[linux-2.6-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))

#define __ipt_flag_op(op, ...)					\
	op(__VA_ARGS__, __IP_TUNNEL_FLAG_NUM)

#define IP_TUNNEL_DECLARE_FLAGS(...)				\
	__ipt_flag_op(DECLARE_BITMAP, __VA_ARGS__)

#define ip_tunnel_flags_zero(...)	__ipt_flag_op(bitmap_zero, __VA_ARGS__)
#define ip_tunnel_flags_copy(...)	__ipt_flag_op(bitmap_copy, __VA_ARGS__)
#define ip_tunnel_flags_and(...)	__ipt_flag_op(bitmap_and, __VA_ARGS__)
#define ip_tunnel_flags_or(...)		__ipt_flag_op(bitmap_or, __VA_ARGS__)

#define ip_tunnel_flags_empty(...)				\
	__ipt_flag_op(bitmap_empty, __VA_ARGS__)
#define ip_tunnel_flags_intersect(...)				\
	__ipt_flag_op(bitmap_intersects, __VA_ARGS__)
#define ip_tunnel_flags_subset(...)				\
	__ipt_flag_op(bitmap_subset, __VA_ARGS__)

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;
	IP_TUNNEL_DECLARE_FLAGS(tun_flags);
	__be32			label;		/* Flow Label for IPv6 */
	u32			nhid;
	u8			tos;		/* TOS for IPv4, TC for IPv6 */
	u8			ttl;		/* TTL for IPv4, HL for IPv6 */
	__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;

/* Kernel-side variant of ip_tunnel_parm */
struct ip_tunnel_parm_kern {
	char			name[IFNAMSIZ];
	IP_TUNNEL_DECLARE_FLAGS(i_flags);
	IP_TUNNEL_DECLARE_FLAGS(o_flags);
	__be32			i_key;
	__be32			o_key;
	int			link;
	struct iphdr		iph;
};

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_kern 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 {
	IP_TUNNEL_DECLARE_FLAGS(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_set_options_present(unsigned long *flags)
{
	IP_TUNNEL_DECLARE_FLAGS(present) = { };

	__set_bit(IP_TUNNEL_GENEVE_OPT_BIT, present);
	__set_bit(IP_TUNNEL_VXLAN_OPT_BIT, present);
	__set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, present);
	__set_bit(IP_TUNNEL_GTP_OPT_BIT, present);
	__set_bit(IP_TUNNEL_PFCP_OPT_BIT, present);

	ip_tunnel_flags_or(flags, flags, present);
}

static inline void ip_tunnel_clear_options_present(unsigned long *flags)
{
	IP_TUNNEL_DECLARE_FLAGS(present) = { };

	__set_bit(IP_TUNNEL_GENEVE_OPT_BIT, present);
	__set_bit(IP_TUNNEL_VXLAN_OPT_BIT, present);
	__set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, present);
	__set_bit(IP_TUNNEL_GTP_OPT_BIT, present);
	__set_bit(IP_TUNNEL_PFCP_OPT_BIT, present);

	__ipt_flag_op(bitmap_andnot, flags, flags, present);
}

static inline bool ip_tunnel_is_options_present(const unsigned long *flags)
{
	IP_TUNNEL_DECLARE_FLAGS(present) = { };

	__set_bit(IP_TUNNEL_GENEVE_OPT_BIT, present);
	__set_bit(IP_TUNNEL_VXLAN_OPT_BIT, present);
	__set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, present);
	__set_bit(IP_TUNNEL_GTP_OPT_BIT, present);
	__set_bit(IP_TUNNEL_PFCP_OPT_BIT, present);

	return ip_tunnel_flags_intersect(flags, present);
}

static inline bool ip_tunnel_flags_is_be16_compat(const unsigned long *flags)
{
	IP_TUNNEL_DECLARE_FLAGS(supp) = { };

	bitmap_set(supp, 0, BITS_PER_TYPE(__be16));
	__set_bit(IP_TUNNEL_VTI_BIT, supp);

	return ip_tunnel_flags_subset(flags, supp);
}

static inline void ip_tunnel_flags_from_be16(unsigned long *dst, __be16 flags)
{
	ip_tunnel_flags_zero(dst);

	bitmap_write(dst, be16_to_cpu(flags), 0, BITS_PER_TYPE(__be16));
	__assign_bit(IP_TUNNEL_VTI_BIT, dst, flags & VTI_ISVTI);
}

static inline __be16 ip_tunnel_flags_to_be16(const unsigned long *flags)
{
	__be16 ret;

	ret = cpu_to_be16(bitmap_read(flags, 0, BITS_PER_TYPE(__be16)));
	if (test_bit(IP_TUNNEL_VTI_BIT, flags))
		ret |= VTI_ISVTI;

	return ret;
}

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,
				      const unsigned long *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;
	ip_tunnel_flags_copy(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;

	return !info || !test_bit(IP_TUNNEL_NOCACHE_BIT, info->key.tun_flags);
}

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,
			   struct list_head *dev_to_kill);

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_kern *p,
		  int cmd);
bool ip_tunnel_parm_from_user(struct ip_tunnel_parm_kern *kp,
			      const void __user *data);
bool ip_tunnel_parm_to_user(void __user *data, struct ip_tunnel_parm_kern *kp);
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, const unsigned long *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_kern *p, __u32 fwmark);
int ip_tunnel_newlink(struct net_device *dev, struct nlattr *tb[],
		      struct ip_tunnel_parm_kern *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_kern *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);
}

/* Variant of pskb_inet_may_pull().
 */
static inline bool skb_vlan_inet_prepare(struct sk_buff *skb,
					 bool inner_proto_inherit)
{
	int nhlen = 0, maclen = inner_proto_inherit ? 0 : ETH_HLEN;
	__be16 type = skb->protocol;

	/* Essentially this is skb_protocol(skb, true)
	 * And we get MAC len.
	 */
	if (eth_type_vlan(type))
		type = __vlan_get_protocol(skb, type, &maclen);

	switch (type) {
#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;
	}
	/* For ETH_P_IPV6/ETH_P_IP we make sure to pull
	 * a base network header in skb->head.
	 */
	if (!pskb_may_pull(skb, maclen + nhlen))
		return false;

	skb_set_network_header(skb, maclen);
	return true;
}

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;
}

/* Propagate 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,
					   const unsigned long *flags)
{
	info->options_len = len;
	if (len > 0) {
		memcpy(ip_tunnel_info_opts(info), from, len);
		ip_tunnel_flags_or(info->key.tun_flags, 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,
					   const unsigned long *flags)
{
	info->options_len = 0;
}

#endif /* CONFIG_INET */

#endif /* __NET_IP_TUNNELS_H */
Commit	Line	Data
	1	/* SPDX-License-Identifier: GPL-2.0 */
	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>
	8	#include <linux/socket.h>
	9	#include <linux/types.h>
	10	#include <linux/u64_stats_sync.h>
	11	#include <linux/bitops.h>
	12
	13	#include <net/dsfield.h>
	14	#include <net/gro_cells.h>
	15	#include <net/inet_ecn.h>
	16	#include <net/netns/generic.h>
	17	#include <net/rtnetlink.h>
	18	#include <net/lwtunnel.h>
	19	#include <net/dst_cache.h>
	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
	30	/* Used to memset ip_tunnel padding. */
	31	#define IP_TUNNEL_KEY_SIZE offsetofend(struct ip_tunnel_key, tp_dst)
	32
	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 \
	36	(sizeof_field(struct ip_tunnel_key, u) - \
	37	sizeof_field(struct ip_tunnel_key, u.ipv4))
	38
	39	#define __ipt_flag_op(op, ...) \
	40	op(__VA_ARGS__, __IP_TUNNEL_FLAG_NUM)
	41
	42	#define IP_TUNNEL_DECLARE_FLAGS(...) \
	43	__ipt_flag_op(DECLARE_BITMAP, __VA_ARGS__)
	44
	45	#define ip_tunnel_flags_zero(...) __ipt_flag_op(bitmap_zero, __VA_ARGS__)
	46	#define ip_tunnel_flags_copy(...) __ipt_flag_op(bitmap_copy, __VA_ARGS__)
	47	#define ip_tunnel_flags_and(...) __ipt_flag_op(bitmap_and, __VA_ARGS__)
	48	#define ip_tunnel_flags_or(...) __ipt_flag_op(bitmap_or, __VA_ARGS__)
	49
	50	#define ip_tunnel_flags_empty(...) \
	51	__ipt_flag_op(bitmap_empty, __VA_ARGS__)
	52	#define ip_tunnel_flags_intersect(...) \
	53	__ipt_flag_op(bitmap_intersects, __VA_ARGS__)
	54	#define ip_tunnel_flags_subset(...) \
	55	__ipt_flag_op(bitmap_subset, __VA_ARGS__)
	56
	57	struct ip_tunnel_key {
	58	__be64 tun_id;
	59	union {
	60	struct {
	61	__be32 src;
	62	__be32 dst;
	63	} ipv4;
	64	struct {
	65	struct in6_addr src;
	66	struct in6_addr dst;
	67	} ipv6;
	68	} u;
	69	IP_TUNNEL_DECLARE_FLAGS(tun_flags);
	70	__be32 label; /* Flow Label for IPv6 */
	71	u32 nhid;
	72	u8 tos; /* TOS for IPv4, TC for IPv6 */
	73	u8 ttl; /* TTL for IPv4, HL for IPv6 */
	74	__be16 tp_src;
	75	__be16 tp_dst;
	76	__u8 flow_flags;
	77	};
	78
	79	struct ip_tunnel_encap {
	80	u16 type;
	81	u16 flags;
	82	__be16 sport;
	83	__be16 dport;
	84	};
	85
	86	/* Flags for ip_tunnel_info mode. */
	87	#define IP_TUNNEL_INFO_TX 0x01 /* represents tx tunnel parameters */
	88	#define IP_TUNNEL_INFO_IPV6 0x02 /* key contains IPv6 addresses */
	89	#define IP_TUNNEL_INFO_BRIDGE 0x04 /* represents a bridged tunnel id */
	90
	91	/* Maximum tunnel options length. */
	92	#define IP_TUNNEL_OPTS_MAX \
	93	GENMASK((sizeof_field(struct ip_tunnel_info, \
	94	options_len) * BITS_PER_BYTE) - 1, 0)
	95
	96	#define ip_tunnel_info_opts(info) \
	97	_Generic(info, \
	98	const struct ip_tunnel_info * : ((const void *)((info) + 1)),\
	99	struct ip_tunnel_info * : ((void *)((info) + 1))\
	100	)
	101
	102	struct ip_tunnel_info {
	103	struct ip_tunnel_key key;
	104	struct ip_tunnel_encap encap;
	105	#ifdef CONFIG_DST_CACHE
	106	struct dst_cache dst_cache;
	107	#endif
	108	u8 options_len;
	109	u8 mode;
	110	};
	111
	112	/* 6rd prefix/relay information */
	113	#ifdef CONFIG_IPV6_SIT_6RD
	114	struct ip_tunnel_6rd_parm {
	115	struct in6_addr prefix;
	116	__be32 relay_prefix;
	117	u16 prefixlen;
	118	u16 relay_prefixlen;
	119	};
	120	#endif
	121
	122	struct ip_tunnel_prl_entry {
	123	struct ip_tunnel_prl_entry __rcu *next;
	124	__be32 addr;
	125	u16 flags;
	126	struct rcu_head rcu_head;
	127	};
	128
	129	struct metadata_dst;
	130
	131	/* Kernel-side variant of ip_tunnel_parm */
	132	struct ip_tunnel_parm_kern {
	133	char name[IFNAMSIZ];
	134	IP_TUNNEL_DECLARE_FLAGS(i_flags);
	135	IP_TUNNEL_DECLARE_FLAGS(o_flags);
	136	__be32 i_key;
	137	__be32 o_key;
	138	int link;
	139	struct iphdr iph;
	140	};
	141
	142	struct ip_tunnel {
	143	struct ip_tunnel __rcu *next;
	144	struct hlist_node hash_node;
	145
	146	struct net_device *dev;
	147	netdevice_tracker dev_tracker;
	148
	149	struct net net; / netns for packet i/o */
	150
	151	unsigned long err_time; /* Time when the last ICMP error
	152	* arrived */
	153	int err_count; /* Number of arrived ICMP errors */
	154
	155	/* These four fields used only by GRE */
	156	u32 i_seqno; /* The last seen seqno */
	157	atomic_t o_seqno; /* The last output seqno */
	158	int tun_hlen; /* Precalculated header length */
	159
	160	/* These four fields used only by ERSPAN */
	161	u32 index; /* ERSPAN type II index */
	162	u8 erspan_ver; /* ERSPAN version */
	163	u8 dir; /* ERSPAN direction */
	164	u16 hwid; /* ERSPAN hardware ID */
	165
	166	struct dst_cache dst_cache;
	167
	168	struct ip_tunnel_parm_kern parms;
	169
	170	int mlink;
	171	int encap_hlen; /* Encap header length (FOU,GUE) */
	172	int hlen; /* tun_hlen + encap_hlen */
	173	struct ip_tunnel_encap encap;
	174
	175	/* for SIT */
	176	#ifdef CONFIG_IPV6_SIT_6RD
	177	struct ip_tunnel_6rd_parm ip6rd;
	178	#endif
	179	struct ip_tunnel_prl_entry __rcu prl; / potential router list */
	180	unsigned int prl_count; /* # of entries in PRL */
	181	unsigned int ip_tnl_net_id;
	182	struct gro_cells gro_cells;
	183	__u32 fwmark;
	184	bool collect_md;
	185	bool ignore_df;
	186	};
	187
	188	struct tnl_ptk_info {
	189	IP_TUNNEL_DECLARE_FLAGS(flags);
	190	__be16 proto;
	191	__be32 key;
	192	__be32 seq;
	193	int hdr_len;
	194	};
	195
	196	#define PACKET_RCVD 0
	197	#define PACKET_REJECT 1
	198	#define PACKET_NEXT 2
	199
	200	#define IP_TNL_HASH_BITS 7
	201	#define IP_TNL_HASH_SIZE (1 << IP_TNL_HASH_BITS)
	202
	203	struct ip_tunnel_net {
	204	struct net_device *fb_tunnel_dev;
	205	struct rtnl_link_ops *rtnl_link_ops;
	206	struct hlist_head tunnels[IP_TNL_HASH_SIZE];
	207	struct ip_tunnel __rcu *collect_md_tun;
	208	int type;
	209	};
	210
	211	static inline void ip_tunnel_set_options_present(unsigned long *flags)
	212	{
	213	IP_TUNNEL_DECLARE_FLAGS(present) = { };
	214
	215	__set_bit(IP_TUNNEL_GENEVE_OPT_BIT, present);
	216	__set_bit(IP_TUNNEL_VXLAN_OPT_BIT, present);
	217	__set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, present);
	218	__set_bit(IP_TUNNEL_GTP_OPT_BIT, present);
	219	__set_bit(IP_TUNNEL_PFCP_OPT_BIT, present);
	220
	221	ip_tunnel_flags_or(flags, flags, present);
	222	}
	223
	224	static inline void ip_tunnel_clear_options_present(unsigned long *flags)
	225	{
	226	IP_TUNNEL_DECLARE_FLAGS(present) = { };
	227
	228	__set_bit(IP_TUNNEL_GENEVE_OPT_BIT, present);
	229	__set_bit(IP_TUNNEL_VXLAN_OPT_BIT, present);
	230	__set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, present);
	231	__set_bit(IP_TUNNEL_GTP_OPT_BIT, present);
	232	__set_bit(IP_TUNNEL_PFCP_OPT_BIT, present);
	233
	234	__ipt_flag_op(bitmap_andnot, flags, flags, present);
	235	}
	236
	237	static inline bool ip_tunnel_is_options_present(const unsigned long *flags)
	238	{
	239	IP_TUNNEL_DECLARE_FLAGS(present) = { };
	240
	241	__set_bit(IP_TUNNEL_GENEVE_OPT_BIT, present);
	242	__set_bit(IP_TUNNEL_VXLAN_OPT_BIT, present);
	243	__set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, present);
	244	__set_bit(IP_TUNNEL_GTP_OPT_BIT, present);
	245	__set_bit(IP_TUNNEL_PFCP_OPT_BIT, present);
	246
	247	return ip_tunnel_flags_intersect(flags, present);
	248	}
	249
	250	static inline bool ip_tunnel_flags_is_be16_compat(const unsigned long *flags)
	251	{
	252	IP_TUNNEL_DECLARE_FLAGS(supp) = { };
	253
	254	bitmap_set(supp, 0, BITS_PER_TYPE(__be16));
	255	__set_bit(IP_TUNNEL_VTI_BIT, supp);
	256
	257	return ip_tunnel_flags_subset(flags, supp);
	258	}
	259
	260	static inline void ip_tunnel_flags_from_be16(unsigned long *dst, __be16 flags)
	261	{
	262	ip_tunnel_flags_zero(dst);
	263
	264	bitmap_write(dst, be16_to_cpu(flags), 0, BITS_PER_TYPE(__be16));
	265	__assign_bit(IP_TUNNEL_VTI_BIT, dst, flags & VTI_ISVTI);
	266	}
	267
	268	static inline __be16 ip_tunnel_flags_to_be16(const unsigned long *flags)
	269	{
	270	__be16 ret;
	271
	272	ret = cpu_to_be16(bitmap_read(flags, 0, BITS_PER_TYPE(__be16)));
	273	if (test_bit(IP_TUNNEL_VTI_BIT, flags))
	274	ret \|= VTI_ISVTI;
	275
	276	return ret;
	277	}
	278
	279	static inline void ip_tunnel_key_init(struct ip_tunnel_key *key,
	280	__be32 saddr, __be32 daddr,
	281	u8 tos, u8 ttl, __be32 label,
	282	__be16 tp_src, __be16 tp_dst,
	283	__be64 tun_id,
	284	const unsigned long *tun_flags)
	285	{
	286	key->tun_id = tun_id;
	287	key->u.ipv4.src = saddr;
	288	key->u.ipv4.dst = daddr;
	289	memset((unsigned char *)key + IP_TUNNEL_KEY_IPV4_PAD,
	290	0, IP_TUNNEL_KEY_IPV4_PAD_LEN);
	291	key->tos = tos;
	292	key->ttl = ttl;
	293	key->label = label;
	294	ip_tunnel_flags_copy(key->tun_flags, tun_flags);
	295
	296	/* For the tunnel types on the top of IPsec, the tp_src and tp_dst of
	297	* the upper tunnel are used.
	298	* E.g: GRE over IPSEC, the tp_src and tp_port are zero.
	299	*/
	300	key->tp_src = tp_src;
	301	key->tp_dst = tp_dst;
	302
	303	/* Clear struct padding. */
	304	if (sizeof(*key) != IP_TUNNEL_KEY_SIZE)
	305	memset((unsigned char *)key + IP_TUNNEL_KEY_SIZE,
	306	0, sizeof(*key) - IP_TUNNEL_KEY_SIZE);
	307	}
	308
	309	static inline bool
	310	ip_tunnel_dst_cache_usable(const struct sk_buff *skb,
	311	const struct ip_tunnel_info *info)
	312	{
	313	if (skb->mark)
	314	return false;
	315
	316	return !info \|\| !test_bit(IP_TUNNEL_NOCACHE_BIT, info->key.tun_flags);
	317	}
	318
	319	static inline unsigned short ip_tunnel_info_af(const struct ip_tunnel_info
	320	*tun_info)
	321	{
	322	return tun_info->mode & IP_TUNNEL_INFO_IPV6 ? AF_INET6 : AF_INET;
	323	}
	324
	325	static inline __be64 key32_to_tunnel_id(__be32 key)
	326	{
	327	#ifdef __BIG_ENDIAN
	328	return (__force __be64)key;
	329	#else
	330	return (__force __be64)((__force u64)key << 32);
	331	#endif
	332	}
	333
	334	/* Returns the least-significant 32 bits of a __be64. */
	335	static inline __be32 tunnel_id_to_key32(__be64 tun_id)
	336	{
	337	#ifdef __BIG_ENDIAN
	338	return (__force __be32)tun_id;
	339	#else
	340	return (__force __be32)((__force u64)tun_id >> 32);
	341	#endif
	342	}
	343
	344	#ifdef CONFIG_INET
	345
	346	static inline void ip_tunnel_init_flow(struct flowi4 *fl4,
	347	int proto,
	348	__be32 daddr, __be32 saddr,
	349	__be32 key, __u8 tos,
	350	struct net *net, int oif,
	351	__u32 mark, __u32 tun_inner_hash,
	352	__u8 flow_flags)
	353	{
	354	memset(fl4, 0, sizeof(*fl4));
	355
	356	if (oif) {
	357	fl4->flowi4_l3mdev = l3mdev_master_upper_ifindex_by_index_rcu(net, oif);
	358	/* Legacy VRF/l3mdev use case */
	359	fl4->flowi4_oif = fl4->flowi4_l3mdev ? 0 : oif;
	360	}
	361
	362	fl4->daddr = daddr;
	363	fl4->saddr = saddr;
	364	fl4->flowi4_tos = tos;
	365	fl4->flowi4_proto = proto;
	366	fl4->fl4_gre_key = key;
	367	fl4->flowi4_mark = mark;
	368	fl4->flowi4_multipath_hash = tun_inner_hash;
	369	fl4->flowi4_flags = flow_flags;
	370	}
	371
	372	int ip_tunnel_init(struct net_device *dev);
	373	void ip_tunnel_uninit(struct net_device *dev);
	374	void ip_tunnel_dellink(struct net_device dev, struct list_head head);
	375	struct net ip_tunnel_get_link_net(const struct net_device dev);
	376	int ip_tunnel_get_iflink(const struct net_device *dev);
	377	int ip_tunnel_init_net(struct net *net, unsigned int ip_tnl_net_id,
	378	struct rtnl_link_ops ops, char devname);
	379
	380	void ip_tunnel_delete_nets(struct list_head *list_net, unsigned int id,
	381	struct rtnl_link_ops *ops,
	382	struct list_head *dev_to_kill);
	383
	384	void ip_tunnel_xmit(struct sk_buff skb, struct net_device dev,
	385	const struct iphdr *tnl_params, const u8 protocol);
	386	void ip_md_tunnel_xmit(struct sk_buff skb, struct net_device dev,
	387	const u8 proto, int tunnel_hlen);
	388	int ip_tunnel_ctl(struct net_device dev, struct ip_tunnel_parm_kern p,
	389	int cmd);
	390	bool ip_tunnel_parm_from_user(struct ip_tunnel_parm_kern *kp,
	391	const void __user *data);
	392	bool ip_tunnel_parm_to_user(void __user data, struct ip_tunnel_parm_kern kp);
	393	int ip_tunnel_siocdevprivate(struct net_device dev, struct ifreq ifr,
	394	void __user *data, int cmd);
	395	int __ip_tunnel_change_mtu(struct net_device *dev, int new_mtu, bool strict);
	396	int ip_tunnel_change_mtu(struct net_device *dev, int new_mtu);
	397
	398	struct ip_tunnel ip_tunnel_lookup(struct ip_tunnel_net itn,
	399	int link, const unsigned long *flags,
	400	__be32 remote, __be32 local,
	401	__be32 key);
	402
	403	void ip_tunnel_md_udp_encap(struct sk_buff skb, struct ip_tunnel_info info);
	404	int ip_tunnel_rcv(struct ip_tunnel tunnel, struct sk_buff skb,
	405	const struct tnl_ptk_info tpi, struct metadata_dst tun_dst,
	406	bool log_ecn_error);
	407	int ip_tunnel_changelink(struct net_device dev, struct nlattr tb[],
	408	struct ip_tunnel_parm_kern *p, __u32 fwmark);
	409	int ip_tunnel_newlink(struct net_device dev, struct nlattr tb[],
	410	struct ip_tunnel_parm_kern *p, __u32 fwmark);
	411	void ip_tunnel_setup(struct net_device *dev, unsigned int net_id);
	412
	413	bool ip_tunnel_netlink_encap_parms(struct nlattr *data[],
	414	struct ip_tunnel_encap *encap);
	415
	416	void ip_tunnel_netlink_parms(struct nlattr *data[],
	417	struct ip_tunnel_parm_kern *parms);
	418
	419	extern const struct header_ops ip_tunnel_header_ops;
	420	__be16 ip_tunnel_parse_protocol(const struct sk_buff *skb);
	421
	422	struct ip_tunnel_encap_ops {
	423	size_t (encap_hlen)(struct ip_tunnel_encap e);
	424	int (build_header)(struct sk_buff skb, struct ip_tunnel_encap *e,
	425	u8 protocol, struct flowi4 fl4);
	426	int (err_handler)(struct sk_buff skb, u32 info);
	427	};
	428
	429	#define MAX_IPTUN_ENCAP_OPS 8
	430
	431	extern const struct ip_tunnel_encap_ops __rcu *
	432	iptun_encaps[MAX_IPTUN_ENCAP_OPS];
	433
	434	int ip_tunnel_encap_add_ops(const struct ip_tunnel_encap_ops *op,
	435	unsigned int num);
	436	int ip_tunnel_encap_del_ops(const struct ip_tunnel_encap_ops *op,
	437	unsigned int num);
	438
	439	int ip_tunnel_encap_setup(struct ip_tunnel *t,
	440	struct ip_tunnel_encap *ipencap);
	441
	442	static inline bool pskb_inet_may_pull(struct sk_buff *skb)
	443	{
	444	int nhlen;
	445
	446	switch (skb->protocol) {
	447	#if IS_ENABLED(CONFIG_IPV6)
	448	case htons(ETH_P_IPV6):
	449	nhlen = sizeof(struct ipv6hdr);
	450	break;
	451	#endif
	452	case htons(ETH_P_IP):
	453	nhlen = sizeof(struct iphdr);
	454	break;
	455	default:
	456	nhlen = 0;
	457	}
	458
	459	return pskb_network_may_pull(skb, nhlen);
	460	}
	461
	462	/* Variant of pskb_inet_may_pull().
	463	*/
	464	static inline bool skb_vlan_inet_prepare(struct sk_buff *skb,
	465	bool inner_proto_inherit)
	466	{
	467	int nhlen = 0, maclen = inner_proto_inherit ? 0 : ETH_HLEN;
	468	__be16 type = skb->protocol;
	469
	470	/* Essentially this is skb_protocol(skb, true)
	471	* And we get MAC len.
	472	*/
	473	if (eth_type_vlan(type))
	474	type = __vlan_get_protocol(skb, type, &maclen);
	475
	476	switch (type) {
	477	#if IS_ENABLED(CONFIG_IPV6)
	478	case htons(ETH_P_IPV6):
	479	nhlen = sizeof(struct ipv6hdr);
	480	break;
	481	#endif
	482	case htons(ETH_P_IP):
	483	nhlen = sizeof(struct iphdr);
	484	break;
	485	}
	486	/* For ETH_P_IPV6/ETH_P_IP we make sure to pull
	487	* a base network header in skb->head.
	488	*/
	489	if (!pskb_may_pull(skb, maclen + nhlen))
	490	return false;
	491
	492	skb_set_network_header(skb, maclen);
	493	return true;
	494	}
	495
	496	static inline int ip_encap_hlen(struct ip_tunnel_encap *e)
	497	{
	498	const struct ip_tunnel_encap_ops *ops;
	499	int hlen = -EINVAL;
	500
	501	if (e->type == TUNNEL_ENCAP_NONE)
	502	return 0;
	503
	504	if (e->type >= MAX_IPTUN_ENCAP_OPS)
	505	return -EINVAL;
	506
	507	rcu_read_lock();
	508	ops = rcu_dereference(iptun_encaps[e->type]);
	509	if (likely(ops && ops->encap_hlen))
	510	hlen = ops->encap_hlen(e);
	511	rcu_read_unlock();
	512
	513	return hlen;
	514	}
	515
	516	static inline int ip_tunnel_encap(struct sk_buff *skb,
	517	struct ip_tunnel_encap *e,
	518	u8 protocol, struct flowi4 fl4)
	519	{
	520	const struct ip_tunnel_encap_ops *ops;
	521	int ret = -EINVAL;
	522
	523	if (e->type == TUNNEL_ENCAP_NONE)
	524	return 0;
	525
	526	if (e->type >= MAX_IPTUN_ENCAP_OPS)
	527	return -EINVAL;
	528
	529	rcu_read_lock();
	530	ops = rcu_dereference(iptun_encaps[e->type]);
	531	if (likely(ops && ops->build_header))
	532	ret = ops->build_header(skb, e, protocol, fl4);
	533	rcu_read_unlock();
	534
	535	return ret;
	536	}
	537
	538	/* Extract dsfield from inner protocol */
	539	static inline u8 ip_tunnel_get_dsfield(const struct iphdr *iph,
	540	const struct sk_buff *skb)
	541	{
	542	__be16 payload_protocol = skb_protocol(skb, true);
	543
	544	if (payload_protocol == htons(ETH_P_IP))
	545	return iph->tos;
	546	else if (payload_protocol == htons(ETH_P_IPV6))
	547	return ipv6_get_dsfield((const struct ipv6hdr *)iph);
	548	else
	549	return 0;
	550	}
	551
	552	static inline __be32 ip_tunnel_get_flowlabel(const struct iphdr *iph,
	553	const struct sk_buff *skb)
	554	{
	555	__be16 payload_protocol = skb_protocol(skb, true);
	556
	557	if (payload_protocol == htons(ETH_P_IPV6))
	558	return ip6_flowlabel((const struct ipv6hdr *)iph);
	559	else
	560	return 0;
	561	}
	562
	563	static inline u8 ip_tunnel_get_ttl(const struct iphdr *iph,
	564	const struct sk_buff *skb)
	565	{
	566	__be16 payload_protocol = skb_protocol(skb, true);
	567
	568	if (payload_protocol == htons(ETH_P_IP))
	569	return iph->ttl;
	570	else if (payload_protocol == htons(ETH_P_IPV6))
	571	return ((const struct ipv6hdr *)iph)->hop_limit;
	572	else
	573	return 0;
	574	}
	575
	576	/* Propagate ECN bits out */
	577	static inline u8 ip_tunnel_ecn_encap(u8 tos, const struct iphdr *iph,
	578	const struct sk_buff *skb)
	579	{
	580	u8 inner = ip_tunnel_get_dsfield(iph, skb);
	581
	582	return INET_ECN_encapsulate(tos, inner);
	583	}
	584
	585	int __iptunnel_pull_header(struct sk_buff *skb, int hdr_len,
	586	__be16 inner_proto, bool raw_proto, bool xnet);
	587
	588	static inline int iptunnel_pull_header(struct sk_buff *skb, int hdr_len,
	589	__be16 inner_proto, bool xnet)
	590	{
	591	return __iptunnel_pull_header(skb, hdr_len, inner_proto, false, xnet);
	592	}
	593
	594	void iptunnel_xmit(struct sock sk, struct rtable rt, struct sk_buff *skb,
	595	__be32 src, __be32 dst, u8 proto,
	596	u8 tos, u8 ttl, __be16 df, bool xnet);
	597	struct metadata_dst iptunnel_metadata_reply(struct metadata_dst md,
	598	gfp_t flags);
	599	int skb_tunnel_check_pmtu(struct sk_buff skb, struct dst_entry encap_dst,
	600	int headroom, bool reply);
	601
	602	int iptunnel_handle_offloads(struct sk_buff *skb, int gso_type_mask);
	603
	604	static inline int iptunnel_pull_offloads(struct sk_buff *skb)
	605	{
	606	if (skb_is_gso(skb)) {
	607	int err;
	608
	609	err = skb_unclone(skb, GFP_ATOMIC);
	610	if (unlikely(err))
	611	return err;
	612	skb_shinfo(skb)->gso_type &= ~(NETIF_F_GSO_ENCAP_ALL >>
	613	NETIF_F_GSO_SHIFT);
	614	}
	615
	616	skb->encapsulation = 0;
	617	return 0;
	618	}
	619
	620	static inline void iptunnel_xmit_stats(struct net_device *dev, int pkt_len)
	621	{
	622	if (pkt_len > 0) {
	623	struct pcpu_sw_netstats *tstats = get_cpu_ptr(dev->tstats);
	624
	625	u64_stats_update_begin(&tstats->syncp);
	626	u64_stats_add(&tstats->tx_bytes, pkt_len);
	627	u64_stats_inc(&tstats->tx_packets);
	628	u64_stats_update_end(&tstats->syncp);
	629	put_cpu_ptr(tstats);
	630	return;
	631	}
	632
	633	if (pkt_len < 0) {
	634	DEV_STATS_INC(dev, tx_errors);
	635	DEV_STATS_INC(dev, tx_aborted_errors);
	636	} else {
	637	DEV_STATS_INC(dev, tx_dropped);
	638	}
	639	}
	640
	641	static inline void ip_tunnel_info_opts_get(void *to,
	642	const struct ip_tunnel_info *info)
	643	{
	644	memcpy(to, info + 1, info->options_len);
	645	}
	646
	647	static inline void ip_tunnel_info_opts_set(struct ip_tunnel_info *info,
	648	const void *from, int len,
	649	const unsigned long *flags)
	650	{
	651	info->options_len = len;
	652	if (len > 0) {
	653	memcpy(ip_tunnel_info_opts(info), from, len);
	654	ip_tunnel_flags_or(info->key.tun_flags, info->key.tun_flags,
	655	flags);
	656	}
	657	}
	658
	659	static inline struct ip_tunnel_info lwt_tun_info(struct lwtunnel_state lwtstate)
	660	{
	661	return (struct ip_tunnel_info *)lwtstate->data;
	662	}
	663
	664	DECLARE_STATIC_KEY_FALSE(ip_tunnel_metadata_cnt);
	665
	666	/* Returns > 0 if metadata should be collected */
	667	static inline int ip_tunnel_collect_metadata(void)
	668	{
	669	return static_branch_unlikely(&ip_tunnel_metadata_cnt);
	670	}
	671
	672	void __init ip_tunnel_core_init(void);
	673
	674	void ip_tunnel_need_metadata(void);
	675	void ip_tunnel_unneed_metadata(void);
	676
	677	#else /* CONFIG_INET */
	678
	679	static inline struct ip_tunnel_info lwt_tun_info(struct lwtunnel_state lwtstate)
	680	{
	681	return NULL;
	682	}
	683
	684	static inline void ip_tunnel_need_metadata(void)
	685	{
	686	}
	687
	688	static inline void ip_tunnel_unneed_metadata(void)
	689	{
	690	}
	691
	692	static inline void ip_tunnel_info_opts_get(void *to,
	693	const struct ip_tunnel_info *info)
	694	{
	695	}
	696
	697	static inline void ip_tunnel_info_opts_set(struct ip_tunnel_info *info,
	698	const void *from, int len,
	699	const unsigned long *flags)
	700	{
	701	info->options_len = 0;
	702	}
	703
	704	#endif /* CONFIG_INET */
	705
	706	#endif /* __NET_IP_TUNNELS_H */