898	324	dev->hw_enc_features = VETH_FEATURES;
607fca9a	325	dev->mpls_features = NETIF_F_HW_CSUM \| NETIF_F_GSO_SOFTWARE;
e314dbdc PE	326	}
	327
	328	/*
	329	* netlink interface
	330	*/
	331
a8b8a889 MS	332	static int veth_validate(struct nlattr tb[], struct nlattr data[],
a8b8a889 MS	333	struct netlink_ext_ack *extack)
e314dbdc PE	334	{
	335	if (tb[IFLA_ADDRESS]) {
	336	if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN)
	337	return -EINVAL;
	338	if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS])))
	339	return -EADDRNOTAVAIL;
	340	}
38d40815 EB	341	if (tb[IFLA_MTU]) {
	342	if (!is_valid_veth_mtu(nla_get_u32(tb[IFLA_MTU])))
	343	return -EINVAL;
	344	}
e314dbdc PE	345	return 0;
	346	}
	347
	348	static struct rtnl_link_ops veth_link_ops;
	349
81adee47	350	static int veth_newlink(struct net src_net, struct net_device dev,
7a3f4a18 MS	351	struct nlattr tb[], struct nlattr data[],
7a3f4a18 MS	352	struct netlink_ext_ack *extack)
e314dbdc PE	353	{
	354	int err;
	355	struct net_device *peer;
	356	struct veth_priv *priv;
	357	char ifname[IFNAMSIZ];
	358	struct nlattr peer_tb[IFLA_MAX + 1], *tbp;
5517750f	359	unsigned char name_assign_type;
3729d502	360	struct ifinfomsg *ifmp;
81adee47	361	struct net *net;
e314dbdc PE	362
	363	/*
	364	* create and register peer first
e314dbdc	365	*/
e314dbdc PE	366	if (data != NULL && data[VETH_INFO_PEER] != NULL) {
	367	struct nlattr *nla_peer;
	368
	369	nla_peer = data[VETH_INFO_PEER];
3729d502	370	ifmp = nla_data(nla_peer);
f7b12606 JP	371	err = rtnl_nla_parse_ifla(peer_tb,
f7b12606 JP	372	nla_data(nla_peer) + sizeof(struct ifinfomsg),
fceb6435 JB	373	nla_len(nla_peer) - sizeof(struct ifinfomsg),
fceb6435 JB	374	NULL);
e314dbdc PE	375	if (err < 0)
	376	return err;
	377
a8b8a889	378	err = veth_validate(peer_tb, NULL, extack);
e314dbdc PE	379	if (err < 0)
	380	return err;
	381
	382	tbp = peer_tb;
3729d502 PM	383	} else {
3729d502 PM	384	ifmp = NULL;
e314dbdc	385	tbp = tb;
3729d502	386	}
e314dbdc	387
191cdb38	388	if (ifmp && tbp[IFLA_IFNAME]) {
e314dbdc	389	nla_strlcpy(ifname, tbp[IFLA_IFNAME], IFNAMSIZ);
5517750f TG	390	name_assign_type = NET_NAME_USER;
5517750f TG	391	} else {
e314dbdc	392	snprintf(ifname, IFNAMSIZ, DRV_NAME "%%d");
5517750f TG	393	name_assign_type = NET_NAME_ENUM;
5517750f TG	394	}
e314dbdc	395
81adee47 EB	396	net = rtnl_link_get_net(src_net, tbp);
	397	if (IS_ERR(net))
	398	return PTR_ERR(net);
	399
5517750f TG	400	peer = rtnl_create_link(net, ifname, name_assign_type,
5517750f TG	401	&veth_link_ops, tbp);
81adee47 EB	402	if (IS_ERR(peer)) {
81adee47 EB	403	put_net(net);
e314dbdc	404	return PTR_ERR(peer);
81adee47	405	}
e314dbdc	406
191cdb38	407	if (!ifmp \|\| !tbp[IFLA_ADDRESS])
f2cedb63	408	eth_hw_addr_random(peer);
e6f8f1a7 PE	409
	410	if (ifmp && (dev->ifindex != 0))
	411	peer->ifindex = ifmp->ifi_index;
e314dbdc	412
5cb71d0d SH	413	peer->gso_max_size = dev->gso_max_size;
	414	peer->gso_max_segs = dev->gso_max_segs;
	415
e314dbdc	416	err = register_netdevice(peer);
81adee47 EB	417	put_net(net);
81adee47 EB	418	net = NULL;
e314dbdc PE	419	if (err < 0)
	420	goto err_register_peer;
	421
	422	netif_carrier_off(peer);
	423
3729d502 PM	424	err = rtnl_configure_link(peer, ifmp);
	425	if (err < 0)
	426	goto err_configure_peer;
	427
e314dbdc PE	428	/*
	429	* register dev last
	430	*
	431	* note, that since we've registered new device the dev's name
	432	* should be re-allocated
	433	*/
	434
	435	if (tb[IFLA_ADDRESS] == NULL)
f2cedb63	436	eth_hw_addr_random(dev);
e314dbdc	437
6c8c4446 JP	438	if (tb[IFLA_IFNAME])
	439	nla_strlcpy(dev->name, tb[IFLA_IFNAME], IFNAMSIZ);
	440	else
	441	snprintf(dev->name, IFNAMSIZ, DRV_NAME "%%d");
	442
e314dbdc PE	443	err = register_netdevice(dev);
	444	if (err < 0)
	445	goto err_register_dev;
	446
	447	netif_carrier_off(dev);
	448
	449	/*
	450	* tie the deviced together
	451	*/
	452
	453	priv = netdev_priv(dev);
d0e2c55e	454	rcu_assign_pointer(priv->peer, peer);
e314dbdc PE	455
e314dbdc PE	456	priv = netdev_priv(peer);
d0e2c55e	457	rcu_assign_pointer(priv->peer, dev);
e314dbdc PE	458	return 0;
	459
	460	err_register_dev:
	461	/* nothing to do */
3729d502	462	err_configure_peer:
e314dbdc PE	463	unregister_netdevice(peer);
	464	return err;
	465
	466	err_register_peer:
	467	free_netdev(peer);
	468	return err;
	469	}
	470
23289a37	471	static void veth_dellink(struct net_device dev, struct list_head head)
e314dbdc PE	472	{
	473	struct veth_priv *priv;
	474	struct net_device *peer;
	475
	476	priv = netdev_priv(dev);
d0e2c55e ED	477	peer = rtnl_dereference(priv->peer);
	478
	479	/* Note : dellink() is called from default_device_exit_batch(),
	480	* before a rcu_synchronize() point. The devices are guaranteed
	481	* not being freed before one RCU grace period.
	482	*/
	483	RCU_INIT_POINTER(priv->peer, NULL);
24540535	484	unregister_netdevice_queue(dev, head);
f45a5c26 ED	485
	486	if (peer) {
	487	priv = netdev_priv(peer);
	488	RCU_INIT_POINTER(priv->peer, NULL);
	489	unregister_netdevice_queue(peer, head);
	490	}
e314dbdc PE	491	}
e314dbdc PE	492
23711438 TG	493	static const struct nla_policy veth_policy[VETH_INFO_MAX + 1] = {
	494	[VETH_INFO_PEER] = { .len = sizeof(struct ifinfomsg) },
	495	};
e314dbdc	496
e5f4e7b9 ND	497	static struct net veth_get_link_net(const struct net_device dev)
	498	{
	499	struct veth_priv *priv = netdev_priv(dev);
	500	struct net_device *peer = rtnl_dereference(priv->peer);
	501
	502	return peer ? dev_net(peer) : dev_net(dev);
	503	}
	504
e314dbdc PE	505	static struct rtnl_link_ops veth_link_ops = {
	506	.kind = DRV_NAME,
	507	.priv_size = sizeof(struct veth_priv),
	508	.setup = veth_setup,
	509	.validate = veth_validate,
	510	.newlink = veth_newlink,
	511	.dellink = veth_dellink,
	512	.policy = veth_policy,
	513	.maxtype = VETH_INFO_MAX,
e5f4e7b9	514	.get_link_net = veth_get_link_net,
e314dbdc PE	515	};
	516
	517	/*
	518	* init/fini
	519	*/
	520
	521	static __init int veth_init(void)
	522	{
	523	return rtnl_link_register(&veth_link_ops);
	524	}
	525
	526	static __exit void veth_exit(void)
	527	{
68365458	528	rtnl_link_unregister(&veth_link_ops);
e314dbdc PE	529	}
	530
	531	module_init(veth_init);
	532	module_exit(veth_exit);
	533
	534	MODULE_DESCRIPTION("Virtual Ethernet Tunnel");
	535	MODULE_LICENSE("GPL v2");
	536	MODULE_ALIAS_RTNL_LINK(DRV_NAME);