ip_tunnel_core.c 11.8 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
/*
 * 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/mroute.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>
35
#include <linux/static_key.h>
36 37 38 39 40 41 42 43 44 45 46 47 48 49

#include <net/ip.h>
#include <net/icmp.h>
#include <net/protocol.h>
#include <net/ip_tunnels.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>

50
int iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb,
51
		  __be32 src, __be32 dst, __u8 proto,
52
		  __u8 tos, __u8 ttl, __be16 df, bool xnet)
53
{
54
	int pkt_len = skb->len - skb_inner_network_offset(skb);
55 56 57
	struct iphdr *iph;
	int err;

58 59
	skb_scrub_packet(skb, xnet);

60
	skb_clear_hash(skb);
61 62 63 64
	skb_dst_set(skb, &rt->dst);
	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));

	/* Push down and install the IP header. */
65
	skb_push(skb, sizeof(struct iphdr));
66 67 68 69 70 71 72 73 74 75 76 77
	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;
78 79
	__ip_select_ident(dev_net(rt->dst.dev), iph,
			  skb_shinfo(skb)->gso_segs ?: 1);
80

81
	err = ip_local_out_sk(sk, skb);
82 83 84 85 86
	if (unlikely(net_xmit_eval(err)))
		pkt_len = 0;
	return pkt_len;
}
EXPORT_SYMBOL_GPL(iptunnel_xmit);
87 88 89 90 91 92 93 94 95

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

	skb_pull_rcsum(skb, hdr_len);

	if (inner_proto == htons(ETH_P_TEB)) {
96
		struct ethhdr *eh;
97 98 99 100

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

101
		eh = (struct ethhdr *)skb->data;
102
		if (likely(eth_proto_is_802_3(eh->h_proto)))
103 104 105 106 107 108 109 110 111 112
			skb->protocol = eh->h_proto;
		else
			skb->protocol = htons(ETH_P_802_2);

	} else {
		skb->protocol = inner_proto;
	}

	nf_reset(skb);
	secpath_reset(skb);
113
	skb_clear_hash_if_not_l4(skb);
P
Pravin B Shelar 已提交
114
	skb_dst_drop(skb);
115 116 117 118 119 120
	skb->vlan_tci = 0;
	skb_set_queue_mapping(skb, 0);
	skb->pkt_type = PACKET_HOST;
	return 0;
}
EXPORT_SYMBOL_GPL(iptunnel_pull_header);
121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140

struct sk_buff *iptunnel_handle_offloads(struct sk_buff *skb,
					 bool csum_help,
					 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_unclone(skb, GFP_ATOMIC);
		if (unlikely(err))
			goto error;
		skb_shinfo(skb)->gso_type |= gso_type_mask;
		return skb;
	}

141 142 143 144 145 146 147 148
	/* If packet is not gso and we are resolving any partial checksum,
	 * clear encapsulation flag. This allows setting CHECKSUM_PARTIAL
	 * on the outer header without confusing devices that implement
	 * NETIF_F_IP_CSUM with encapsulation.
	 */
	if (csum_help)
		skb->encapsulation = 0;

149 150 151 152 153 154 155 156 157 158 159 160 161
	if (skb->ip_summed == CHECKSUM_PARTIAL && csum_help) {
		err = skb_checksum_help(skb);
		if (unlikely(err))
			goto error;
	} else if (skb->ip_summed != CHECKSUM_PARTIAL)
		skb->ip_summed = CHECKSUM_NONE;

	return skb;
error:
	kfree_skb(skb);
	return ERR_PTR(err);
}
EXPORT_SYMBOL_GPL(iptunnel_handle_offloads);
162 163 164 165 166 167 168

/* 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;

169 170
	netdev_stats_to_stats64(tot, &dev->stats);

171 172 173 174 175 176 177
	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 {
178
			start = u64_stats_fetch_begin_irq(&tstats->syncp);
179 180 181 182
			rx_packets = tstats->rx_packets;
			tx_packets = tstats->tx_packets;
			rx_bytes = tstats->rx_bytes;
			tx_bytes = tstats->tx_bytes;
183
		} while (u64_stats_fetch_retry_irq(&tstats->syncp, start));
184 185 186 187 188 189 190 191 192 193

		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);
194

195 196 197 198 199 200 201 202 203
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_SPORT]	= { .type = NLA_U16 },
	[LWTUNNEL_IP_DPORT]	= { .type = NLA_U16 },
	[LWTUNNEL_IP_FLAGS]	= { .type = NLA_U16 },
204 205 206
};

static int ip_tun_build_state(struct net_device *dev, struct nlattr *attr,
207
			      unsigned int family, const void *cfg,
208 209 210 211
			      struct lwtunnel_state **ts)
{
	struct ip_tunnel_info *tun_info;
	struct lwtunnel_state *new_state;
212
	struct nlattr *tb[LWTUNNEL_IP_MAX + 1];
213 214
	int err;

215
	err = nla_parse_nested(tb, LWTUNNEL_IP_MAX, attr, ip_tun_policy);
216 217 218 219 220 221 222 223 224 225 226
	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);

227 228
	if (tb[LWTUNNEL_IP_ID])
		tun_info->key.tun_id = nla_get_u64(tb[LWTUNNEL_IP_ID]);
229

230
	if (tb[LWTUNNEL_IP_DST])
231
		tun_info->key.u.ipv4.dst = nla_get_be32(tb[LWTUNNEL_IP_DST]);
232

233
	if (tb[LWTUNNEL_IP_SRC])
234
		tun_info->key.u.ipv4.src = nla_get_be32(tb[LWTUNNEL_IP_SRC]);
235

236
	if (tb[LWTUNNEL_IP_TTL])
237
		tun_info->key.ttl = nla_get_u8(tb[LWTUNNEL_IP_TTL]);
238

239
	if (tb[LWTUNNEL_IP_TOS])
240
		tun_info->key.tos = nla_get_u8(tb[LWTUNNEL_IP_TOS]);
241

242 243
	if (tb[LWTUNNEL_IP_SPORT])
		tun_info->key.tp_src = nla_get_be16(tb[LWTUNNEL_IP_SPORT]);
244

245 246
	if (tb[LWTUNNEL_IP_DPORT])
		tun_info->key.tp_dst = nla_get_be16(tb[LWTUNNEL_IP_DPORT]);
247

248 249
	if (tb[LWTUNNEL_IP_FLAGS])
		tun_info->key.tun_flags = nla_get_u16(tb[LWTUNNEL_IP_FLAGS]);
250 251 252 253 254 255 256 257 258 259 260 261 262 263

	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);

264
	if (nla_put_u64(skb, LWTUNNEL_IP_ID, tun_info->key.tun_id) ||
265 266
	    nla_put_be32(skb, LWTUNNEL_IP_DST, tun_info->key.u.ipv4.dst) ||
	    nla_put_be32(skb, LWTUNNEL_IP_SRC, tun_info->key.u.ipv4.src) ||
267 268
	    nla_put_u8(skb, LWTUNNEL_IP_TOS, tun_info->key.tos) ||
	    nla_put_u8(skb, LWTUNNEL_IP_TTL, tun_info->key.ttl) ||
269 270 271
	    nla_put_u16(skb, LWTUNNEL_IP_SPORT, tun_info->key.tp_src) ||
	    nla_put_u16(skb, LWTUNNEL_IP_DPORT, tun_info->key.tp_dst) ||
	    nla_put_u16(skb, LWTUNNEL_IP_FLAGS, tun_info->key.tun_flags))
272 273 274 275 276 277 278
		return -ENOMEM;

	return 0;
}

static int ip_tun_encap_nlsize(struct lwtunnel_state *lwtstate)
{
279 280 281 282 283 284 285 286
	return nla_total_size(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_SPORT */
		+ nla_total_size(2)	/* LWTUNNEL_IP_DPORT */
		+ nla_total_size(2);	/* LWTUNNEL_IP_FLAGS */
287 288
}

289 290 291 292 293 294
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));
}

295 296 297 298
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,
299
	.cmp_encap = ip_tun_cmp_encap,
300 301
};

302 303 304 305 306 307 308 309 310 311 312 313
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_SPORT]		= { .type = NLA_U16 },
	[LWTUNNEL_IP6_DPORT]		= { .type = NLA_U16 },
	[LWTUNNEL_IP6_FLAGS]		= { .type = NLA_U16 },
};

static int ip6_tun_build_state(struct net_device *dev, struct nlattr *attr,
314
			       unsigned int family, const void *cfg,
315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357
			       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_u64(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_SPORT])
		tun_info->key.tp_src = nla_get_be16(tb[LWTUNNEL_IP6_SPORT]);

	if (tb[LWTUNNEL_IP6_DPORT])
		tun_info->key.tp_dst = nla_get_be16(tb[LWTUNNEL_IP6_DPORT]);

	if (tb[LWTUNNEL_IP6_FLAGS])
		tun_info->key.tun_flags = nla_get_u16(tb[LWTUNNEL_IP6_FLAGS]);

358
	tun_info->mode = IP_TUNNEL_INFO_TX | IP_TUNNEL_INFO_IPV6;
359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402
	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_u64(skb, LWTUNNEL_IP6_ID, tun_info->key.tun_id) ||
	    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_HOPLIMIT, tun_info->key.tos) ||
	    nla_put_u8(skb, LWTUNNEL_IP6_TC, tun_info->key.ttl) ||
	    nla_put_u16(skb, LWTUNNEL_IP6_SPORT, tun_info->key.tp_src) ||
	    nla_put_u16(skb, LWTUNNEL_IP6_DPORT, tun_info->key.tp_dst) ||
	    nla_put_u16(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(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_SPORT */
		+ nla_total_size(2)	/* LWTUNNEL_IP6_DPORT */
		+ 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,
};

403
void __init ip_tunnel_core_init(void)
404 405
{
	lwtunnel_encap_add_ops(&ip_tun_lwt_ops, LWTUNNEL_ENCAP_IP);
406
	lwtunnel_encap_add_ops(&ip6_tun_lwt_ops, LWTUNNEL_ENCAP_IP6);
407
}
408 409 410 411 412 413 414 415 416 417 418 419 420 421 422

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);