[linux-block.git] / include / net / ip_tunnels.h

/* SPDX-License-Identifier: GPL-2.0 */
#ifndef __NET_IP_TUNNELS_H
#define __NET_IP_TUNNELS_H 1

#include <linux/if_tunnel.h>
#include <linux/netdevice.h>
#include <linux/skbuff.h>
#include <linux/socket.h>
#include <linux/types.h>
#include <linux/u64_stats_sync.h>
#include <linux/bitops.h>

#include <net/dsfield.h>
#include <net/gro_cells.h>
#include <net/inet_ecn.h>
#include <net/netns/generic.h>
#include <net/rtnetlink.h>
#include <net/lwtunnel.h>
#include <net/dst_cache.h>

#if IS_ENABLED(CONFIG_IPV6)
#include <net/ipv6.h>
#include <net/ip6_fib.h>
#include <net/ip6_route.h>
#endif

/* Keep error state on tunnel for 30 sec */
#define IPTUNNEL_ERR_TIMEO	(30*HZ)

/* Used to memset ip_tunnel padding. */
#define IP_TUNNEL_KEY_SIZE	offsetofend(struct ip_tunnel_key, tp_dst)

/* Used to memset ipv4 address padding. */
#define IP_TUNNEL_KEY_IPV4_PAD	offsetofend(struct ip_tunnel_key, u.ipv4.dst)
#define IP_TUNNEL_KEY_IPV4_PAD_LEN				\
	(sizeof_field(struct ip_tunnel_key, u) -		\
	 sizeof_field(struct ip_tunnel_key, u.ipv4))

struct ip_tunnel_key {
	__be64			tun_id;
	union {
		struct {
			__be32	src;
			__be32	dst;
		} ipv4;
		struct {
			struct in6_addr src;
			struct in6_addr dst;
		} ipv6;
	} u;
	__be16			tun_flags;
	u8			tos;		/* TOS for IPv4, TC for IPv6 */
	u8			ttl;		/* TTL for IPv4, HL for IPv6 */
	__be32			label;		/* Flow Label for IPv6 */
	u32			nhid;
	__be16			tp_src;
	__be16			tp_dst;
	__u8			flow_flags;
};

struct ip_tunnel_encap {
	u16			type;
	u16			flags;
	__be16			sport;
	__be16			dport;
};

/* Flags for ip_tunnel_info mode. */
#define IP_TUNNEL_INFO_TX	0x01	/* represents tx tunnel parameters */
#define IP_TUNNEL_INFO_IPV6	0x02	/* key contains IPv6 addresses */
#define IP_TUNNEL_INFO_BRIDGE	0x04	/* represents a bridged tunnel id */

/* Maximum tunnel options length. */
#define IP_TUNNEL_OPTS_MAX					\
	GENMASK((sizeof_field(struct ip_tunnel_info,		\
			      options_len) * BITS_PER_BYTE) - 1, 0)

#define ip_tunnel_info_opts(info)				\
	_Generic(info,						\
		 const struct ip_tunnel_info * : ((const void *)((info) + 1)),\
		 struct ip_tunnel_info * : ((void *)((info) + 1))\
	)

struct ip_tunnel_info {
	struct ip_tunnel_key	key;
	struct ip_tunnel_encap	encap;
#ifdef CONFIG_DST_CACHE
	struct dst_cache	dst_cache;
#endif
	u8			options_len;
	u8			mode;
};

/* 6rd prefix/relay information */
#ifdef CONFIG_IPV6_SIT_6RD
struct ip_tunnel_6rd_parm {
	struct in6_addr		prefix;
	__be32			relay_prefix;
	u16			prefixlen;
	u16			relay_prefixlen;
};
#endif

struct ip_tunnel_prl_entry {
	struct ip_tunnel_prl_entry __rcu *next;
	__be32				addr;
	u16				flags;
	struct rcu_head			rcu_head;
};

struct metadata_dst;

struct ip_tunnel {
	struct ip_tunnel __rcu	*next;
	struct hlist_node hash_node;

	struct net_device	*dev;
	netdevice_tracker	dev_tracker;

	struct net		*net;	/* netns for packet i/o */

	unsigned long	err_time;	/* Time when the last ICMP error
					 * arrived */
	int		err_count;	/* Number of arrived ICMP errors */

	/* These four fields used only by GRE */
	u32		i_seqno;	/* The last seen seqno	*/
	atomic_t	o_seqno;	/* The last output seqno */
	int		tun_hlen;	/* Precalculated header length */

	/* These four fields used only by ERSPAN */
	u32		index;		/* ERSPAN type II index */
	u8		erspan_ver;	/* ERSPAN version */
	u8		dir;		/* ERSPAN direction */
	u16		hwid;		/* ERSPAN hardware ID */

	struct dst_cache dst_cache;

	struct ip_tunnel_parm parms;

	int		mlink;
	int		encap_hlen;	/* Encap header length (FOU,GUE) */
	int		hlen;		/* tun_hlen + encap_hlen */
	struct ip_tunnel_encap encap;

	/* for SIT */
#ifdef CONFIG_IPV6_SIT_6RD
	struct ip_tunnel_6rd_parm ip6rd;
#endif
	struct ip_tunnel_prl_entry __rcu *prl;	/* potential router list */
	unsigned int		prl_count;	/* # of entries in PRL */
	unsigned int		ip_tnl_net_id;
	struct gro_cells	gro_cells;
	__u32			fwmark;
	bool			collect_md;
	bool			ignore_df;
};

struct tnl_ptk_info {
	__be16 flags;
	__be16 proto;
	__be32 key;
	__be32 seq;
	int hdr_len;
};

#define PACKET_RCVD	0
#define PACKET_REJECT	1
#define PACKET_NEXT	2

#define IP_TNL_HASH_BITS   7
#define IP_TNL_HASH_SIZE   (1 << IP_TNL_HASH_BITS)

struct ip_tunnel_net {
	struct net_device *fb_tunnel_dev;
	struct rtnl_link_ops *rtnl_link_ops;
	struct hlist_head tunnels[IP_TNL_HASH_SIZE];
	struct ip_tunnel __rcu *collect_md_tun;
	int type;
};

static inline void ip_tunnel_key_init(struct ip_tunnel_key *key,
				      __be32 saddr, __be32 daddr,
				      u8 tos, u8 ttl, __be32 label,
				      __be16 tp_src, __be16 tp_dst,
				      __be64 tun_id, __be16 tun_flags)
{
	key->tun_id = tun_id;
	key->u.ipv4.src = saddr;
	key->u.ipv4.dst = daddr;
	memset((unsigned char *)key + IP_TUNNEL_KEY_IPV4_PAD,
	       0, IP_TUNNEL_KEY_IPV4_PAD_LEN);
	key->tos = tos;
	key->ttl = ttl;
	key->label = label;
	key->tun_flags = tun_flags;

	/* For the tunnel types on the top of IPsec, the tp_src and tp_dst of
	 * the upper tunnel are used.
	 * E.g: GRE over IPSEC, the tp_src and tp_port are zero.
	 */
	key->tp_src = tp_src;
	key->tp_dst = tp_dst;

	/* Clear struct padding. */
	if (sizeof(*key) != IP_TUNNEL_KEY_SIZE)
		memset((unsigned char *)key + IP_TUNNEL_KEY_SIZE,
		       0, sizeof(*key) - IP_TUNNEL_KEY_SIZE);
}

static inline bool
ip_tunnel_dst_cache_usable(const struct sk_buff *skb,
			   const struct ip_tunnel_info *info)
{
	if (skb->mark)
		return false;
	if (!info)
		return true;
	if (info->key.tun_flags & TUNNEL_NOCACHE)
		return false;

	return true;
}

static inline unsigned short ip_tunnel_info_af(const struct ip_tunnel_info
					       *tun_info)
{
	return tun_info->mode & IP_TUNNEL_INFO_IPV6 ? AF_INET6 : AF_INET;
}

static inline __be64 key32_to_tunnel_id(__be32 key)
{
#ifdef __BIG_ENDIAN
	return (__force __be64)key;
#else
	return (__force __be64)((__force u64)key << 32);
#endif
}

/* Returns the least-significant 32 bits of a __be64. */
static inline __be32 tunnel_id_to_key32(__be64 tun_id)
{
#ifdef __BIG_ENDIAN
	return (__force __be32)tun_id;
#else
	return (__force __be32)((__force u64)tun_id >> 32);
#endif
}

#ifdef CONFIG_INET

static inline void ip_tunnel_init_flow(struct flowi4 *fl4,
				       int proto,
				       __be32 daddr, __be32 saddr,
				       __be32 key, __u8 tos,
				       struct net *net, int oif,
				       __u32 mark, __u32 tun_inner_hash,
				       __u8 flow_flags)
{
	memset(fl4, 0, sizeof(*fl4));

	if (oif) {
		fl4->flowi4_l3mdev = l3mdev_master_upper_ifindex_by_index_rcu(net, oif);
		/* Legacy VRF/l3mdev use case */
		fl4->flowi4_oif = fl4->flowi4_l3mdev ? 0 : oif;
	}

	fl4->daddr = daddr;
	fl4->saddr = saddr;
	fl4->flowi4_tos = tos;
	fl4->flowi4_proto = proto;
	fl4->fl4_gre_key = key;
	fl4->flowi4_mark = mark;
	fl4->flowi4_multipath_hash = tun_inner_hash;
	fl4->flowi4_flags = flow_flags;
}

int ip_tunnel_init(struct net_device *dev);
void ip_tunnel_uninit(struct net_device *dev);
void  ip_tunnel_dellink(struct net_device *dev, struct list_head *head);
struct net *ip_tunnel_get_link_net(const struct net_device *dev);
int ip_tunnel_get_iflink(const struct net_device *dev);
int ip_tunnel_init_net(struct net *net, unsigned int ip_tnl_net_id,
		       struct rtnl_link_ops *ops, char *devname);

void ip_tunnel_delete_nets(struct list_head *list_net, unsigned int id,
			   struct rtnl_link_ops *ops);

void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
		    const struct iphdr *tnl_params, const u8 protocol);
void ip_md_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
		       const u8 proto, int tunnel_hlen);
int ip_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd);
int ip_tunnel_siocdevprivate(struct net_device *dev, struct ifreq *ifr,
			     void __user *data, int cmd);
int __ip_tunnel_change_mtu(struct net_device *dev, int new_mtu, bool strict);
int ip_tunnel_change_mtu(struct net_device *dev, int new_mtu);

struct ip_tunnel *ip_tunnel_lookup(struct ip_tunnel_net *itn,
				   int link, __be16 flags,
				   __be32 remote, __be32 local,
				   __be32 key);

void ip_tunnel_md_udp_encap(struct sk_buff *skb, struct ip_tunnel_info *info);
int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb,
		  const struct tnl_ptk_info *tpi, struct metadata_dst *tun_dst,
		  bool log_ecn_error);
int ip_tunnel_changelink(struct net_device *dev, struct nlattr *tb[],
			 struct ip_tunnel_parm *p, __u32 fwmark);
int ip_tunnel_newlink(struct net_device *dev, struct nlattr *tb[],
		      struct ip_tunnel_parm *p, __u32 fwmark);
void ip_tunnel_setup(struct net_device *dev, unsigned int net_id);

bool ip_tunnel_netlink_encap_parms(struct nlattr *data[],
				   struct ip_tunnel_encap *encap);

void ip_tunnel_netlink_parms(struct nlattr *data[],
			     struct ip_tunnel_parm *parms);

extern const struct header_ops ip_tunnel_header_ops;
__be16 ip_tunnel_parse_protocol(const struct sk_buff *skb);

struct ip_tunnel_encap_ops {
	size_t (*encap_hlen)(struct ip_tunnel_encap *e);
	int (*build_header)(struct sk_buff *skb, struct ip_tunnel_encap *e,
			    u8 *protocol, struct flowi4 *fl4);
	int (*err_handler)(struct sk_buff *skb, u32 info);
};

#define MAX_IPTUN_ENCAP_OPS 8

extern const struct ip_tunnel_encap_ops __rcu *
		iptun_encaps[MAX_IPTUN_ENCAP_OPS];

int ip_tunnel_encap_add_ops(const struct ip_tunnel_encap_ops *op,
			    unsigned int num);
int ip_tunnel_encap_del_ops(const struct ip_tunnel_encap_ops *op,
			    unsigned int num);

int ip_tunnel_encap_setup(struct ip_tunnel *t,
			  struct ip_tunnel_encap *ipencap);

static inline bool pskb_inet_may_pull(struct sk_buff *skb)
{
	int nhlen;

	switch (skb->protocol) {
#if IS_ENABLED(CONFIG_IPV6)
	case htons(ETH_P_IPV6):
		nhlen = sizeof(struct ipv6hdr);
		break;
#endif
	case htons(ETH_P_IP):
		nhlen = sizeof(struct iphdr);
		break;
	default:
		nhlen = 0;
	}

	return pskb_network_may_pull(skb, nhlen);
}

static inline int ip_encap_hlen(struct ip_tunnel_encap *e)
{
	const struct ip_tunnel_encap_ops *ops;
	int hlen = -EINVAL;

	if (e->type == TUNNEL_ENCAP_NONE)
		return 0;

	if (e->type >= MAX_IPTUN_ENCAP_OPS)
		return -EINVAL;

	rcu_read_lock();
	ops = rcu_dereference(iptun_encaps[e->type]);
	if (likely(ops && ops->encap_hlen))
		hlen = ops->encap_hlen(e);
	rcu_read_unlock();

	return hlen;
}

static inline int ip_tunnel_encap(struct sk_buff *skb,
				  struct ip_tunnel_encap *e,
				  u8 *protocol, struct flowi4 *fl4)
{
	const struct ip_tunnel_encap_ops *ops;
	int ret = -EINVAL;

	if (e->type == TUNNEL_ENCAP_NONE)
		return 0;

	if (e->type >= MAX_IPTUN_ENCAP_OPS)
		return -EINVAL;

	rcu_read_lock();
	ops = rcu_dereference(iptun_encaps[e->type]);
	if (likely(ops && ops->build_header))
		ret = ops->build_header(skb, e, protocol, fl4);
	rcu_read_unlock();

	return ret;
}

/* Extract dsfield from inner protocol */
static inline u8 ip_tunnel_get_dsfield(const struct iphdr *iph,
				       const struct sk_buff *skb)
{
	__be16 payload_protocol = skb_protocol(skb, true);

	if (payload_protocol == htons(ETH_P_IP))
		return iph->tos;
	else if (payload_protocol == htons(ETH_P_IPV6))
		return ipv6_get_dsfield((const struct ipv6hdr *)iph);
	else
		return 0;
}

static inline __be32 ip_tunnel_get_flowlabel(const struct iphdr *iph,
					     const struct sk_buff *skb)
{
	__be16 payload_protocol = skb_protocol(skb, true);

	if (payload_protocol == htons(ETH_P_IPV6))
		return ip6_flowlabel((const struct ipv6hdr *)iph);
	else
		return 0;
}

static inline u8 ip_tunnel_get_ttl(const struct iphdr *iph,
				       const struct sk_buff *skb)
{
	__be16 payload_protocol = skb_protocol(skb, true);

	if (payload_protocol == htons(ETH_P_IP))
		return iph->ttl;
	else if (payload_protocol == htons(ETH_P_IPV6))
		return ((const struct ipv6hdr *)iph)->hop_limit;
	else
		return 0;
}

/* Propogate ECN bits out */
static inline u8 ip_tunnel_ecn_encap(u8 tos, const struct iphdr *iph,
				     const struct sk_buff *skb)
{
	u8 inner = ip_tunnel_get_dsfield(iph, skb);

	return INET_ECN_encapsulate(tos, inner);
}

int __iptunnel_pull_header(struct sk_buff *skb, int hdr_len,
			   __be16 inner_proto, bool raw_proto, bool xnet);

static inline int iptunnel_pull_header(struct sk_buff *skb, int hdr_len,
				       __be16 inner_proto, bool xnet)
{
	return __iptunnel_pull_header(skb, hdr_len, inner_proto, false, xnet);
}

void iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb,
		   __be32 src, __be32 dst, u8 proto,
		   u8 tos, u8 ttl, __be16 df, bool xnet);
struct metadata_dst *iptunnel_metadata_reply(struct metadata_dst *md,
					     gfp_t flags);
int skb_tunnel_check_pmtu(struct sk_buff *skb, struct dst_entry *encap_dst,
			  int headroom, bool reply);

int iptunnel_handle_offloads(struct sk_buff *skb, int gso_type_mask);

static inline int iptunnel_pull_offloads(struct sk_buff *skb)
{
	if (skb_is_gso(skb)) {
		int err;

		err = skb_unclone(skb, GFP_ATOMIC);
		if (unlikely(err))
			return err;
		skb_shinfo(skb)->gso_type &= ~(NETIF_F_GSO_ENCAP_ALL >>
					       NETIF_F_GSO_SHIFT);
	}

	skb->encapsulation = 0;
	return 0;
}

static inline void iptunnel_xmit_stats(struct net_device *dev, int pkt_len)
{
	if (pkt_len > 0) {
		struct pcpu_sw_netstats *tstats = get_cpu_ptr(dev->tstats);

		u64_stats_update_begin(&tstats->syncp);
		u64_stats_add(&tstats->tx_bytes, pkt_len);
		u64_stats_inc(&tstats->tx_packets);
		u64_stats_update_end(&tstats->syncp);
		put_cpu_ptr(tstats);
		return;
	}

	if (pkt_len < 0) {
		DEV_STATS_INC(dev, tx_errors);
		DEV_STATS_INC(dev, tx_aborted_errors);
	} else {
		DEV_STATS_INC(dev, tx_dropped);
	}
}

static inline void ip_tunnel_info_opts_get(void *to,
					   const struct ip_tunnel_info *info)
{
	memcpy(to, info + 1, info->options_len);
}

static inline void ip_tunnel_info_opts_set(struct ip_tunnel_info *info,
					   const void *from, int len,
					   __be16 flags)
{
	info->options_len = len;
	if (len > 0) {
		memcpy(ip_tunnel_info_opts(info), from, len);
		info->key.tun_flags |= flags;
	}
}

static inline struct ip_tunnel_info *lwt_tun_info(struct lwtunnel_state *lwtstate)
{
	return (struct ip_tunnel_info *)lwtstate->data;
}

DECLARE_STATIC_KEY_FALSE(ip_tunnel_metadata_cnt);

/* Returns > 0 if metadata should be collected */
static inline int ip_tunnel_collect_metadata(void)
{
	return static_branch_unlikely(&ip_tunnel_metadata_cnt);
}

void __init ip_tunnel_core_init(void);

void ip_tunnel_need_metadata(void);
void ip_tunnel_unneed_metadata(void);

#else /* CONFIG_INET */

static inline struct ip_tunnel_info *lwt_tun_info(struct lwtunnel_state *lwtstate)
{
	return NULL;
}

static inline void ip_tunnel_need_metadata(void)
{
}

static inline void ip_tunnel_unneed_metadata(void)
{
}

static inline void ip_tunnel_info_opts_get(void *to,
					   const struct ip_tunnel_info *info)
{
}

static inline void ip_tunnel_info_opts_set(struct ip_tunnel_info *info,
					   const void *from, int len,
					   __be16 flags)
{
	info->options_len = 0;
}

#endif /* CONFIG_INET */

#endif /* __NET_IP_TUNNELS_H */
Commit	Line	Data
b2441318	1	/* SPDX-License-Identifier: GPL-2.0 */
c5441932 PS	2	#ifndef __NET_IP_TUNNELS_H
	3	#define __NET_IP_TUNNELS_H 1
	4
	5	#include <linux/if_tunnel.h>
	6	#include <linux/netdevice.h>
	7	#include <linux/skbuff.h>
7f9562a1	8	#include <linux/socket.h>
c5441932 PS	9	#include <linux/types.h>
c5441932 PS	10	#include <linux/u64_stats_sync.h>
fca5fdf6 DB	11	#include <linux/bitops.h>
fca5fdf6 DB	12
c5441932 PS	13	#include <net/dsfield.h>
	14	#include <net/gro_cells.h>
	15	#include <net/inet_ecn.h>
56328486	16	#include <net/netns/generic.h>
c5441932	17	#include <net/rtnetlink.h>
3093fbe7	18	#include <net/lwtunnel.h>
e09acddf	19	#include <net/dst_cache.h>
c5441932 PS	20
	21	#if IS_ENABLED(CONFIG_IPV6)
	22	#include <net/ipv6.h>
	23	#include <net/ip6_fib.h>
	24	#include <net/ip6_route.h>
	25	#endif
	26
	27	/* Keep error state on tunnel for 30 sec */
	28	#define IPTUNNEL_ERR_TIMEO (30*HZ)
	29
1d8fff90	30	/* Used to memset ip_tunnel padding. */
376534a3	31	#define IP_TUNNEL_KEY_SIZE offsetofend(struct ip_tunnel_key, tp_dst)
1d8fff90	32
c1ea5d67 JB	33	/* Used to memset ipv4 address padding. */
	34	#define IP_TUNNEL_KEY_IPV4_PAD offsetofend(struct ip_tunnel_key, u.ipv4.dst)
	35	#define IP_TUNNEL_KEY_IPV4_PAD_LEN \
c593642c PB	36	(sizeof_field(struct ip_tunnel_key, u) - \
c593642c PB	37	sizeof_field(struct ip_tunnel_key, u.ipv4))
c1ea5d67	38
1d8fff90 TG	39	struct ip_tunnel_key {
1d8fff90 TG	40	__be64 tun_id;
c1ea5d67 JB	41	union {
	42	struct {
	43	__be32 src;
	44	__be32 dst;
	45	} ipv4;
	46	struct {
	47	struct in6_addr src;
	48	struct in6_addr dst;
	49	} ipv6;
	50	} u;
1d8fff90	51	__be16 tun_flags;
7c383fb2 JB	52	u8 tos; /* TOS for IPv4, TC for IPv6 */
7c383fb2 JB	53	u8 ttl; /* TTL for IPv4, HL for IPv6 */
13461144	54	__be32 label; /* Flow Label for IPv6 */
8bb5e825	55	u32 nhid;
1d8fff90 TG	56	__be16 tp_src;
1d8fff90 TG	57	__be16 tp_dst;
451ef36b	58	__u8 flow_flags;
ac1cf399	59	};
1d8fff90	60
ac931d4c CE	61	struct ip_tunnel_encap {
	62	u16 type;
	63	u16 flags;
	64	__be16 sport;
	65	__be16 dport;
	66	};
	67
46fa062a JB	68	/* Flags for ip_tunnel_info mode. */
46fa062a JB	69	#define IP_TUNNEL_INFO_TX 0x01 /* represents tx tunnel parameters */
7f9562a1	70	#define IP_TUNNEL_INFO_IPV6 0x02 /* key contains IPv6 addresses */
f35581d6	71	#define IP_TUNNEL_INFO_BRIDGE 0x04 /* represents a bridged tunnel id */
ee122c79	72
fca5fdf6 DB	73	/* Maximum tunnel options length. */
fca5fdf6 DB	74	#define IP_TUNNEL_OPTS_MAX \
c593642c	75	GENMASK((sizeof_field(struct ip_tunnel_info, \
fca5fdf6 DB	76	options_len) * BITS_PER_BYTE) - 1, 0)
fca5fdf6 DB	77
bc9d003d GL	78	#define ip_tunnel_info_opts(info) \
	79	_Generic(info, \
	80	const struct ip_tunnel_info * : ((const void *)((info) + 1)),\
	81	struct ip_tunnel_info * : ((void *)((info) + 1))\
	82	)
	83
1d8fff90 TG	84	struct ip_tunnel_info {
1d8fff90 TG	85	struct ip_tunnel_key key;
ac931d4c	86	struct ip_tunnel_encap encap;
d71785ff PA	87	#ifdef CONFIG_DST_CACHE
	88	struct dst_cache dst_cache;
	89	#endif
1d8fff90	90	u8 options_len;
ee122c79	91	u8 mode;
1d8fff90 TG	92	};
1d8fff90 TG	93
c5441932 PS	94	/* 6rd prefix/relay information */
	95	#ifdef CONFIG_IPV6_SIT_6RD
	96	struct ip_tunnel_6rd_parm {
	97	struct in6_addr prefix;
	98	__be32 relay_prefix;
	99	u16 prefixlen;
	100	u16 relay_prefixlen;
	101	};
	102	#endif
	103
	104	struct ip_tunnel_prl_entry {
	105	struct ip_tunnel_prl_entry __rcu *next;
	106	__be32 addr;
	107	u16 flags;
	108	struct rcu_head rcu_head;
	109	};
	110
2e15ea39 PS	111	struct metadata_dst;
2e15ea39 PS	112
c5441932 PS	113	struct ip_tunnel {
	114	struct ip_tunnel __rcu *next;
	115	struct hlist_node hash_node;
c0fd407a	116
c5441932	117	struct net_device *dev;
c0fd407a ED	118	netdevice_tracker dev_tracker;
c0fd407a ED	119
5e6700b3	120	struct net net; / netns for packet i/o */
c5441932	121
c5441932 PS	122	unsigned long err_time; /* Time when the last ICMP error
c5441932 PS	123	* arrived */
f38ba953	124	int err_count; /* Number of arrived ICMP errors */
c5441932 PS	125
c5441932 PS	126	/* These four fields used only by GRE */
6b8847c5	127	u32 i_seqno; /* The last seen seqno */
31c417c9	128	atomic_t o_seqno; /* The last output seqno */
56328486	129	int tun_hlen; /* Precalculated header length */
c5441932	130
f551c91d	131	/* These four fields used only by ERSPAN */
84e54fe0	132	u32 index; /* ERSPAN type II index */
f551c91d WT	133	u8 erspan_ver; /* ERSPAN version */
	134	u8 dir; /* ERSPAN direction */
	135	u16 hwid; /* ERSPAN hardware ID */
84e54fe0	136
e09acddf	137	struct dst_cache dst_cache;
7d442fab	138
c5441932 PS	139	struct ip_tunnel_parm parms;
c5441932 PS	140
f38ba953	141	int mlink;
56328486	142	int encap_hlen; /* Encap header length (FOU,GUE) */
56328486	143	int hlen; /* tun_hlen + encap_hlen */
f38ba953	144	struct ip_tunnel_encap encap;
56328486	145
c5441932 PS	146	/* for SIT */
	147	#ifdef CONFIG_IPV6_SIT_6RD
	148	struct ip_tunnel_6rd_parm ip6rd;
	149	#endif
	150	struct ip_tunnel_prl_entry __rcu prl; / potential router list */
	151	unsigned int prl_count; /* # of entries in PRL */
c7d03a00	152	unsigned int ip_tnl_net_id;
c5441932	153	struct gro_cells gro_cells;
9830ad4c	154	__u32 fwmark;
2e15ea39	155	bool collect_md;
22a59be8	156	bool ignore_df;
c5441932 PS	157	};
c5441932 PS	158
c5441932 PS	159	struct tnl_ptk_info {
	160	__be16 flags;
	161	__be16 proto;
	162	__be32 key;
	163	__be32 seq;
9b8c6d7b	164	int hdr_len;
c5441932 PS	165	};
	166
	167	#define PACKET_RCVD 0
	168	#define PACKET_REJECT 1
125372fa	169	#define PACKET_NEXT 2
c5441932	170
6261d983	171	#define IP_TNL_HASH_BITS 7
c5441932 PS	172	#define IP_TNL_HASH_SIZE (1 << IP_TNL_HASH_BITS)
	173
	174	struct ip_tunnel_net {
c5441932	175	struct net_device *fb_tunnel_dev;
79134e6c	176	struct rtnl_link_ops *rtnl_link_ops;
6261d983	177	struct hlist_head tunnels[IP_TNL_HASH_SIZE];
2e15ea39	178	struct ip_tunnel __rcu *collect_md_tun;
79134e6c	179	int type;
c5441932 PS	180	};
c5441932 PS	181
4c222798 PS	182	static inline void ip_tunnel_key_init(struct ip_tunnel_key *key,
4c222798 PS	183	__be32 saddr, __be32 daddr,
13461144	184	u8 tos, u8 ttl, __be32 label,
4c222798 PS	185	__be16 tp_src, __be16 tp_dst,
4c222798 PS	186	__be64 tun_id, __be16 tun_flags)
1d8fff90	187	{
4c222798 PS	188	key->tun_id = tun_id;
	189	key->u.ipv4.src = saddr;
	190	key->u.ipv4.dst = daddr;
	191	memset((unsigned char *)key + IP_TUNNEL_KEY_IPV4_PAD,
c1ea5d67	192	0, IP_TUNNEL_KEY_IPV4_PAD_LEN);
4c222798 PS	193	key->tos = tos;
4c222798 PS	194	key->ttl = ttl;
13461144	195	key->label = label;
4c222798	196	key->tun_flags = tun_flags;
1d8fff90 TG	197
	198	/* For the tunnel types on the top of IPsec, the tp_src and tp_dst of
	199	* the upper tunnel are used.
	200	* E.g: GRE over IPSEC, the tp_src and tp_port are zero.
	201	*/
4c222798 PS	202	key->tp_src = tp_src;
4c222798 PS	203	key->tp_dst = tp_dst;
1d8fff90 TG	204
1d8fff90 TG	205	/* Clear struct padding. */
4c222798 PS	206	if (sizeof(*key) != IP_TUNNEL_KEY_SIZE)
	207	memset((unsigned char *)key + IP_TUNNEL_KEY_SIZE,
	208	0, sizeof(*key) - IP_TUNNEL_KEY_SIZE);
1d8fff90 TG	209	}
1d8fff90 TG	210
db3c6139 DB	211	static inline bool
	212	ip_tunnel_dst_cache_usable(const struct sk_buff *skb,
	213	const struct ip_tunnel_info *info)
	214	{
	215	if (skb->mark)
	216	return false;
	217	if (!info)
	218	return true;
	219	if (info->key.tun_flags & TUNNEL_NOCACHE)
	220	return false;
	221
	222	return true;
	223	}
	224
7f9562a1 JB	225	static inline unsigned short ip_tunnel_info_af(const struct ip_tunnel_info
	226	*tun_info)
	227	{
	228	return tun_info->mode & IP_TUNNEL_INFO_IPV6 ? AF_INET6 : AF_INET;
	229	}
	230
d817f432 AV	231	static inline __be64 key32_to_tunnel_id(__be32 key)
	232	{
	233	#ifdef __BIG_ENDIAN
	234	return (__force __be64)key;
	235	#else
	236	return (__force __be64)((__force u64)key << 32);
	237	#endif
	238	}
	239
	240	/* Returns the least-significant 32 bits of a __be64. */
	241	static inline __be32 tunnel_id_to_key32(__be64 tun_id)
	242	{
	243	#ifdef __BIG_ENDIAN
	244	return (__force __be32)tun_id;
	245	#else
	246	return (__force __be32)((__force u64)tun_id >> 32);
	247	#endif
	248	}
	249
5243b6ac JG	250	#ifdef CONFIG_INET
5243b6ac JG	251
b0066da5 PM	252	static inline void ip_tunnel_init_flow(struct flowi4 *fl4,
	253	int proto,
	254	__be32 daddr, __be32 saddr,
db53cd3d DA	255	__be32 key, __u8 tos,
db53cd3d DA	256	struct net *net, int oif,
7ec9fce4 EB	257	__u32 mark, __u32 tun_inner_hash,
7ec9fce4 EB	258	__u8 flow_flags)
b0066da5 PM	259	{
b0066da5 PM	260	memset(fl4, 0, sizeof(*fl4));
db53cd3d DA	261
	262	if (oif) {
	263	fl4->flowi4_l3mdev = l3mdev_master_upper_ifindex_by_index_rcu(net, oif);
	264	/* Legacy VRF/l3mdev use case */
	265	fl4->flowi4_oif = fl4->flowi4_l3mdev ? 0 : oif;
	266	}
	267
b0066da5 PM	268	fl4->daddr = daddr;
	269	fl4->saddr = saddr;
	270	fl4->flowi4_tos = tos;
	271	fl4->flowi4_proto = proto;
	272	fl4->fl4_gre_key = key;
	273	fl4->flowi4_mark = mark;
24ba1440	274	fl4->flowi4_multipath_hash = tun_inner_hash;
7ec9fce4	275	fl4->flowi4_flags = flow_flags;
b0066da5 PM	276	}
b0066da5 PM	277
c5441932 PS	278	int ip_tunnel_init(struct net_device *dev);
	279	void ip_tunnel_uninit(struct net_device *dev);
	280	void ip_tunnel_dellink(struct net_device dev, struct list_head head);
1728d4fa	281	struct net ip_tunnel_get_link_net(const struct net_device dev);
1e99584b	282	int ip_tunnel_get_iflink(const struct net_device *dev);
c7d03a00	283	int ip_tunnel_init_net(struct net *net, unsigned int ip_tnl_net_id,
d3b6f614	284	struct rtnl_link_ops ops, char devname);
c5441932	285
64bc1781 ED	286	void ip_tunnel_delete_nets(struct list_head *list_net, unsigned int id,
64bc1781 ED	287	struct rtnl_link_ops *ops);
c5441932 PS	288
c5441932 PS	289	void ip_tunnel_xmit(struct sk_buff skb, struct net_device dev,
bf3d6a8f	290	const struct iphdr *tnl_params, const u8 protocol);
cfc7381b	291	void ip_md_tunnel_xmit(struct sk_buff skb, struct net_device dev,
c8b34e68	292	const u8 proto, int tunnel_hlen);
607259a6	293	int ip_tunnel_ctl(struct net_device dev, struct ip_tunnel_parm p, int cmd);
3e7a1c7c AB	294	int ip_tunnel_siocdevprivate(struct net_device dev, struct ifreq ifr,
3e7a1c7c AB	295	void __user *data, int cmd);
7e059158	296	int __ip_tunnel_change_mtu(struct net_device *dev, int new_mtu, bool strict);
c5441932 PS	297	int ip_tunnel_change_mtu(struct net_device *dev, int new_mtu);
c5441932 PS	298
c5441932 PS	299	struct ip_tunnel ip_tunnel_lookup(struct ip_tunnel_net itn,
	300	int link, __be16 flags,
	301	__be32 remote, __be32 local,
	302	__be32 key);
	303
ac931d4c	304	void ip_tunnel_md_udp_encap(struct sk_buff skb, struct ip_tunnel_info info);
c5441932	305	int ip_tunnel_rcv(struct ip_tunnel tunnel, struct sk_buff skb,
2e15ea39 PS	306	const struct tnl_ptk_info tpi, struct metadata_dst tun_dst,
2e15ea39 PS	307	bool log_ecn_error);
c5441932	308	int ip_tunnel_changelink(struct net_device dev, struct nlattr tb[],
9830ad4c	309	struct ip_tunnel_parm *p, __u32 fwmark);
c5441932	310	int ip_tunnel_newlink(struct net_device dev, struct nlattr tb[],
9830ad4c	311	struct ip_tunnel_parm *p, __u32 fwmark);
c7d03a00	312	void ip_tunnel_setup(struct net_device *dev, unsigned int net_id);
55c2bc14	313
537dd2d9 LJ	314	bool ip_tunnel_netlink_encap_parms(struct nlattr *data[],
	315	struct ip_tunnel_encap *encap);
	316
b86fca80 LJ	317	void ip_tunnel_netlink_parms(struct nlattr *data[],
	318	struct ip_tunnel_parm *parms);
	319
2606aff9 JD	320	extern const struct header_ops ip_tunnel_header_ops;
	321	__be16 ip_tunnel_parse_protocol(const struct sk_buff *skb);
	322
55c2bc14 TH	323	struct ip_tunnel_encap_ops {
	324	size_t (encap_hlen)(struct ip_tunnel_encap e);
	325	int (build_header)(struct sk_buff skb, struct ip_tunnel_encap *e,
	326	u8 protocol, struct flowi4 fl4);
e7cc0824	327	int (err_handler)(struct sk_buff skb, u32 info);
55c2bc14 TH	328	};
	329
	330	#define MAX_IPTUN_ENCAP_OPS 8
	331
	332	extern const struct ip_tunnel_encap_ops __rcu *
	333	iptun_encaps[MAX_IPTUN_ENCAP_OPS];
	334
	335	int ip_tunnel_encap_add_ops(const struct ip_tunnel_encap_ops *op,
	336	unsigned int num);
	337	int ip_tunnel_encap_del_ops(const struct ip_tunnel_encap_ops *op,
	338	unsigned int num);
	339
56328486 TH	340	int ip_tunnel_encap_setup(struct ip_tunnel *t,
56328486 TH	341	struct ip_tunnel_encap *ipencap);
c5441932	342
cb9f1b78 WB	343	static inline bool pskb_inet_may_pull(struct sk_buff *skb)
	344	{
	345	int nhlen;
	346
	347	switch (skb->protocol) {
	348	#if IS_ENABLED(CONFIG_IPV6)
	349	case htons(ETH_P_IPV6):
	350	nhlen = sizeof(struct ipv6hdr);
	351	break;
	352	#endif
	353	case htons(ETH_P_IP):
	354	nhlen = sizeof(struct iphdr);
	355	break;
	356	default:
	357	nhlen = 0;
	358	}
	359
	360	return pskb_network_may_pull(skb, nhlen);
	361	}
	362
55c2bc14 TH	363	static inline int ip_encap_hlen(struct ip_tunnel_encap *e)
	364	{
	365	const struct ip_tunnel_encap_ops *ops;
	366	int hlen = -EINVAL;
	367
	368	if (e->type == TUNNEL_ENCAP_NONE)
	369	return 0;
	370
	371	if (e->type >= MAX_IPTUN_ENCAP_OPS)
	372	return -EINVAL;
	373
	374	rcu_read_lock();
	375	ops = rcu_dereference(iptun_encaps[e->type]);
	376	if (likely(ops && ops->encap_hlen))
	377	hlen = ops->encap_hlen(e);
	378	rcu_read_unlock();
	379
	380	return hlen;
	381	}
	382
ac931d4c CE	383	static inline int ip_tunnel_encap(struct sk_buff *skb,
ac931d4c CE	384	struct ip_tunnel_encap *e,
55c2bc14 TH	385	u8 protocol, struct flowi4 fl4)
	386	{
	387	const struct ip_tunnel_encap_ops *ops;
	388	int ret = -EINVAL;
	389
ac931d4c	390	if (e->type == TUNNEL_ENCAP_NONE)
55c2bc14 TH	391	return 0;
55c2bc14 TH	392
ac931d4c	393	if (e->type >= MAX_IPTUN_ENCAP_OPS)
55c2bc14 TH	394	return -EINVAL;
	395
	396	rcu_read_lock();
ac931d4c	397	ops = rcu_dereference(iptun_encaps[e->type]);
55c2bc14	398	if (likely(ops && ops->build_header))
ac931d4c	399	ret = ops->build_header(skb, e, protocol, fl4);
55c2bc14 TH	400	rcu_read_unlock();
	401
	402	return ret;
	403	}
	404
c5441932 PS	405	/* Extract dsfield from inner protocol */
	406	static inline u8 ip_tunnel_get_dsfield(const struct iphdr *iph,
	407	const struct sk_buff *skb)
	408	{
7074732c MM	409	__be16 payload_protocol = skb_protocol(skb, true);
	410
	411	if (payload_protocol == htons(ETH_P_IP))
c5441932	412	return iph->tos;
7074732c	413	else if (payload_protocol == htons(ETH_P_IPV6))
c5441932 PS	414	return ipv6_get_dsfield((const struct ipv6hdr *)iph);
	415	else
	416	return 0;
	417	}
	418
c6e9dba3 AL	419	static inline __be32 ip_tunnel_get_flowlabel(const struct iphdr *iph,
	420	const struct sk_buff *skb)
	421	{
	422	__be16 payload_protocol = skb_protocol(skb, true);
	423
	424	if (payload_protocol == htons(ETH_P_IPV6))
	425	return ip6_flowlabel((const struct ipv6hdr *)iph);
	426	else
	427	return 0;
	428	}
	429
72f6d71e HL	430	static inline u8 ip_tunnel_get_ttl(const struct iphdr *iph,
	431	const struct sk_buff *skb)
	432	{
7074732c MM	433	__be16 payload_protocol = skb_protocol(skb, true);
	434
	435	if (payload_protocol == htons(ETH_P_IP))
72f6d71e	436	return iph->ttl;
7074732c	437	else if (payload_protocol == htons(ETH_P_IPV6))
72f6d71e HL	438	return ((const struct ipv6hdr *)iph)->hop_limit;
	439	else
	440	return 0;
	441	}
	442
c5441932 PS	443	/* Propogate ECN bits out */
	444	static inline u8 ip_tunnel_ecn_encap(u8 tos, const struct iphdr *iph,
	445	const struct sk_buff *skb)
	446	{
	447	u8 inner = ip_tunnel_get_dsfield(iph, skb);
	448
	449	return INET_ECN_encapsulate(tos, inner);
	450	}
	451
a6d5bbf3 JB	452	int __iptunnel_pull_header(struct sk_buff *skb, int hdr_len,
	453	__be16 inner_proto, bool raw_proto, bool xnet);
	454
	455	static inline int iptunnel_pull_header(struct sk_buff *skb, int hdr_len,
	456	__be16 inner_proto, bool xnet)
	457	{
	458	return __iptunnel_pull_header(skb, hdr_len, inner_proto, false, xnet);
	459	}
	460
039f5062 PS	461	void iptunnel_xmit(struct sock sk, struct rtable rt, struct sk_buff *skb,
	462	__be32 src, __be32 dst, u8 proto,
	463	u8 tos, u8 ttl, __be16 df, bool xnet);
63d008a4 JB	464	struct metadata_dst iptunnel_metadata_reply(struct metadata_dst md,
63d008a4 JB	465	gfp_t flags);
4cb47a86 SB	466	int skb_tunnel_check_pmtu(struct sk_buff skb, struct dst_entry encap_dst,
4cb47a86 SB	467	int headroom, bool reply);
0e6fbc5b	468
aed069df	469	int iptunnel_handle_offloads(struct sk_buff *skb, int gso_type_mask);
2d26f0a3	470
a09a4c8d JG	471	static inline int iptunnel_pull_offloads(struct sk_buff *skb)
	472	{
	473	if (skb_is_gso(skb)) {
	474	int err;
	475
	476	err = skb_unclone(skb, GFP_ATOMIC);
	477	if (unlikely(err))
	478	return err;
	479	skb_shinfo(skb)->gso_type &= ~(NETIF_F_GSO_ENCAP_ALL >>
	480	NETIF_F_GSO_SHIFT);
	481	}
	482
	483	skb->encapsulation = 0;
	484	return 0;
	485	}
	486
039f5062	487	static inline void iptunnel_xmit_stats(struct net_device *dev, int pkt_len)
c5441932	488	{
039f5062 PS	489	if (pkt_len > 0) {
039f5062 PS	490	struct pcpu_sw_netstats *tstats = get_cpu_ptr(dev->tstats);
c5441932	491
c5441932	492	u64_stats_update_begin(&tstats->syncp);
9962acef ED	493	u64_stats_add(&tstats->tx_bytes, pkt_len);
9962acef ED	494	u64_stats_inc(&tstats->tx_packets);
c5441932	495	u64_stats_update_end(&tstats->syncp);
b4fe85f9	496	put_cpu_ptr(tstats);
9b271eba ED	497	return;
	498	}
	499
	500	if (pkt_len < 0) {
	501	DEV_STATS_INC(dev, tx_errors);
	502	DEV_STATS_INC(dev, tx_aborted_errors);
c5441932	503	} else {
9b271eba	504	DEV_STATS_INC(dev, tx_dropped);
c5441932 PS	505	}
c5441932 PS	506	}
5243b6ac	507
4c222798 PS	508	static inline void ip_tunnel_info_opts_get(void *to,
	509	const struct ip_tunnel_info *info)
	510	{
	511	memcpy(to, info + 1, info->options_len);
	512	}
	513
	514	static inline void ip_tunnel_info_opts_set(struct ip_tunnel_info *info,
256c87c1 PJV	515	const void *from, int len,
256c87c1 PJV	516	__be16 flags)
4c222798	517	{
4c222798	518	info->options_len = len;
9c2e14b4 YHW	519	if (len > 0) {
	520	memcpy(ip_tunnel_info_opts(info), from, len);
	521	info->key.tun_flags \|= flags;
	522	}
4c222798 PS	523	}
4c222798 PS	524
3093fbe7 TG	525	static inline struct ip_tunnel_info lwt_tun_info(struct lwtunnel_state lwtstate)
	526	{
	527	return (struct ip_tunnel_info *)lwtstate->data;
	528	}
	529
5263a98f	530	DECLARE_STATIC_KEY_FALSE(ip_tunnel_metadata_cnt);
e7030878 TG	531
	532	/* Returns > 0 if metadata should be collected */
	533	static inline int ip_tunnel_collect_metadata(void)
	534	{
5263a98f	535	return static_branch_unlikely(&ip_tunnel_metadata_cnt);
e7030878 TG	536	}
e7030878 TG	537
045a0fa0 TG	538	void __init ip_tunnel_core_init(void);
045a0fa0 TG	539
e7030878 TG	540	void ip_tunnel_need_metadata(void);
	541	void ip_tunnel_unneed_metadata(void);
	542
05283187 TG	543	#else /* CONFIG_INET */
	544
	545	static inline struct ip_tunnel_info lwt_tun_info(struct lwtunnel_state lwtstate)
	546	{
	547	return NULL;
	548	}
	549
	550	static inline void ip_tunnel_need_metadata(void)
	551	{
	552	}
	553
	554	static inline void ip_tunnel_unneed_metadata(void)
	555	{
	556	}
	557
e28e87ed DB	558	static inline void ip_tunnel_info_opts_get(void *to,
	559	const struct ip_tunnel_info *info)
	560	{
	561	}
	562
	563	static inline void ip_tunnel_info_opts_set(struct ip_tunnel_info *info,
256c87c1 PJV	564	const void *from, int len,
256c87c1 PJV	565	__be16 flags)
e28e87ed DB	566	{
	567	info->options_len = 0;
	568	}
	569
5243b6ac JG	570	#endif /* CONFIG_INET */
5243b6ac JG	571
c5441932	572	#endif /* __NET_IP_TUNNELS_H */