[mirror_ubuntu-zesty-kernel.git] / net / ipv4 / ip_tunnel_core.c

/*
 * Copyright (c) 2013 Nicira, Inc.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of version 2 of the GNU General Public
 * License as published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful, but
 * WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
 * General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
 * 02110-1301, USA
 */

#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt

#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/skbuff.h>
#include <linux/netdevice.h>
#include <linux/in.h>
#include <linux/if_arp.h>
#include <linux/init.h>
#include <linux/in6.h>
#include <linux/inetdevice.h>
#include <linux/netfilter_ipv4.h>
#include <linux/etherdevice.h>
#include <linux/if_ether.h>
#include <linux/if_vlan.h>
#include <linux/static_key.h>

#include <net/ip.h>
#include <net/icmp.h>
#include <net/protocol.h>
#include <net/ip_tunnels.h>
#include <net/ip6_tunnel.h>
#include <net/arp.h>
#include <net/checksum.h>
#include <net/dsfield.h>
#include <net/inet_ecn.h>
#include <net/xfrm.h>
#include <net/net_namespace.h>
#include <net/netns/generic.h>
#include <net/rtnetlink.h>
#include <net/dst_metadata.h>

const struct ip_tunnel_encap_ops __rcu *
		iptun_encaps[MAX_IPTUN_ENCAP_OPS] __read_mostly;
EXPORT_SYMBOL(iptun_encaps);

const struct ip6_tnl_encap_ops __rcu *
		ip6tun_encaps[MAX_IPTUN_ENCAP_OPS] __read_mostly;
EXPORT_SYMBOL(ip6tun_encaps);

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)
{
	int pkt_len = skb->len - skb_inner_network_offset(skb);
	struct net *net = dev_net(rt->dst.dev);
	struct net_device *dev = skb->dev;
	int skb_iif = skb->skb_iif;
	struct iphdr *iph;
	int err;

	skb_scrub_packet(skb, xnet);

	skb_clear_hash(skb);
	skb_dst_set(skb, &rt->dst);
	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));

	if (skb_iif && !(df & htons(IP_DF))) {
		/* Arrived from an ingress interface, got encapsulated, with
		 * fragmentation of encapulating frames allowed.
		 * If skb is gso, the resulting encapsulated network segments
		 * may exceed dst mtu.
		 * Allow IP Fragmentation of segments.
		 */
		IPCB(skb)->flags |= IPSKB_FRAG_SEGS;
	}

	/* Push down and install the IP header. */
	skb_push(skb, sizeof(struct iphdr));
	skb_reset_network_header(skb);

	iph = ip_hdr(skb);

	iph->version	=	4;
	iph->ihl	=	sizeof(struct iphdr) >> 2;
	iph->frag_off	=	df;
	iph->protocol	=	proto;
	iph->tos	=	tos;
	iph->daddr	=	dst;
	iph->saddr	=	src;
	iph->ttl	=	ttl;
	__ip_select_ident(net, iph, skb_shinfo(skb)->gso_segs ?: 1);

	err = ip_local_out(net, sk, skb);
	if (unlikely(net_xmit_eval(err)))
		pkt_len = 0;
	iptunnel_xmit_stats(dev, pkt_len);
}
EXPORT_SYMBOL_GPL(iptunnel_xmit);

int __iptunnel_pull_header(struct sk_buff *skb, int hdr_len,
			   __be16 inner_proto, bool raw_proto, bool xnet)
{
	if (unlikely(!pskb_may_pull(skb, hdr_len)))
		return -ENOMEM;

	skb_pull_rcsum(skb, hdr_len);

	if (!raw_proto && inner_proto == htons(ETH_P_TEB)) {
		struct ethhdr *eh;

		if (unlikely(!pskb_may_pull(skb, ETH_HLEN)))
			return -ENOMEM;

		eh = (struct ethhdr *)skb->data;
		if (likely(eth_proto_is_802_3(eh->h_proto)))
			skb->protocol = eh->h_proto;
		else
			skb->protocol = htons(ETH_P_802_2);

	} else {
		skb->protocol = inner_proto;
	}

	skb_clear_hash_if_not_l4(skb);
	skb->vlan_tci = 0;
	skb_set_queue_mapping(skb, 0);
	skb_scrub_packet(skb, xnet);

	return iptunnel_pull_offloads(skb);
}
EXPORT_SYMBOL_GPL(__iptunnel_pull_header);

struct metadata_dst *iptunnel_metadata_reply(struct metadata_dst *md,
					     gfp_t flags)
{
	struct metadata_dst *res;
	struct ip_tunnel_info *dst, *src;

	if (!md || md->u.tun_info.mode & IP_TUNNEL_INFO_TX)
		return NULL;

	res = metadata_dst_alloc(0, flags);
	if (!res)
		return NULL;

	dst = &res->u.tun_info;
	src = &md->u.tun_info;
	dst->key.tun_id = src->key.tun_id;
	if (src->mode & IP_TUNNEL_INFO_IPV6)
		memcpy(&dst->key.u.ipv6.dst, &src->key.u.ipv6.src,
		       sizeof(struct in6_addr));
	else
		dst->key.u.ipv4.dst = src->key.u.ipv4.src;
	dst->mode = src->mode | IP_TUNNEL_INFO_TX;

	return res;
}
EXPORT_SYMBOL_GPL(iptunnel_metadata_reply);

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

	if (likely(!skb->encapsulation)) {
		skb_reset_inner_headers(skb);
		skb->encapsulation = 1;
	}

	if (skb_is_gso(skb)) {
		err = skb_header_unclone(skb, GFP_ATOMIC);
		if (unlikely(err))
			return err;
		skb_shinfo(skb)->gso_type |= gso_type_mask;
		return 0;
	}

	if (skb->ip_summed != CHECKSUM_PARTIAL) {
		skb->ip_summed = CHECKSUM_NONE;
		/* We clear encapsulation here to prevent badly-written
		 * drivers potentially deciding to offload an inner checksum
		 * if we set CHECKSUM_PARTIAL on the outer header.
		 * This should go away when the drivers are all fixed.
		 */
		skb->encapsulation = 0;
	}

	return 0;
}
EXPORT_SYMBOL_GPL(iptunnel_handle_offloads);

/* Often modified stats are per cpu, other are shared (netdev->stats) */
struct rtnl_link_stats64 *ip_tunnel_get_stats64(struct net_device *dev,
						struct rtnl_link_stats64 *tot)
{
	int i;

	netdev_stats_to_stats64(tot, &dev->stats);

	for_each_possible_cpu(i) {
		const struct pcpu_sw_netstats *tstats =
						   per_cpu_ptr(dev->tstats, i);
		u64 rx_packets, rx_bytes, tx_packets, tx_bytes;
		unsigned int start;

		do {
			start = u64_stats_fetch_begin_irq(&tstats->syncp);
			rx_packets = tstats->rx_packets;
			tx_packets = tstats->tx_packets;
			rx_bytes = tstats->rx_bytes;
			tx_bytes = tstats->tx_bytes;
		} while (u64_stats_fetch_retry_irq(&tstats->syncp, start));

		tot->rx_packets += rx_packets;
		tot->tx_packets += tx_packets;
		tot->rx_bytes   += rx_bytes;
		tot->tx_bytes   += tx_bytes;
	}

	return tot;
}
EXPORT_SYMBOL_GPL(ip_tunnel_get_stats64);

static const struct nla_policy ip_tun_policy[LWTUNNEL_IP_MAX + 1] = {
	[LWTUNNEL_IP_ID]	= { .type = NLA_U64 },
	[LWTUNNEL_IP_DST]	= { .type = NLA_U32 },
	[LWTUNNEL_IP_SRC]	= { .type = NLA_U32 },
	[LWTUNNEL_IP_TTL]	= { .type = NLA_U8 },
	[LWTUNNEL_IP_TOS]	= { .type = NLA_U8 },
	[LWTUNNEL_IP_FLAGS]	= { .type = NLA_U16 },
};

static int ip_tun_build_state(struct net_device *dev, struct nlattr *attr,
			      unsigned int family, const void *cfg,
			      struct lwtunnel_state **ts)
{
	struct ip_tunnel_info *tun_info;
	struct lwtunnel_state *new_state;
	struct nlattr *tb[LWTUNNEL_IP_MAX + 1];
	int err;

	err = nla_parse_nested(tb, LWTUNNEL_IP_MAX, attr, ip_tun_policy);
	if (err < 0)
		return err;

	new_state = lwtunnel_state_alloc(sizeof(*tun_info));
	if (!new_state)
		return -ENOMEM;

	new_state->type = LWTUNNEL_ENCAP_IP;

	tun_info = lwt_tun_info(new_state);

	if (tb[LWTUNNEL_IP_ID])
		tun_info->key.tun_id = nla_get_be64(tb[LWTUNNEL_IP_ID]);

	if (tb[LWTUNNEL_IP_DST])
		tun_info->key.u.ipv4.dst = nla_get_in_addr(tb[LWTUNNEL_IP_DST]);

	if (tb[LWTUNNEL_IP_SRC])
		tun_info->key.u.ipv4.src = nla_get_in_addr(tb[LWTUNNEL_IP_SRC]);

	if (tb[LWTUNNEL_IP_TTL])
		tun_info->key.ttl = nla_get_u8(tb[LWTUNNEL_IP_TTL]);

	if (tb[LWTUNNEL_IP_TOS])
		tun_info->key.tos = nla_get_u8(tb[LWTUNNEL_IP_TOS]);

	if (tb[LWTUNNEL_IP_FLAGS])
		tun_info->key.tun_flags = nla_get_be16(tb[LWTUNNEL_IP_FLAGS]);

	tun_info->mode = IP_TUNNEL_INFO_TX;
	tun_info->options_len = 0;

	*ts = new_state;

	return 0;
}

static int ip_tun_fill_encap_info(struct sk_buff *skb,
				  struct lwtunnel_state *lwtstate)
{
	struct ip_tunnel_info *tun_info = lwt_tun_info(lwtstate);

	if (nla_put_be64(skb, LWTUNNEL_IP_ID, tun_info->key.tun_id,
			 LWTUNNEL_IP_PAD) ||
	    nla_put_in_addr(skb, LWTUNNEL_IP_DST, tun_info->key.u.ipv4.dst) ||
	    nla_put_in_addr(skb, LWTUNNEL_IP_SRC, tun_info->key.u.ipv4.src) ||
	    nla_put_u8(skb, LWTUNNEL_IP_TOS, tun_info->key.tos) ||
	    nla_put_u8(skb, LWTUNNEL_IP_TTL, tun_info->key.ttl) ||
	    nla_put_be16(skb, LWTUNNEL_IP_FLAGS, tun_info->key.tun_flags))
		return -ENOMEM;

	return 0;
}

static int ip_tun_encap_nlsize(struct lwtunnel_state *lwtstate)
{
	return nla_total_size_64bit(8)	/* LWTUNNEL_IP_ID */
		+ nla_total_size(4)	/* LWTUNNEL_IP_DST */
		+ nla_total_size(4)	/* LWTUNNEL_IP_SRC */
		+ nla_total_size(1)	/* LWTUNNEL_IP_TOS */
		+ nla_total_size(1)	/* LWTUNNEL_IP_TTL */
		+ nla_total_size(2);	/* LWTUNNEL_IP_FLAGS */
}

static int ip_tun_cmp_encap(struct lwtunnel_state *a, struct lwtunnel_state *b)
{
	return memcmp(lwt_tun_info(a), lwt_tun_info(b),
		      sizeof(struct ip_tunnel_info));
}

static const struct lwtunnel_encap_ops ip_tun_lwt_ops = {
	.build_state = ip_tun_build_state,
	.fill_encap = ip_tun_fill_encap_info,
	.get_encap_size = ip_tun_encap_nlsize,
	.cmp_encap = ip_tun_cmp_encap,
};

static const struct nla_policy ip6_tun_policy[LWTUNNEL_IP6_MAX + 1] = {
	[LWTUNNEL_IP6_ID]		= { .type = NLA_U64 },
	[LWTUNNEL_IP6_DST]		= { .len = sizeof(struct in6_addr) },
	[LWTUNNEL_IP6_SRC]		= { .len = sizeof(struct in6_addr) },
	[LWTUNNEL_IP6_HOPLIMIT]		= { .type = NLA_U8 },
	[LWTUNNEL_IP6_TC]		= { .type = NLA_U8 },
	[LWTUNNEL_IP6_FLAGS]		= { .type = NLA_U16 },
};

static int ip6_tun_build_state(struct net_device *dev, struct nlattr *attr,
			       unsigned int family, const void *cfg,
			       struct lwtunnel_state **ts)
{
	struct ip_tunnel_info *tun_info;
	struct lwtunnel_state *new_state;
	struct nlattr *tb[LWTUNNEL_IP6_MAX + 1];
	int err;

	err = nla_parse_nested(tb, LWTUNNEL_IP6_MAX, attr, ip6_tun_policy);
	if (err < 0)
		return err;

	new_state = lwtunnel_state_alloc(sizeof(*tun_info));
	if (!new_state)
		return -ENOMEM;

	new_state->type = LWTUNNEL_ENCAP_IP6;

	tun_info = lwt_tun_info(new_state);

	if (tb[LWTUNNEL_IP6_ID])
		tun_info->key.tun_id = nla_get_be64(tb[LWTUNNEL_IP6_ID]);

	if (tb[LWTUNNEL_IP6_DST])
		tun_info->key.u.ipv6.dst = nla_get_in6_addr(tb[LWTUNNEL_IP6_DST]);

	if (tb[LWTUNNEL_IP6_SRC])
		tun_info->key.u.ipv6.src = nla_get_in6_addr(tb[LWTUNNEL_IP6_SRC]);

	if (tb[LWTUNNEL_IP6_HOPLIMIT])
		tun_info->key.ttl = nla_get_u8(tb[LWTUNNEL_IP6_HOPLIMIT]);

	if (tb[LWTUNNEL_IP6_TC])
		tun_info->key.tos = nla_get_u8(tb[LWTUNNEL_IP6_TC]);

	if (tb[LWTUNNEL_IP6_FLAGS])
		tun_info->key.tun_flags = nla_get_be16(tb[LWTUNNEL_IP6_FLAGS]);

	tun_info->mode = IP_TUNNEL_INFO_TX | IP_TUNNEL_INFO_IPV6;
	tun_info->options_len = 0;

	*ts = new_state;

	return 0;
}

static int ip6_tun_fill_encap_info(struct sk_buff *skb,
				   struct lwtunnel_state *lwtstate)
{
	struct ip_tunnel_info *tun_info = lwt_tun_info(lwtstate);

	if (nla_put_be64(skb, LWTUNNEL_IP6_ID, tun_info->key.tun_id,
			 LWTUNNEL_IP6_PAD) ||
	    nla_put_in6_addr(skb, LWTUNNEL_IP6_DST, &tun_info->key.u.ipv6.dst) ||
	    nla_put_in6_addr(skb, LWTUNNEL_IP6_SRC, &tun_info->key.u.ipv6.src) ||
	    nla_put_u8(skb, LWTUNNEL_IP6_TC, tun_info->key.tos) ||
	    nla_put_u8(skb, LWTUNNEL_IP6_HOPLIMIT, tun_info->key.ttl) ||
	    nla_put_be16(skb, LWTUNNEL_IP6_FLAGS, tun_info->key.tun_flags))
		return -ENOMEM;

	return 0;
}

static int ip6_tun_encap_nlsize(struct lwtunnel_state *lwtstate)
{
	return nla_total_size_64bit(8)	/* LWTUNNEL_IP6_ID */
		+ nla_total_size(16)	/* LWTUNNEL_IP6_DST */
		+ nla_total_size(16)	/* LWTUNNEL_IP6_SRC */
		+ nla_total_size(1)	/* LWTUNNEL_IP6_HOPLIMIT */
		+ nla_total_size(1)	/* LWTUNNEL_IP6_TC */
		+ nla_total_size(2);	/* LWTUNNEL_IP6_FLAGS */
}

static const struct lwtunnel_encap_ops ip6_tun_lwt_ops = {
	.build_state = ip6_tun_build_state,
	.fill_encap = ip6_tun_fill_encap_info,
	.get_encap_size = ip6_tun_encap_nlsize,
	.cmp_encap = ip_tun_cmp_encap,
};

void __init ip_tunnel_core_init(void)
{
	/* If you land here, make sure whether increasing ip_tunnel_info's
	 * options_len is a reasonable choice with its usage in front ends
	 * (f.e., it's part of flow keys, etc).
	 */
	BUILD_BUG_ON(IP_TUNNEL_OPTS_MAX != 255);

	lwtunnel_encap_add_ops(&ip_tun_lwt_ops, LWTUNNEL_ENCAP_IP);
	lwtunnel_encap_add_ops(&ip6_tun_lwt_ops, LWTUNNEL_ENCAP_IP6);
}

struct static_key ip_tunnel_metadata_cnt = STATIC_KEY_INIT_FALSE;
EXPORT_SYMBOL(ip_tunnel_metadata_cnt);

void ip_tunnel_need_metadata(void)
{
	static_key_slow_inc(&ip_tunnel_metadata_cnt);
}
EXPORT_SYMBOL_GPL(ip_tunnel_need_metadata);

void ip_tunnel_unneed_metadata(void)
{
	static_key_slow_dec(&ip_tunnel_metadata_cnt);
}
EXPORT_SYMBOL_GPL(ip_tunnel_unneed_metadata);
Commit	Line	Data
0e6fbc5b PS	1	/*
	2	* Copyright (c) 2013 Nicira, Inc.
	3	*
	4	* This program is free software; you can redistribute it and/or
	5	* modify it under the terms of version 2 of the GNU General Public
	6	* License as published by the Free Software Foundation.
	7	*
	8	* This program is distributed in the hope that it will be useful, but
	9	* WITHOUT ANY WARRANTY; without even the implied warranty of
	10	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
	11	* General Public License for more details.
	12	*
	13	* You should have received a copy of the GNU General Public License
	14	* along with this program; if not, write to the Free Software
	15	* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
	16	* 02110-1301, USA
	17	*/
	18
	19	#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
	20
	21	#include <linux/types.h>
	22	#include <linux/kernel.h>
	23	#include <linux/skbuff.h>
	24	#include <linux/netdevice.h>
	25	#include <linux/in.h>
	26	#include <linux/if_arp.h>
0e6fbc5b PS	27	#include <linux/init.h>
	28	#include <linux/in6.h>
	29	#include <linux/inetdevice.h>
	30	#include <linux/netfilter_ipv4.h>
	31	#include <linux/etherdevice.h>
	32	#include <linux/if_ether.h>
	33	#include <linux/if_vlan.h>
e7030878	34	#include <linux/static_key.h>
0e6fbc5b PS	35
	36	#include <net/ip.h>
	37	#include <net/icmp.h>
	38	#include <net/protocol.h>
	39	#include <net/ip_tunnels.h>
058214a4	40	#include <net/ip6_tunnel.h>
0e6fbc5b PS	41	#include <net/arp.h>
	42	#include <net/checksum.h>
	43	#include <net/dsfield.h>
	44	#include <net/inet_ecn.h>
	45	#include <net/xfrm.h>
	46	#include <net/net_namespace.h>
	47	#include <net/netns/generic.h>
	48	#include <net/rtnetlink.h>
63d008a4	49	#include <net/dst_metadata.h>
0e6fbc5b	50
55c2bc14 TH	51	const struct ip_tunnel_encap_ops __rcu *
	52	iptun_encaps[MAX_IPTUN_ENCAP_OPS] __read_mostly;
	53	EXPORT_SYMBOL(iptun_encaps);
	54
058214a4 TH	55	const struct ip6_tnl_encap_ops __rcu *
	56	ip6tun_encaps[MAX_IPTUN_ENCAP_OPS] __read_mostly;
	57	EXPORT_SYMBOL(ip6tun_encaps);
	58
039f5062 PS	59	void iptunnel_xmit(struct sock sk, struct rtable rt, struct sk_buff *skb,
	60	__be32 src, __be32 dst, __u8 proto,
	61	__u8 tos, __u8 ttl, __be16 df, bool xnet)
0e6fbc5b	62	{
bc22a0e2	63	int pkt_len = skb->len - skb_inner_network_offset(skb);
f859b0f6	64	struct net *net = dev_net(rt->dst.dev);
039f5062	65	struct net_device *dev = skb->dev;
b8247f09	66	int skb_iif = skb->skb_iif;
0e6fbc5b PS	67	struct iphdr *iph;
	68	int err;
	69
963a88b3 ND	70	skb_scrub_packet(skb, xnet);
963a88b3 ND	71
7539fadc	72	skb_clear_hash(skb);
0e6fbc5b PS	73	skb_dst_set(skb, &rt->dst);
	74	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
	75
c0451fe1 SL	76	if (skb_iif && !(df & htons(IP_DF))) {
	77	/* Arrived from an ingress interface, got encapsulated, with
	78	* fragmentation of encapulating frames allowed.
	79	* If skb is gso, the resulting encapsulated network segments
	80	* may exceed dst mtu.
b8247f09 SL	81	* Allow IP Fragmentation of segments.
	82	*/
	83	IPCB(skb)->flags \|= IPSKB_FRAG_SEGS;
	84	}
	85
0e6fbc5b	86	/* Push down and install the IP header. */
78a3694d	87	skb_push(skb, sizeof(struct iphdr));
0e6fbc5b PS	88	skb_reset_network_header(skb);
	89
	90	iph = ip_hdr(skb);
	91
	92	iph->version = 4;
	93	iph->ihl = sizeof(struct iphdr) >> 2;
	94	iph->frag_off = df;
	95	iph->protocol = proto;
	96	iph->tos = tos;
	97	iph->daddr = dst;
	98	iph->saddr = src;
	99	iph->ttl = ttl;
f859b0f6	100	__ip_select_ident(net, iph, skb_shinfo(skb)->gso_segs ?: 1);
0e6fbc5b	101
33224b16	102	err = ip_local_out(net, sk, skb);
0e6fbc5b PS	103	if (unlikely(net_xmit_eval(err)))
0e6fbc5b PS	104	pkt_len = 0;
039f5062	105	iptunnel_xmit_stats(dev, pkt_len);
0e6fbc5b PS	106	}
0e6fbc5b PS	107	EXPORT_SYMBOL_GPL(iptunnel_xmit);
3d7b46cd	108
a6d5bbf3 JB	109	int __iptunnel_pull_header(struct sk_buff *skb, int hdr_len,
a6d5bbf3 JB	110	__be16 inner_proto, bool raw_proto, bool xnet)
3d7b46cd PS	111	{
	112	if (unlikely(!pskb_may_pull(skb, hdr_len)))
	113	return -ENOMEM;
	114
	115	skb_pull_rcsum(skb, hdr_len);
	116
a6d5bbf3	117	if (!raw_proto && inner_proto == htons(ETH_P_TEB)) {
1245dfc8	118	struct ethhdr *eh;
3d7b46cd PS	119
	120	if (unlikely(!pskb_may_pull(skb, ETH_HLEN)))
	121	return -ENOMEM;
	122
1245dfc8	123	eh = (struct ethhdr *)skb->data;
d181ddca	124	if (likely(eth_proto_is_802_3(eh->h_proto)))
3d7b46cd PS	125	skb->protocol = eh->h_proto;
	126	else
	127	skb->protocol = htons(ETH_P_802_2);
	128
	129	} else {
	130	skb->protocol = inner_proto;
	131	}
	132
7539fadc	133	skb_clear_hash_if_not_l4(skb);
3d7b46cd PS	134	skb->vlan_tci = 0;
3d7b46cd PS	135	skb_set_queue_mapping(skb, 0);
7f290c94	136	skb_scrub_packet(skb, xnet);
a09a4c8d JG	137
a09a4c8d JG	138	return iptunnel_pull_offloads(skb);
3d7b46cd	139	}
a6d5bbf3	140	EXPORT_SYMBOL_GPL(__iptunnel_pull_header);
2d26f0a3	141
63d008a4 JB	142	struct metadata_dst iptunnel_metadata_reply(struct metadata_dst md,
	143	gfp_t flags)
	144	{
	145	struct metadata_dst *res;
	146	struct ip_tunnel_info dst, src;
	147
	148	if (!md \|\| md->u.tun_info.mode & IP_TUNNEL_INFO_TX)
	149	return NULL;
	150
	151	res = metadata_dst_alloc(0, flags);
	152	if (!res)
	153	return NULL;
	154
	155	dst = &res->u.tun_info;
	156	src = &md->u.tun_info;
	157	dst->key.tun_id = src->key.tun_id;
	158	if (src->mode & IP_TUNNEL_INFO_IPV6)
	159	memcpy(&dst->key.u.ipv6.dst, &src->key.u.ipv6.src,
	160	sizeof(struct in6_addr));
	161	else
	162	dst->key.u.ipv4.dst = src->key.u.ipv4.src;
	163	dst->mode = src->mode \| IP_TUNNEL_INFO_TX;
	164
	165	return res;
	166	}
	167	EXPORT_SYMBOL_GPL(iptunnel_metadata_reply);
	168
aed069df AD	169	int iptunnel_handle_offloads(struct sk_buff *skb,
aed069df AD	170	int gso_type_mask)
2d26f0a3 ED	171	{
	172	int err;
	173
	174	if (likely(!skb->encapsulation)) {
	175	skb_reset_inner_headers(skb);
	176	skb->encapsulation = 1;
	177	}
	178
	179	if (skb_is_gso(skb)) {
9580bf2e	180	err = skb_header_unclone(skb, GFP_ATOMIC);
2d26f0a3	181	if (unlikely(err))
aed069df	182	return err;
2d26f0a3	183	skb_shinfo(skb)->gso_type \|= gso_type_mask;
aed069df	184	return 0;
2d26f0a3 ED	185	}
2d26f0a3 ED	186
6fa79666	187	if (skb->ip_summed != CHECKSUM_PARTIAL) {
2d26f0a3	188	skb->ip_summed = CHECKSUM_NONE;
6fa79666 EC	189	/* We clear encapsulation here to prevent badly-written
	190	* drivers potentially deciding to offload an inner checksum
	191	* if we set CHECKSUM_PARTIAL on the outer header.
	192	* This should go away when the drivers are all fixed.
	193	*/
179bc67f EC	194	skb->encapsulation = 0;
179bc67f EC	195	}
2d26f0a3	196
aed069df	197	return 0;
2d26f0a3 ED	198	}
2d26f0a3 ED	199	EXPORT_SYMBOL_GPL(iptunnel_handle_offloads);
ebe44f35 DM	200
	201	/* Often modified stats are per cpu, other are shared (netdev->stats) */
	202	struct rtnl_link_stats64 ip_tunnel_get_stats64(struct net_device dev,
	203	struct rtnl_link_stats64 *tot)
	204	{
	205	int i;
	206
c24a5964 AD	207	netdev_stats_to_stats64(tot, &dev->stats);
c24a5964 AD	208
ebe44f35 DM	209	for_each_possible_cpu(i) {
	210	const struct pcpu_sw_netstats *tstats =
	211	per_cpu_ptr(dev->tstats, i);
	212	u64 rx_packets, rx_bytes, tx_packets, tx_bytes;
	213	unsigned int start;
	214
	215	do {
57a7744e	216	start = u64_stats_fetch_begin_irq(&tstats->syncp);
ebe44f35 DM	217	rx_packets = tstats->rx_packets;
	218	tx_packets = tstats->tx_packets;
	219	rx_bytes = tstats->rx_bytes;
	220	tx_bytes = tstats->tx_bytes;
57a7744e	221	} while (u64_stats_fetch_retry_irq(&tstats->syncp, start));
ebe44f35 DM	222
	223	tot->rx_packets += rx_packets;
	224	tot->tx_packets += tx_packets;
	225	tot->rx_bytes += rx_bytes;
	226	tot->tx_bytes += tx_bytes;
	227	}
	228
ebe44f35 DM	229	return tot;
	230	}
	231	EXPORT_SYMBOL_GPL(ip_tunnel_get_stats64);
3093fbe7	232
a1c234f9 JB	233	static const struct nla_policy ip_tun_policy[LWTUNNEL_IP_MAX + 1] = {
	234	[LWTUNNEL_IP_ID] = { .type = NLA_U64 },
	235	[LWTUNNEL_IP_DST] = { .type = NLA_U32 },
	236	[LWTUNNEL_IP_SRC] = { .type = NLA_U32 },
	237	[LWTUNNEL_IP_TTL] = { .type = NLA_U8 },
	238	[LWTUNNEL_IP_TOS] = { .type = NLA_U8 },
a1c234f9	239	[LWTUNNEL_IP_FLAGS] = { .type = NLA_U16 },
3093fbe7 TG	240	};
	241
	242	static int ip_tun_build_state(struct net_device dev, struct nlattr attr,
127eb7cd	243	unsigned int family, const void *cfg,
3093fbe7 TG	244	struct lwtunnel_state **ts)
	245	{
	246	struct ip_tunnel_info *tun_info;
	247	struct lwtunnel_state *new_state;
a1c234f9	248	struct nlattr *tb[LWTUNNEL_IP_MAX + 1];
3093fbe7 TG	249	int err;
3093fbe7 TG	250
a1c234f9	251	err = nla_parse_nested(tb, LWTUNNEL_IP_MAX, attr, ip_tun_policy);
3093fbe7 TG	252	if (err < 0)
	253	return err;
	254
	255	new_state = lwtunnel_state_alloc(sizeof(*tun_info));
	256	if (!new_state)
	257	return -ENOMEM;
	258
	259	new_state->type = LWTUNNEL_ENCAP_IP;
	260
	261	tun_info = lwt_tun_info(new_state);
	262
a1c234f9	263	if (tb[LWTUNNEL_IP_ID])
30d3d83a	264	tun_info->key.tun_id = nla_get_be64(tb[LWTUNNEL_IP_ID]);
3093fbe7	265
a1c234f9	266	if (tb[LWTUNNEL_IP_DST])
7822ce73	267	tun_info->key.u.ipv4.dst = nla_get_in_addr(tb[LWTUNNEL_IP_DST]);
3093fbe7	268
a1c234f9	269	if (tb[LWTUNNEL_IP_SRC])
7822ce73	270	tun_info->key.u.ipv4.src = nla_get_in_addr(tb[LWTUNNEL_IP_SRC]);
3093fbe7	271
a1c234f9	272	if (tb[LWTUNNEL_IP_TTL])
7c383fb2	273	tun_info->key.ttl = nla_get_u8(tb[LWTUNNEL_IP_TTL]);
3093fbe7	274
a1c234f9	275	if (tb[LWTUNNEL_IP_TOS])
7c383fb2	276	tun_info->key.tos = nla_get_u8(tb[LWTUNNEL_IP_TOS]);
3093fbe7	277
a1c234f9	278	if (tb[LWTUNNEL_IP_FLAGS])
30d3d83a	279	tun_info->key.tun_flags = nla_get_be16(tb[LWTUNNEL_IP_FLAGS]);
3093fbe7 TG	280
3093fbe7 TG	281	tun_info->mode = IP_TUNNEL_INFO_TX;
3093fbe7 TG	282	tun_info->options_len = 0;
	283
	284	*ts = new_state;
	285
	286	return 0;
	287	}
	288
	289	static int ip_tun_fill_encap_info(struct sk_buff *skb,
	290	struct lwtunnel_state *lwtstate)
	291	{
	292	struct ip_tunnel_info *tun_info = lwt_tun_info(lwtstate);
	293
b46f6ded ND	294	if (nla_put_be64(skb, LWTUNNEL_IP_ID, tun_info->key.tun_id,
b46f6ded ND	295	LWTUNNEL_IP_PAD) \|\|
7822ce73 HY	296	nla_put_in_addr(skb, LWTUNNEL_IP_DST, tun_info->key.u.ipv4.dst) \|\|
7822ce73 HY	297	nla_put_in_addr(skb, LWTUNNEL_IP_SRC, tun_info->key.u.ipv4.src) \|\|
7c383fb2 JB	298	nla_put_u8(skb, LWTUNNEL_IP_TOS, tun_info->key.tos) \|\|
7c383fb2 JB	299	nla_put_u8(skb, LWTUNNEL_IP_TTL, tun_info->key.ttl) \|\|
30d3d83a	300	nla_put_be16(skb, LWTUNNEL_IP_FLAGS, tun_info->key.tun_flags))
3093fbe7 TG	301	return -ENOMEM;
	302
	303	return 0;
	304	}
	305
	306	static int ip_tun_encap_nlsize(struct lwtunnel_state *lwtstate)
	307	{
b46f6ded	308	return nla_total_size_64bit(8) /* LWTUNNEL_IP_ID */
a1c234f9 JB	309	+ nla_total_size(4) /* LWTUNNEL_IP_DST */
	310	+ nla_total_size(4) /* LWTUNNEL_IP_SRC */
	311	+ nla_total_size(1) /* LWTUNNEL_IP_TOS */
	312	+ nla_total_size(1) /* LWTUNNEL_IP_TTL */
a1c234f9	313	+ nla_total_size(2); /* LWTUNNEL_IP_FLAGS */
3093fbe7 TG	314	}
3093fbe7 TG	315
2d798499 JB	316	static int ip_tun_cmp_encap(struct lwtunnel_state a, struct lwtunnel_state b)
	317	{
	318	return memcmp(lwt_tun_info(a), lwt_tun_info(b),
	319	sizeof(struct ip_tunnel_info));
	320	}
	321
3093fbe7 TG	322	static const struct lwtunnel_encap_ops ip_tun_lwt_ops = {
	323	.build_state = ip_tun_build_state,
	324	.fill_encap = ip_tun_fill_encap_info,
	325	.get_encap_size = ip_tun_encap_nlsize,
2d798499	326	.cmp_encap = ip_tun_cmp_encap,
3093fbe7 TG	327	};
3093fbe7 TG	328
32a2b002 JB	329	static const struct nla_policy ip6_tun_policy[LWTUNNEL_IP6_MAX + 1] = {
	330	[LWTUNNEL_IP6_ID] = { .type = NLA_U64 },
	331	[LWTUNNEL_IP6_DST] = { .len = sizeof(struct in6_addr) },
	332	[LWTUNNEL_IP6_SRC] = { .len = sizeof(struct in6_addr) },
	333	[LWTUNNEL_IP6_HOPLIMIT] = { .type = NLA_U8 },
	334	[LWTUNNEL_IP6_TC] = { .type = NLA_U8 },
32a2b002 JB	335	[LWTUNNEL_IP6_FLAGS] = { .type = NLA_U16 },
	336	};
	337
	338	static int ip6_tun_build_state(struct net_device dev, struct nlattr attr,
127eb7cd	339	unsigned int family, const void *cfg,
32a2b002 JB	340	struct lwtunnel_state **ts)
	341	{
	342	struct ip_tunnel_info *tun_info;
	343	struct lwtunnel_state *new_state;
	344	struct nlattr *tb[LWTUNNEL_IP6_MAX + 1];
	345	int err;
	346
	347	err = nla_parse_nested(tb, LWTUNNEL_IP6_MAX, attr, ip6_tun_policy);
	348	if (err < 0)
	349	return err;
	350
	351	new_state = lwtunnel_state_alloc(sizeof(*tun_info));
	352	if (!new_state)
	353	return -ENOMEM;
	354
	355	new_state->type = LWTUNNEL_ENCAP_IP6;
	356
	357	tun_info = lwt_tun_info(new_state);
	358
	359	if (tb[LWTUNNEL_IP6_ID])
30d3d83a	360	tun_info->key.tun_id = nla_get_be64(tb[LWTUNNEL_IP6_ID]);
32a2b002 JB	361
	362	if (tb[LWTUNNEL_IP6_DST])
	363	tun_info->key.u.ipv6.dst = nla_get_in6_addr(tb[LWTUNNEL_IP6_DST]);
	364
	365	if (tb[LWTUNNEL_IP6_SRC])
	366	tun_info->key.u.ipv6.src = nla_get_in6_addr(tb[LWTUNNEL_IP6_SRC]);
	367
	368	if (tb[LWTUNNEL_IP6_HOPLIMIT])
	369	tun_info->key.ttl = nla_get_u8(tb[LWTUNNEL_IP6_HOPLIMIT]);
	370
	371	if (tb[LWTUNNEL_IP6_TC])
	372	tun_info->key.tos = nla_get_u8(tb[LWTUNNEL_IP6_TC]);
	373
32a2b002	374	if (tb[LWTUNNEL_IP6_FLAGS])
30d3d83a	375	tun_info->key.tun_flags = nla_get_be16(tb[LWTUNNEL_IP6_FLAGS]);
32a2b002	376
7f9562a1	377	tun_info->mode = IP_TUNNEL_INFO_TX \| IP_TUNNEL_INFO_IPV6;
32a2b002 JB	378	tun_info->options_len = 0;
	379
	380	*ts = new_state;
	381
	382	return 0;
	383	}
	384
	385	static int ip6_tun_fill_encap_info(struct sk_buff *skb,
	386	struct lwtunnel_state *lwtstate)
	387	{
	388	struct ip_tunnel_info *tun_info = lwt_tun_info(lwtstate);
	389
b46f6ded ND	390	if (nla_put_be64(skb, LWTUNNEL_IP6_ID, tun_info->key.tun_id,
b46f6ded ND	391	LWTUNNEL_IP6_PAD) \|\|
32a2b002 JB	392	nla_put_in6_addr(skb, LWTUNNEL_IP6_DST, &tun_info->key.u.ipv6.dst) \|\|
32a2b002 JB	393	nla_put_in6_addr(skb, LWTUNNEL_IP6_SRC, &tun_info->key.u.ipv6.src) \|\|
995096a0 QA	394	nla_put_u8(skb, LWTUNNEL_IP6_TC, tun_info->key.tos) \|\|
995096a0 QA	395	nla_put_u8(skb, LWTUNNEL_IP6_HOPLIMIT, tun_info->key.ttl) \|\|
30d3d83a	396	nla_put_be16(skb, LWTUNNEL_IP6_FLAGS, tun_info->key.tun_flags))
32a2b002 JB	397	return -ENOMEM;
	398
	399	return 0;
	400	}
	401
	402	static int ip6_tun_encap_nlsize(struct lwtunnel_state *lwtstate)
	403	{
b46f6ded	404	return nla_total_size_64bit(8) /* LWTUNNEL_IP6_ID */
32a2b002 JB	405	+ nla_total_size(16) /* LWTUNNEL_IP6_DST */
	406	+ nla_total_size(16) /* LWTUNNEL_IP6_SRC */
	407	+ nla_total_size(1) /* LWTUNNEL_IP6_HOPLIMIT */
	408	+ nla_total_size(1) /* LWTUNNEL_IP6_TC */
32a2b002 JB	409	+ nla_total_size(2); /* LWTUNNEL_IP6_FLAGS */
	410	}
	411
	412	static const struct lwtunnel_encap_ops ip6_tun_lwt_ops = {
	413	.build_state = ip6_tun_build_state,
	414	.fill_encap = ip6_tun_fill_encap_info,
	415	.get_encap_size = ip6_tun_encap_nlsize,
	416	.cmp_encap = ip_tun_cmp_encap,
	417	};
	418
045a0fa0	419	void __init ip_tunnel_core_init(void)
3093fbe7	420	{
fca5fdf6 DB	421	/* If you land here, make sure whether increasing ip_tunnel_info's
	422	* options_len is a reasonable choice with its usage in front ends
	423	* (f.e., it's part of flow keys, etc).
	424	*/
	425	BUILD_BUG_ON(IP_TUNNEL_OPTS_MAX != 255);
	426
3093fbe7	427	lwtunnel_encap_add_ops(&ip_tun_lwt_ops, LWTUNNEL_ENCAP_IP);
32a2b002	428	lwtunnel_encap_add_ops(&ip6_tun_lwt_ops, LWTUNNEL_ENCAP_IP6);
3093fbe7	429	}
e7030878 TG	430
	431	struct static_key ip_tunnel_metadata_cnt = STATIC_KEY_INIT_FALSE;
	432	EXPORT_SYMBOL(ip_tunnel_metadata_cnt);
	433
	434	void ip_tunnel_need_metadata(void)
	435	{
	436	static_key_slow_inc(&ip_tunnel_metadata_cnt);
	437	}
	438	EXPORT_SYMBOL_GPL(ip_tunnel_need_metadata);
	439
	440	void ip_tunnel_unneed_metadata(void)
	441	{
	442	static_key_slow_dec(&ip_tunnel_metadata_cnt);
	443	}
	444	EXPORT_SYMBOL_GPL(ip_tunnel_unneed_metadata);