l2tp_core.c 51.4 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
/*
 * L2TP core.
 *
 * Copyright (c) 2008,2009,2010 Katalix Systems Ltd
 *
 * This file contains some code of the original L2TPv2 pppol2tp
 * driver, which has the following copyright:
 *
 * Authors:	Martijn van Oosterhout <kleptog@svana.org>
 *		James Chapman (jchapman@katalix.com)
 * Contributors:
 *		Michal Ostrowski <mostrows@speakeasy.net>
 *		Arnaldo Carvalho de Melo <acme@xconectiva.com.br>
 *		David S. Miller (davem@redhat.com)
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 */

21 22
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt

23 24 25
#include <linux/module.h>
#include <linux/string.h>
#include <linux/list.h>
James Chapman's avatar
James Chapman committed
26
#include <linux/rculist.h>
27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
#include <linux/uaccess.h>

#include <linux/kernel.h>
#include <linux/spinlock.h>
#include <linux/kthread.h>
#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/errno.h>
#include <linux/jiffies.h>

#include <linux/netdevice.h>
#include <linux/net.h>
#include <linux/inetdevice.h>
#include <linux/skbuff.h>
#include <linux/init.h>
42
#include <linux/in.h>
43 44
#include <linux/ip.h>
#include <linux/udp.h>
45
#include <linux/l2tp.h>
46 47 48 49 50 51 52 53 54
#include <linux/hash.h>
#include <linux/sort.h>
#include <linux/file.h>
#include <linux/nsproxy.h>
#include <net/net_namespace.h>
#include <net/netns/generic.h>
#include <net/dst.h>
#include <net/ip.h>
#include <net/udp.h>
Tom Herbert's avatar
Tom Herbert committed
55
#include <net/udp_tunnel.h>
56
#include <net/inet_common.h>
57
#include <net/xfrm.h>
58
#include <net/protocol.h>
59 60 61
#include <net/inet6_connection_sock.h>
#include <net/inet_ecn.h>
#include <net/ip6_route.h>
62
#include <net/ip6_checksum.h>
63 64

#include <asm/byteorder.h>
Arun Sharma's avatar
Arun Sharma committed
65
#include <linux/atomic.h>
66 67 68 69 70 71 72 73 74 75 76 77 78 79

#include "l2tp_core.h"

#define L2TP_DRV_VERSION	"V2.0"

/* L2TP header constants */
#define L2TP_HDRFLAG_T	   0x8000
#define L2TP_HDRFLAG_L	   0x4000
#define L2TP_HDRFLAG_S	   0x0800
#define L2TP_HDRFLAG_O	   0x0200
#define L2TP_HDRFLAG_P	   0x0100

#define L2TP_HDR_VER_MASK  0x000F
#define L2TP_HDR_VER_2	   0x0002
80
#define L2TP_HDR_VER_3	   0x0003
81 82 83 84 85 86 87 88 89 90 91 92 93 94

/* L2TPv3 default L2-specific sublayer */
#define L2TP_SLFLAG_S	   0x40000000
#define L2TP_SL_SEQ_MASK   0x00ffffff

#define L2TP_HDR_SIZE_SEQ		10
#define L2TP_HDR_SIZE_NOSEQ		6

/* Default trace flags */
#define L2TP_DEFAULT_DEBUG_FLAGS	0

/* Private data stored for received packets in the skb.
 */
struct l2tp_skb_cb {
95
	u32			ns;
96 97 98 99 100 101 102 103 104
	u16			has_seq;
	u16			length;
	unsigned long		expires;
};

#define L2TP_SKB_CB(skb)	((struct l2tp_skb_cb *) &skb->cb[sizeof(struct inet_skb_parm)])

static atomic_t l2tp_tunnel_count;
static atomic_t l2tp_session_count;
105
static struct workqueue_struct *l2tp_wq;
106 107 108 109 110

/* per-net private data for this module */
static unsigned int l2tp_net_id;
struct l2tp_net {
	struct list_head l2tp_tunnel_list;
James Chapman's avatar
James Chapman committed
111
	spinlock_t l2tp_tunnel_list_lock;
112
	struct hlist_head l2tp_session_hlist[L2TP_HASH_SIZE_2];
James Chapman's avatar
James Chapman committed
113
	spinlock_t l2tp_session_hlist_lock;
114 115
};

116 117 118 119 120 121 122
#if IS_ENABLED(CONFIG_IPV6)
static bool l2tp_sk_is_v6(struct sock *sk)
{
	return sk->sk_family == PF_INET6 &&
	       !ipv6_addr_v4mapped(&sk->sk_v6_daddr);
}
#endif
123

124 125 126 127 128
static inline struct l2tp_tunnel *l2tp_tunnel(struct sock *sk)
{
	return sk->sk_user_data;
}

129
static inline struct l2tp_net *l2tp_pernet(const struct net *net)
130 131 132 133 134 135
{
	BUG_ON(!net);

	return net_generic(net, l2tp_net_id);
}

136 137 138 139 140 141 142 143 144 145 146 147
/* Session hash global list for L2TPv3.
 * The session_id SHOULD be random according to RFC3931, but several
 * L2TP implementations use incrementing session_ids.  So we do a real
 * hash on the session_id, rather than a simple bitmask.
 */
static inline struct hlist_head *
l2tp_session_id_hash_2(struct l2tp_net *pn, u32 session_id)
{
	return &pn->l2tp_session_hlist[hash_32(session_id, L2TP_HASH_BITS_2)];

}

148 149 150 151
/* Lookup the tunnel socket, possibly involving the fs code if the socket is
 * owned by userspace.  A struct sock returned from this function must be
 * released using l2tp_tunnel_sock_put once you're done with it.
 */
152
static struct sock *l2tp_tunnel_sock_lookup(struct l2tp_tunnel *tunnel)
153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171
{
	int err = 0;
	struct socket *sock = NULL;
	struct sock *sk = NULL;

	if (!tunnel)
		goto out;

	if (tunnel->fd >= 0) {
		/* Socket is owned by userspace, who might be in the process
		 * of closing it.  Look the socket up using the fd to ensure
		 * consistency.
		 */
		sock = sockfd_lookup(tunnel->fd, &err);
		if (sock)
			sk = sock->sk;
	} else {
		/* Socket is owned by kernelspace */
		sk = tunnel->sock;
172
		sock_hold(sk);
173 174 175 176 177 178 179
	}

out:
	return sk;
}

/* Drop a reference to a tunnel socket obtained via. l2tp_tunnel_sock_put */
180
static void l2tp_tunnel_sock_put(struct sock *sk)
181 182 183 184 185 186 187 188 189
{
	struct l2tp_tunnel *tunnel = l2tp_sock_to_tunnel(sk);
	if (tunnel) {
		if (tunnel->fd >= 0) {
			/* Socket is owned by userspace */
			sockfd_put(sk->sk_socket);
		}
		sock_put(sk);
	}
190
	sock_put(sk);
191 192
}

193 194 195 196 197 198 199 200 201 202 203 204
/* Session hash list.
 * The session_id SHOULD be random according to RFC2661, but several
 * L2TP implementations (Cisco and Microsoft) use incrementing
 * session_ids.  So we do a real hash on the session_id, rather than a
 * simple bitmask.
 */
static inline struct hlist_head *
l2tp_session_id_hash(struct l2tp_tunnel *tunnel, u32 session_id)
{
	return &tunnel->session_hlist[hash_32(session_id, L2TP_HASH_BITS)];
}

205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225
/* Lookup a tunnel. A new reference is held on the returned tunnel. */
struct l2tp_tunnel *l2tp_tunnel_get(const struct net *net, u32 tunnel_id)
{
	const struct l2tp_net *pn = l2tp_pernet(net);
	struct l2tp_tunnel *tunnel;

	rcu_read_lock_bh();
	list_for_each_entry_rcu(tunnel, &pn->l2tp_tunnel_list, list) {
		if (tunnel->tunnel_id == tunnel_id) {
			l2tp_tunnel_inc_refcount(tunnel);
			rcu_read_unlock_bh();

			return tunnel;
		}
	}
	rcu_read_unlock_bh();

	return NULL;
}
EXPORT_SYMBOL_GPL(l2tp_tunnel_get);

226
/* Lookup a session. A new reference is held on the returned session.
227 228
 * Optionally calls session->ref() too if do_ref is true.
 */
229
struct l2tp_session *l2tp_session_get(const struct net *net,
230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274
				      struct l2tp_tunnel *tunnel,
				      u32 session_id, bool do_ref)
{
	struct hlist_head *session_list;
	struct l2tp_session *session;

	if (!tunnel) {
		struct l2tp_net *pn = l2tp_pernet(net);

		session_list = l2tp_session_id_hash_2(pn, session_id);

		rcu_read_lock_bh();
		hlist_for_each_entry_rcu(session, session_list, global_hlist) {
			if (session->session_id == session_id) {
				l2tp_session_inc_refcount(session);
				if (do_ref && session->ref)
					session->ref(session);
				rcu_read_unlock_bh();

				return session;
			}
		}
		rcu_read_unlock_bh();

		return NULL;
	}

	session_list = l2tp_session_id_hash(tunnel, session_id);
	read_lock_bh(&tunnel->hlist_lock);
	hlist_for_each_entry(session, session_list, hlist) {
		if (session->session_id == session_id) {
			l2tp_session_inc_refcount(session);
			if (do_ref && session->ref)
				session->ref(session);
			read_unlock_bh(&tunnel->hlist_lock);

			return session;
		}
	}
	read_unlock_bh(&tunnel->hlist_lock);

	return NULL;
}
EXPORT_SYMBOL_GPL(l2tp_session_get);

275 276
struct l2tp_session *l2tp_session_get_nth(struct l2tp_tunnel *tunnel, int nth,
					  bool do_ref)
277 278 279 280 281 282 283
{
	int hash;
	struct l2tp_session *session;
	int count = 0;

	read_lock_bh(&tunnel->hlist_lock);
	for (hash = 0; hash < L2TP_HASH_SIZE; hash++) {
284
		hlist_for_each_entry(session, &tunnel->session_hlist[hash], hlist) {
285
			if (++count > nth) {
286 287 288
				l2tp_session_inc_refcount(session);
				if (do_ref && session->ref)
					session->ref(session);
289 290 291 292 293 294 295 296 297 298
				read_unlock_bh(&tunnel->hlist_lock);
				return session;
			}
		}
	}

	read_unlock_bh(&tunnel->hlist_lock);

	return NULL;
}
299
EXPORT_SYMBOL_GPL(l2tp_session_get_nth);
300

301 302 303
/* Lookup a session by interface name.
 * This is very inefficient but is only used by management interfaces.
 */
304 305
struct l2tp_session *l2tp_session_get_by_ifname(const struct net *net,
						const char *ifname,
306
						bool do_ref)
307 308 309 310 311
{
	struct l2tp_net *pn = l2tp_pernet(net);
	int hash;
	struct l2tp_session *session;

James Chapman's avatar
James Chapman committed
312
	rcu_read_lock_bh();
313
	for (hash = 0; hash < L2TP_HASH_SIZE_2; hash++) {
314
		hlist_for_each_entry_rcu(session, &pn->l2tp_session_hlist[hash], global_hlist) {
315
			if (!strcmp(session->ifname, ifname)) {
316 317 318
				l2tp_session_inc_refcount(session);
				if (do_ref && session->ref)
					session->ref(session);
James Chapman's avatar
James Chapman committed
319
				rcu_read_unlock_bh();
320

321 322 323 324 325
				return session;
			}
		}
	}

James Chapman's avatar
James Chapman committed
326
	rcu_read_unlock_bh();
327 328 329

	return NULL;
}
330
EXPORT_SYMBOL_GPL(l2tp_session_get_by_ifname);
331

332 333 334 335 336 337 338
static int l2tp_session_add_to_tunnel(struct l2tp_tunnel *tunnel,
				      struct l2tp_session *session)
{
	struct l2tp_session *session_walk;
	struct hlist_head *g_head;
	struct hlist_head *head;
	struct l2tp_net *pn;
339
	int err;
340 341 342 343

	head = l2tp_session_id_hash(tunnel, session->session_id);

	write_lock_bh(&tunnel->hlist_lock);
344 345 346 347 348
	if (!tunnel->acpt_newsess) {
		err = -ENODEV;
		goto err_tlock;
	}

349
	hlist_for_each_entry(session_walk, head, hlist)
350 351 352 353
		if (session_walk->session_id == session->session_id) {
			err = -EEXIST;
			goto err_tlock;
		}
354 355 356 357 358 359 360

	if (tunnel->version == L2TP_HDR_VER_3) {
		pn = l2tp_pernet(tunnel->l2tp_net);
		g_head = l2tp_session_id_hash_2(l2tp_pernet(tunnel->l2tp_net),
						session->session_id);

		spin_lock_bh(&pn->l2tp_session_hlist_lock);
361

362
		hlist_for_each_entry(session_walk, g_head, global_hlist)
363 364 365 366
			if (session_walk->session_id == session->session_id) {
				err = -EEXIST;
				goto err_tlock_pnlock;
			}
367

368 369
		l2tp_tunnel_inc_refcount(tunnel);
		sock_hold(tunnel->sock);
370
		hlist_add_head_rcu(&session->global_hlist, g_head);
371

372
		spin_unlock_bh(&pn->l2tp_session_hlist_lock);
373 374 375
	} else {
		l2tp_tunnel_inc_refcount(tunnel);
		sock_hold(tunnel->sock);
376 377 378 379 380 381 382
	}

	hlist_add_head(&session->hlist, head);
	write_unlock_bh(&tunnel->hlist_lock);

	return 0;

383
err_tlock_pnlock:
384
	spin_unlock_bh(&pn->l2tp_session_hlist_lock);
385
err_tlock:
386 387
	write_unlock_bh(&tunnel->hlist_lock);

388
	return err;
389 390
}

391 392
/* Lookup a tunnel by id
 */
393
struct l2tp_tunnel *l2tp_tunnel_find(const struct net *net, u32 tunnel_id)
394 395 396 397
{
	struct l2tp_tunnel *tunnel;
	struct l2tp_net *pn = l2tp_pernet(net);

James Chapman's avatar
James Chapman committed
398 399
	rcu_read_lock_bh();
	list_for_each_entry_rcu(tunnel, &pn->l2tp_tunnel_list, list) {
400
		if (tunnel->tunnel_id == tunnel_id) {
James Chapman's avatar
James Chapman committed
401
			rcu_read_unlock_bh();
402 403 404
			return tunnel;
		}
	}
James Chapman's avatar
James Chapman committed
405
	rcu_read_unlock_bh();
406 407 408 409 410

	return NULL;
}
EXPORT_SYMBOL_GPL(l2tp_tunnel_find);

411
struct l2tp_tunnel *l2tp_tunnel_find_nth(const struct net *net, int nth)
412 413 414 415 416
{
	struct l2tp_net *pn = l2tp_pernet(net);
	struct l2tp_tunnel *tunnel;
	int count = 0;

James Chapman's avatar
James Chapman committed
417 418
	rcu_read_lock_bh();
	list_for_each_entry_rcu(tunnel, &pn->l2tp_tunnel_list, list) {
419
		if (++count > nth) {
James Chapman's avatar
James Chapman committed
420
			rcu_read_unlock_bh();
421 422 423 424
			return tunnel;
		}
	}

James Chapman's avatar
James Chapman committed
425
	rcu_read_unlock_bh();
426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441

	return NULL;
}
EXPORT_SYMBOL_GPL(l2tp_tunnel_find_nth);

/*****************************************************************************
 * Receive data handling
 *****************************************************************************/

/* Queue a skb in order. We come here only if the skb has an L2TP sequence
 * number.
 */
static void l2tp_recv_queue_skb(struct l2tp_session *session, struct sk_buff *skb)
{
	struct sk_buff *skbp;
	struct sk_buff *tmp;
442
	u32 ns = L2TP_SKB_CB(skb)->ns;
443 444 445 446 447

	spin_lock_bh(&session->reorder_q.lock);
	skb_queue_walk_safe(&session->reorder_q, skbp, tmp) {
		if (L2TP_SKB_CB(skbp)->ns > ns) {
			__skb_queue_before(&session->reorder_q, skbp, skb);
448 449 450 451
			l2tp_dbg(session, L2TP_MSG_SEQ,
				 "%s: pkt %hu, inserted before %hu, reorder_q len=%d\n",
				 session->name, ns, L2TP_SKB_CB(skbp)->ns,
				 skb_queue_len(&session->reorder_q));
452
			atomic_long_inc(&session->stats.rx_oos_packets);
453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474
			goto out;
		}
	}

	__skb_queue_tail(&session->reorder_q, skb);

out:
	spin_unlock_bh(&session->reorder_q.lock);
}

/* Dequeue a single skb.
 */
static void l2tp_recv_dequeue_skb(struct l2tp_session *session, struct sk_buff *skb)
{
	struct l2tp_tunnel *tunnel = session->tunnel;
	int length = L2TP_SKB_CB(skb)->length;

	/* We're about to requeue the skb, so return resources
	 * to its current owner (a socket receive buffer).
	 */
	skb_orphan(skb);

475 476 477 478
	atomic_long_inc(&tunnel->stats.rx_packets);
	atomic_long_add(length, &tunnel->stats.rx_bytes);
	atomic_long_inc(&session->stats.rx_packets);
	atomic_long_add(length, &session->stats.rx_bytes);
479 480 481 482

	if (L2TP_SKB_CB(skb)->has_seq) {
		/* Bump our Nr */
		session->nr++;
483
		session->nr &= session->nr_max;
484

485 486
		l2tp_dbg(session, L2TP_MSG_SEQ, "%s: updated nr to %hu\n",
			 session->name, session->nr);
487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510
	}

	/* call private receive handler */
	if (session->recv_skb != NULL)
		(*session->recv_skb)(session, skb, L2TP_SKB_CB(skb)->length);
	else
		kfree_skb(skb);

	if (session->deref)
		(*session->deref)(session);
}

/* Dequeue skbs from the session's reorder_q, subject to packet order.
 * Skbs that have been in the queue for too long are simply discarded.
 */
static void l2tp_recv_dequeue(struct l2tp_session *session)
{
	struct sk_buff *skb;
	struct sk_buff *tmp;

	/* If the pkt at the head of the queue has the nr that we
	 * expect to send up next, dequeue it and any other
	 * in-sequence packets behind it.
	 */
511
start:
512 513 514
	spin_lock_bh(&session->reorder_q.lock);
	skb_queue_walk_safe(&session->reorder_q, skb, tmp) {
		if (time_after(jiffies, L2TP_SKB_CB(skb)->expires)) {
515 516
			atomic_long_inc(&session->stats.rx_seq_discards);
			atomic_long_inc(&session->stats.rx_errors);
517 518 519 520 521
			l2tp_dbg(session, L2TP_MSG_SEQ,
				 "%s: oos pkt %u len %d discarded (too old), waiting for %u, reorder_q_len=%d\n",
				 session->name, L2TP_SKB_CB(skb)->ns,
				 L2TP_SKB_CB(skb)->length, session->nr,
				 skb_queue_len(&session->reorder_q));
522
			session->reorder_skip = 1;
523 524 525 526 527 528 529 530
			__skb_unlink(skb, &session->reorder_q);
			kfree_skb(skb);
			if (session->deref)
				(*session->deref)(session);
			continue;
		}

		if (L2TP_SKB_CB(skb)->has_seq) {
531
			if (session->reorder_skip) {
532 533 534 535
				l2tp_dbg(session, L2TP_MSG_SEQ,
					 "%s: advancing nr to next pkt: %u -> %u",
					 session->name, session->nr,
					 L2TP_SKB_CB(skb)->ns);
536 537 538
				session->reorder_skip = 0;
				session->nr = L2TP_SKB_CB(skb)->ns;
			}
539
			if (L2TP_SKB_CB(skb)->ns != session->nr) {
540 541 542 543 544
				l2tp_dbg(session, L2TP_MSG_SEQ,
					 "%s: holding oos pkt %u len %d, waiting for %u, reorder_q_len=%d\n",
					 session->name, L2TP_SKB_CB(skb)->ns,
					 L2TP_SKB_CB(skb)->length, session->nr,
					 skb_queue_len(&session->reorder_q));
545 546 547 548 549 550 551 552 553 554
				goto out;
			}
		}
		__skb_unlink(skb, &session->reorder_q);

		/* Process the skb. We release the queue lock while we
		 * do so to let other contexts process the queue.
		 */
		spin_unlock_bh(&session->reorder_q.lock);
		l2tp_recv_dequeue_skb(session, skb);
555
		goto start;
556 557 558 559 560 561
	}

out:
	spin_unlock_bh(&session->reorder_q.lock);
}

562 563 564 565 566 567 568 569 570 571 572 573
static int l2tp_seq_check_rx_window(struct l2tp_session *session, u32 nr)
{
	u32 nws;

	if (nr >= session->nr)
		nws = nr - session->nr;
	else
		nws = (session->nr_max + 1) - (session->nr - nr);

	return nws < session->nr_window_size;
}

574 575 576 577 578
/* If packet has sequence numbers, queue it if acceptable. Returns 0 if
 * acceptable, else non-zero.
 */
static int l2tp_recv_data_seq(struct l2tp_session *session, struct sk_buff *skb)
{
579 580 581 582 583 584 585 586 587 588 589
	if (!l2tp_seq_check_rx_window(session, L2TP_SKB_CB(skb)->ns)) {
		/* Packet sequence number is outside allowed window.
		 * Discard it.
		 */
		l2tp_dbg(session, L2TP_MSG_SEQ,
			 "%s: pkt %u len %d discarded, outside window, nr=%u\n",
			 session->name, L2TP_SKB_CB(skb)->ns,
			 L2TP_SKB_CB(skb)->length, session->nr);
		goto discard;
	}

590 591 592 593 594
	if (session->reorder_timeout != 0) {
		/* Packet reordering enabled. Add skb to session's
		 * reorder queue, in order of ns.
		 */
		l2tp_recv_queue_skb(session, skb);
595 596 597 598 599 600 601 602 603 604
		goto out;
	}

	/* Packet reordering disabled. Discard out-of-sequence packets, while
	 * tracking the number if in-sequence packets after the first OOS packet
	 * is seen. After nr_oos_count_max in-sequence packets, reset the
	 * sequence number to re-enable packet reception.
	 */
	if (L2TP_SKB_CB(skb)->ns == session->nr) {
		skb_queue_tail(&session->reorder_q, skb);
605
	} else {
606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621
		u32 nr_oos = L2TP_SKB_CB(skb)->ns;
		u32 nr_next = (session->nr_oos + 1) & session->nr_max;

		if (nr_oos == nr_next)
			session->nr_oos_count++;
		else
			session->nr_oos_count = 0;

		session->nr_oos = nr_oos;
		if (session->nr_oos_count > session->nr_oos_count_max) {
			session->reorder_skip = 1;
			l2tp_dbg(session, L2TP_MSG_SEQ,
				 "%s: %d oos packets received. Resetting sequence numbers\n",
				 session->name, session->nr_oos_count);
		}
		if (!session->reorder_skip) {
622 623 624 625 626 627 628 629 630 631 632
			atomic_long_inc(&session->stats.rx_seq_discards);
			l2tp_dbg(session, L2TP_MSG_SEQ,
				 "%s: oos pkt %u len %d discarded, waiting for %u, reorder_q_len=%d\n",
				 session->name, L2TP_SKB_CB(skb)->ns,
				 L2TP_SKB_CB(skb)->length, session->nr,
				 skb_queue_len(&session->reorder_q));
			goto discard;
		}
		skb_queue_tail(&session->reorder_q, skb);
	}

633
out:
634 635 636 637 638 639
	return 0;

discard:
	return 1;
}

640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698
/* Do receive processing of L2TP data frames. We handle both L2TPv2
 * and L2TPv3 data frames here.
 *
 * L2TPv2 Data Message Header
 *
 *  0                   1                   2                   3
 *  0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |T|L|x|x|S|x|O|P|x|x|x|x|  Ver  |          Length (opt)         |
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |           Tunnel ID           |           Session ID          |
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |             Ns (opt)          |             Nr (opt)          |
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |      Offset Size (opt)        |    Offset pad... (opt)
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 *
 * Data frames are marked by T=0. All other fields are the same as
 * those in L2TP control frames.
 *
 * L2TPv3 Data Message Header
 *
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |                      L2TP Session Header                      |
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |                      L2-Specific Sublayer                     |
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |                        Tunnel Payload                      ...
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 *
 * L2TPv3 Session Header Over IP
 *
 *  0                   1                   2                   3
 *  0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |                           Session ID                          |
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |               Cookie (optional, maximum 64 bits)...
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 *                                                                 |
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 *
 * L2TPv3 L2-Specific Sublayer Format
 *
 *  0                   1                   2                   3
 *  0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |x|S|x|x|x|x|x|x|              Sequence Number                  |
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 *
 * Cookie value, sublayer format and offset (pad) are negotiated with
 * the peer when the session is set up. Unlike L2TPv2, we do not need
 * to parse the packet header to determine if optional fields are
 * present.
 *
 * Caller must already have parsed the frame and determined that it is
 * a data (not control) frame before coming here. Fields up to the
 * session-id have already been parsed and ptr points to the data
 * after the session-id.
699 700 701
 *
 * session->ref() must have been called prior to l2tp_recv_common().
 * session->deref() will be called automatically after skb is processed.
702
 */
703 704 705
void l2tp_recv_common(struct l2tp_session *session, struct sk_buff *skb,
		      unsigned char *ptr, unsigned char *optr, u16 hdrflags,
		      int length, int (*payload_hook)(struct sk_buff *skb))
706
{
707
	struct l2tp_tunnel *tunnel = session->tunnel;
708
	int offset;
709
	u32 ns, nr;
710

711 712 713
	/* Parse and check optional cookie */
	if (session->peer_cookie_len > 0) {
		if (memcmp(ptr, &session->peer_cookie[0], session->peer_cookie_len)) {
714 715 716 717
			l2tp_info(tunnel, L2TP_MSG_DATA,
				  "%s: cookie mismatch (%u/%u). Discarding.\n",
				  tunnel->name, tunnel->tunnel_id,
				  session->session_id);
718
			atomic_long_inc(&session->stats.rx_cookie_discards);
719 720 721 722 723
			goto discard;
		}
		ptr += session->peer_cookie_len;
	}

724 725 726 727 728 729 730 731 732
	/* Handle the optional sequence numbers. Sequence numbers are
	 * in different places for L2TPv2 and L2TPv3.
	 *
	 * If we are the LAC, enable/disable sequence numbers under
	 * the control of the LNS.  If no sequence numbers present but
	 * we were expecting them, discard frame.
	 */
	ns = nr = 0;
	L2TP_SKB_CB(skb)->has_seq = 0;
733 734 735 736 737 738
	if (tunnel->version == L2TP_HDR_VER_2) {
		if (hdrflags & L2TP_HDRFLAG_S) {
			ns = ntohs(*(__be16 *) ptr);
			ptr += 2;
			nr = ntohs(*(__be16 *) ptr);
			ptr += 2;
739

740 741 742
			/* Store L2TP info in the skb */
			L2TP_SKB_CB(skb)->ns = ns;
			L2TP_SKB_CB(skb)->has_seq = 1;
743

744 745 746
			l2tp_dbg(session, L2TP_MSG_SEQ,
				 "%s: recv data ns=%u, nr=%u, session nr=%u\n",
				 session->name, ns, nr, session->nr);
747 748 749 750 751 752 753 754 755 756 757
		}
	} else if (session->l2specific_type == L2TP_L2SPECTYPE_DEFAULT) {
		u32 l2h = ntohl(*(__be32 *) ptr);

		if (l2h & 0x40000000) {
			ns = l2h & 0x00ffffff;

			/* Store L2TP info in the skb */
			L2TP_SKB_CB(skb)->ns = ns;
			L2TP_SKB_CB(skb)->has_seq = 1;

758 759 760
			l2tp_dbg(session, L2TP_MSG_SEQ,
				 "%s: recv data ns=%u, session nr=%u\n",
				 session->name, ns, session->nr);
761
		}
762 763
	}

764 765 766
	/* Advance past L2-specific header, if present */
	ptr += session->l2specific_len;

767 768 769 770 771 772
	if (L2TP_SKB_CB(skb)->has_seq) {
		/* Received a packet with sequence numbers. If we're the LNS,
		 * check if we sre sending sequence numbers and if not,
		 * configure it so.
		 */
		if ((!session->lns_mode) && (!session->send_seq)) {
773 774 775
			l2tp_info(session, L2TP_MSG_SEQ,
				  "%s: requested to enable seq numbers by LNS\n",
				  session->name);
776
			session->send_seq = 1;
777
			l2tp_session_set_header_len(session, tunnel->version);
778 779 780 781 782 783
		}
	} else {
		/* No sequence numbers.
		 * If user has configured mandatory sequence numbers, discard.
		 */
		if (session->recv_seq) {
784 785 786
			l2tp_warn(session, L2TP_MSG_SEQ,
				  "%s: recv data has no seq numbers when required. Discarding.\n",
				  session->name);
787
			atomic_long_inc(&session->stats.rx_seq_discards);
788 789 790 791 792 793 794 795 796
			goto discard;
		}

		/* If we're the LAC and we're sending sequence numbers, the
		 * LNS has requested that we no longer send sequence numbers.
		 * If we're the LNS and we're sending sequence numbers, the
		 * LAC is broken. Discard the frame.
		 */
		if ((!session->lns_mode) && (session->send_seq)) {
797 798 799
			l2tp_info(session, L2TP_MSG_SEQ,
				  "%s: requested to disable seq numbers by LNS\n",
				  session->name);
800
			session->send_seq = 0;
801
			l2tp_session_set_header_len(session, tunnel->version);
802
		} else if (session->send_seq) {
803 804 805
			l2tp_warn(session, L2TP_MSG_SEQ,
				  "%s: recv data has no seq numbers when required. Discarding.\n",
				  session->name);
806
			atomic_long_inc(&session->stats.rx_seq_discards);
807 808 809 810
			goto discard;
		}
	}

811 812 813 814 815 816 817 818 819 820 821 822 823
	/* Session data offset is handled differently for L2TPv2 and
	 * L2TPv3. For L2TPv2, there is an optional 16-bit value in
	 * the header. For L2TPv3, the offset is negotiated using AVPs
	 * in the session setup control protocol.
	 */
	if (tunnel->version == L2TP_HDR_VER_2) {
		/* If offset bit set, skip it. */
		if (hdrflags & L2TP_HDRFLAG_O) {
			offset = ntohs(*(__be16 *)ptr);
			ptr += 2 + offset;
		}
	} else
		ptr += session->offset;
824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849

	offset = ptr - optr;
	if (!pskb_may_pull(skb, offset))
		goto discard;

	__skb_pull(skb, offset);

	/* If caller wants to process the payload before we queue the
	 * packet, do so now.
	 */
	if (payload_hook)
		if ((*payload_hook)(skb))
			goto discard;

	/* Prepare skb for adding to the session's reorder_q.  Hold
	 * packets for max reorder_timeout or 1 second if not
	 * reordering.
	 */
	L2TP_SKB_CB(skb)->length = length;
	L2TP_SKB_CB(skb)->expires = jiffies +
		(session->reorder_timeout ? session->reorder_timeout : HZ);

	/* Add packet to the session's receive queue. Reordering is done here, if
	 * enabled. Saved L2TP protocol info is stored in skb->sb[].
	 */
	if (L2TP_SKB_CB(skb)->has_seq) {
850 851
		if (l2tp_recv_data_seq(session, skb))
			goto discard;
852 853 854 855 856 857 858 859 860 861 862
	} else {
		/* No sequence numbers. Add the skb to the tail of the
		 * reorder queue. This ensures that it will be
		 * delivered after all previous sequenced skbs.
		 */
		skb_queue_tail(&session->reorder_q, skb);
	}

	/* Try to dequeue as many skbs from reorder_q as we can. */
	l2tp_recv_dequeue(session);

863
	return;
864 865

discard:
866
	atomic_long_inc(&session->stats.rx_errors);
867 868 869 870
	kfree_skb(skb);

	if (session->deref)
		(*session->deref)(session);
871 872 873
}
EXPORT_SYMBOL(l2tp_recv_common);

874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890
/* Drop skbs from the session's reorder_q
 */
int l2tp_session_queue_purge(struct l2tp_session *session)
{
	struct sk_buff *skb = NULL;
	BUG_ON(!session);
	BUG_ON(session->magic != L2TP_SESSION_MAGIC);
	while ((skb = skb_dequeue(&session->reorder_q))) {
		atomic_long_inc(&session->stats.rx_errors);
		kfree_skb(skb);
		if (session->deref)
			(*session->deref)(session);
	}
	return 0;
}
EXPORT_SYMBOL_GPL(l2tp_session_queue_purge);

891 892 893 894 895 896
/* Internal UDP receive frame. Do the real work of receiving an L2TP data frame
 * here. The skb is not on a list when we get here.
 * Returns 0 if the packet was a data packet and was successfully passed on.
 * Returns 1 if the packet was not a good data packet and could not be
 * forwarded.  All such packets are passed up to userspace to deal with.
 */
897 898
static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb,
			      int (*payload_hook)(struct sk_buff *skb))
899 900 901 902 903 904 905 906
{
	struct l2tp_session *session = NULL;
	unsigned char *ptr, *optr;
	u16 hdrflags;
	u32 tunnel_id, session_id;
	u16 version;
	int length;

907
	/* UDP has verifed checksum */
908 909 910 911 912 913

	/* UDP always verifies the packet length. */
	__skb_pull(skb, sizeof(struct udphdr));

	/* Short packet? */
	if (!pskb_may_pull(skb, L2TP_HDR_SIZE_SEQ)) {
914 915 916
		l2tp_info(tunnel, L2TP_MSG_DATA,
			  "%s: recv short packet (len=%d)\n",
			  tunnel->name, skb->len);
917 918 919 920 921 922 923 924 925
		goto error;
	}

	/* Trace packet contents, if enabled */
	if (tunnel->debug & L2TP_MSG_DATA) {
		length = min(32u, skb->len);
		if (!pskb_may_pull(skb, length))
			goto error;

926 927
		pr_debug("%s: recv\n", tunnel->name);
		print_hex_dump_bytes("", DUMP_PREFIX_OFFSET, skb->data, length);
928 929
	}

930 931 932
	/* Point to L2TP header */
	optr = ptr = skb->data;

933 934 935 936 937 938
	/* Get L2TP header flags */
	hdrflags = ntohs(*(__be16 *) ptr);

	/* Check protocol version */
	version = hdrflags & L2TP_HDR_VER_MASK;
	if (version != tunnel->version) {
939 940 941
		l2tp_info(tunnel, L2TP_MSG_DATA,
			  "%s: recv protocol version mismatch: got %d expected %d\n",
			  tunnel->name, version, tunnel->version);
942 943 944 945 946 947 948 949
		goto error;
	}

	/* Get length of L2TP packet */
	length = skb->len;

	/* If type is control packet, it is handled by userspace. */
	if (hdrflags & L2TP_HDRFLAG_T) {
950 951 952
		l2tp_dbg(tunnel, L2TP_MSG_DATA,
			 "%s: recv control packet, len=%d\n",
			 tunnel->name, length);
953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976
		goto error;
	}

	/* Skip flags */
	ptr += 2;

	if (tunnel->version == L2TP_HDR_VER_2) {
		/* If length is present, skip it */
		if (hdrflags & L2TP_HDRFLAG_L)
			ptr += 2;

		/* Extract tunnel and session ID */
		tunnel_id = ntohs(*(__be16 *) ptr);
		ptr += 2;
		session_id = ntohs(*(__be16 *) ptr);
		ptr += 2;
	} else {
		ptr += 2;	/* skip reserved bits */
		tunnel_id = tunnel->tunnel_id;
		session_id = ntohl(*(__be32 *) ptr);
		ptr += 4;
	}

	/* Find the session context */
977
	session = l2tp_session_get(tunnel->l2tp_net, tunnel, session_id, true);
978
	if (!session || !session->recv_skb) {
979 980 981 982 983 984
		if (session) {
			if (session->deref)
				session->deref(session);
			l2tp_session_dec_refcount(session);
		}

985
		/* Not found? Pass to userspace to deal with */
986 987 988
		l2tp_info(tunnel, L2TP_MSG_DATA,
			  "%s: no session found (%u/%u). Passing up.\n",
			  tunnel->name, tunnel_id, session_id);
989 990 991 992
		goto error;
	}

	l2tp_recv_common(session, skb, ptr, optr, hdrflags, length, payload_hook);
993
	l2tp_session_dec_refcount(session);
994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017

	return 0;

error:
	/* Put UDP header back */
	__skb_push(skb, sizeof(struct udphdr));

	return 1;
}

/* UDP encapsulation receive handler. See net/ipv4/udp.c.
 * Return codes:
 * 0 : success.
 * <0: error
 * >0: skb should be passed up to userspace as UDP.
 */
int l2tp_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
{
	struct l2tp_tunnel *tunnel;

	tunnel = l2tp_sock_to_tunnel(sk);
	if (tunnel == NULL)
		goto pass_up;

1018 1019
	l2tp_dbg(tunnel, L2TP_MSG_DATA, "%s: received %d bytes\n",
		 tunnel->name, skb->len);
1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039

	if (l2tp_udp_recv_core(tunnel, skb, tunnel->recv_payload_hook))
		goto pass_up_put;

	sock_put(sk);
	return 0;

pass_up_put:
	sock_put(sk);
pass_up:
	return 1;
}
EXPORT_SYMBOL_GPL(l2tp_udp_encap_recv);

/************************************************************************
 * Transmit handling
 ***********************************************************************/

/* Build an L2TP header for the session into the buffer provided.
 */
1040
static int l2tp_build_l2tpv2_header(struct l2tp_session *session, void *buf)
1041
{
1042
	struct l2tp_tunnel *tunnel = session->tunnel;
1043
	__be16 *bufp = buf;
1044
	__be16 *optr = buf;
1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059
	u16 flags = L2TP_HDR_VER_2;
	u32 tunnel_id = tunnel->peer_tunnel_id;
	u32 session_id = session->peer_session_id;

	if (session->send_seq)
		flags |= L2TP_HDRFLAG_S;

	/* Setup L2TP header. */
	*bufp++ = htons(flags);
	*bufp++ = htons(tunnel_id);
	*bufp++ = htons(session_id);
	if (session->send_seq) {
		*bufp++ = htons(session->ns);
		*bufp++ = 0;
		session->ns++;
1060
		session->ns &= 0xffff;
1061 1062
		l2tp_dbg(session, L2TP_MSG_SEQ, "%s: updated ns to %u\n",
			 session->name, session->ns);
1063
	}
1064 1065

	return bufp - optr;
1066 1067
}

1068
static int l2tp_build_l2tpv3_header(struct l2tp_session *session, void *buf)
1069
{
1070
	struct l2tp_tunnel *tunnel = session->tunnel;
1071 1072 1073
	char *bufp = buf;
	char *optr = bufp;

1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084
	/* Setup L2TP header. The header differs slightly for UDP and
	 * IP encapsulations. For UDP, there is 4 bytes of flags.
	 */
	if (tunnel->encap == L2TP_ENCAPTYPE_UDP) {
		u16 flags = L2TP_HDR_VER_3;
		*((__be16 *) bufp) = htons(flags);
		bufp += 2;
		*((__be16 *) bufp) = 0;
		bufp += 2;
	}

1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097
	*((__be32 *) bufp) = htonl(session->peer_session_id);
	bufp += 4;
	if (session->cookie_len) {
		memcpy(bufp, &session->cookie[0], session->cookie_len);
		bufp += session->cookie_len;
	}
	if (session->l2specific_len) {
		if (session->l2specific_type == L2TP_L2SPECTYPE_DEFAULT) {
			u32 l2h = 0;
			if (session->send_seq) {
				l2h = 0x40000000 | session->ns;
				session->ns++;
				session->ns &= 0xffffff;
1098 1099 1100
				l2tp_dbg(session, L2TP_MSG_SEQ,
					 "%s: updated ns to %u\n",
					 session->name, session->ns);
1101 1102 1103 1104 1105 1106 1107 1108
			}

			*((__be32 *) bufp) = htonl(l2h);
		}
		bufp += session->l2specific_len;
	}
	if (session->offset)
		bufp += session->offset;
1109

1110
	return bufp - optr;
1111 1112
}

1113
static int l2tp_xmit_core(struct l2tp_session *session, struct sk_buff *skb,
1114
			  struct flowi *fl, size_t data_len)
1115 1116 1117 1118 1119 1120 1121
{
	struct l2tp_tunnel *tunnel = session->tunnel;
	unsigned int len = skb->len;
	int error;

	/* Debug */
	if (session->send_seq)
1122
		l2tp_dbg(session, L2TP_MSG_DATA, "%s: send %zd bytes, ns=%u\n",
1123
			 session->name, data_len, session->ns - 1);
1124
	else
1125
		l2tp_dbg(session, L2TP_MSG_DATA, "%s: send %zd bytes\n",
1126
			 session->name, data_len);
1127 1128

	if (session->debug & L2TP_MSG_DATA) {
1129 1130
		int uhlen = (tunnel->encap == L2TP_ENCAPTYPE_UDP) ? sizeof(struct udphdr) : 0;
		unsigned char *datap = skb->data + uhlen;
1131

1132 1133 1134
		pr_debug("%s: xmit\n", session->name);
		print_hex_dump_bytes("", DUMP_PREFIX_OFFSET,
				     datap, min_t(size_t, 32, len - uhlen));
1135 1136 1137
	}

	/* Queue the packet to IP for output */
1138
	skb->ignore_df = 1;
1139
#if IS_ENABLED(CONFIG_IPV6)
1140
	if (l2tp_sk_is_v6(tunnel->sock))
1141
		error = inet6_csk_xmit(tunnel->sock, skb, NULL);
1142 1143
	else
#endif
1144
		error = ip_queue_xmit(tunnel->sock, skb, fl);
1145 1146 1147

	/* Update stats */
	if (error >= 0) {
1148 1149 1150 1151
		atomic_long_inc(&tunnel->stats.tx_packets);
		atomic_long_add(len, &tunnel->stats.tx_bytes);
		atomic_long_inc(&session->stats.tx_packets);
		atomic_long_add(len, &session->stats.tx_bytes);
1152
	} else {
1153 1154
		atomic_long_inc(&tunnel->stats.tx_errors);
		atomic_long_inc(&session->stats.tx_errors);
1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165
	}

	return 0;
}

/* If caller requires the skb to have a ppp header, the header must be
 * inserted in the skb data before calling this function.
 */
int l2tp_xmit_skb(struct l2tp_session *session, struct sk_buff *skb, int hdr_len)
{
	int data_len = skb->len;
1166 1167
	struct l2tp_tunnel *tunnel = session->tunnel;
	struct sock *sk = tunnel->sock;
1168
	struct flowi *fl;
1169 1170 1171
	struct udphdr *uh;
	struct inet_sock *inet;
	int headroom;
1172 1173
	int uhlen = (tunnel->encap == L2TP_ENCAPTYPE_UDP) ? sizeof(struct udphdr) : 0;
	int udp_len;
1174
	int ret = NET_XMIT_SUCCESS;
1175 1176 1177 1178 1179 1180

	/* Check that there's enough headroom in the skb to insert IP,
	 * UDP and L2TP headers. If not enough, expand it to
	 * make room. Adjust truesize.
	 */
	headroom = NET_SKB_PAD + sizeof(struct iphdr) +
1181
		uhlen + hdr_len;
1182
	if (skb_cow_head(skb, headroom)) {
1183 1184
		kfree_skb(skb);
		return NET_XMIT_DROP;
1185
	}
1186 1187

	/* Setup L2TP header */
1188
	session->build_header(session, __skb_push(skb, hdr_len));
1189

1190
	/* Reset skb netfilter state */
1191 1192 1193 1194 1195
	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
	IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED |
			      IPSKB_REROUTED);
	nf_reset(skb);

1196 1197
	bh_lock_sock(sk);
	if (sock_owned_by_user(sk)) {
1198 1199
		kfree_skb(skb);
		ret = NET_XMIT_DROP;
1200 1201 1202
		goto out_unlock;
	}

1203 1204 1205 1206 1207 1208 1209 1210 1211
	/* The user-space may change the connection status for the user-space
	 * provided socket at run time: we must check it under the socket lock
	 */
	if (tunnel->fd >= 0 && sk->sk_state != TCP_ESTABLISHED) {
		kfree_skb(skb);
		ret = NET_XMIT_DROP;
		goto out_unlock;
	}

1212 1213
	/* Get routing info from the tunnel socket */
	skb_dst_drop(skb);
1214
	skb_dst_set(skb, sk_dst_check(sk, 0));
1215

1216 1217
	inet = inet_sk(sk);
	fl = &inet->cork.fl;
1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229
	switch (tunnel->encap) {
	case L2TP_ENCAPTYPE_UDP:
		/* Setup UDP header */
		__skb_push(skb, sizeof(*uh));
		skb_reset_transport_header(skb);
		uh = udp_hdr(skb);
		uh->source = inet->inet_sport;
		uh->dest = inet->inet_dport;
		udp_len = uhlen + hdr_len + data_len;
		uh->len = htons(udp_len);

		/* Calculate UDP checksum if configured to do so */
1230
#if IS_ENABLED(CONFIG_IPV6)
1231
		if (l2tp_sk_is_v6(sk))
Tom Herbert's avatar
Tom Herbert committed
1232 1233 1234
			udp6_set_csum(udp_get_no_check6_tx(sk),
				      skb, &inet6_sk(sk)->saddr,
				      &sk->sk_v6_daddr, udp_len);
1235 1236
		else
#endif
Tom Herbert's avatar
Tom Herbert committed
1237 1238
		udp_set_csum(sk->sk_no_check_tx, skb, inet->inet_saddr,
			     inet->inet_daddr, udp_len);
1239 1240 1241 1242
		break;

	case L2TP_ENCAPTYPE_IP:
		break;
1243 1244
	}

1245
	l2tp_xmit_core(session, skb, fl, data_len);
1246 1247
out_unlock:
	bh_unlock_sock(sk);
1248

1249
	return ret;
1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260
}
EXPORT_SYMBOL_GPL(l2tp_xmit_skb);

/*****************************************************************************
 * Tinnel and session create/destroy.
 *****************************************************************************/

/* Tunnel socket destruct hook.
 * The tunnel context is deleted only when all session sockets have been
 * closed.
 */
1261
static void l2tp_tunnel_destruct(struct sock *sk)
1262
{
1263
	struct l2tp_tunnel *tunnel = l2tp_tunnel(sk);
1264
	struct l2tp_net *pn;
1265 1266 1267 1268

	if (tunnel == NULL)
		goto end;

1269
	l2tp_info(tunnel, L2TP_MSG_CONTROL, "%s: closing...\n", tunnel->name);
1270 1271


1272
	/* Disable udp encapsulation */
1273 1274 1275 1276 1277
	switch (tunnel->encap) {
	case L2TP_ENCAPTYPE_UDP:
		/* No longer an encapsulation socket. See net/ipv4/udp.c */
		(udp_sk(sk))->encap_type = 0;
		(udp_sk(sk))->encap_rcv = NULL;
1278
		(udp_sk(sk))->encap_destroy = NULL;
1279 1280 1281 1282
		break;
	case L2TP_ENCAPTYPE_IP:
		break;
	}
1283 1284 1285 1286 1287

	/* Remove hooks into tunnel socket */
	sk->sk_destruct = tunnel->old_sk_destruct;
	sk->sk_user_data = NULL;

1288 1289 1290 1291 1292 1293
	/* Remove the tunnel struct from the tunnel list */
	pn = l2tp_pernet(tunnel->l2tp_net);
	spin_lock_bh(&pn->l2tp_tunnel_list_lock);
	list_del_rcu(&tunnel->list);
	spin_unlock_bh(&pn->l2tp_tunnel_list_lock);
	atomic_dec(&l2tp_tunnel_count);
1294

1295
	l2tp_tunnel_closeall(tunnel);
1296 1297

	tunnel->sock = NULL;
1298 1299
	l2tp_tunnel_dec_refcount(tunnel);

1300 1301 1302
	/* Call the original destructor */
	if (sk->sk_destruct)
		(*sk->sk_destruct)(sk);
1303 1304 1305 1306 1307 1308
end:
	return;
}

/* When the tunnel is closed, all the attached sessions need to go too.
 */
1309
void l2tp_tunnel_closeall(struct l2tp_tunnel *tunnel)
1310 1311 1312 1313 1314 1315 1316 1317
{
	int hash;
	struct hlist_node *walk;
	struct hlist_node *tmp;
	struct l2tp_session *session;

	BUG_ON(tunnel == NULL);

1318 1319
	l2tp_info(tunnel, L2TP_MSG_CONTROL, "%s: closing all sessions...\n",
		  tunnel->name);
1320 1321

	write_lock_bh(&tunnel->hlist_lock);
1322
	tunnel->acpt_newsess = false;
1323 1324 1325 1326 1327
	for (hash = 0; hash < L2TP_HASH_SIZE; hash++) {
again:
		hlist_for_each_safe(walk, tmp, &tunnel->session_hlist[hash]) {
			session = hlist_entry(walk, struct l2tp_session, hlist);

1328 1329
			l2tp_info(session, L2TP_MSG_CONTROL,
				  "%s: closing session\n", session->name);
1330 1331 1332

			hlist_del_init(&session->hlist);

1333 1334 1335
			if (test_and_set_bit(0, &session->dead))
				goto again;

1336 1337 1338 1339 1340
			if (session->ref != NULL)
				(*session->ref)(session);

			write_unlock_bh(&tunnel->hlist_lock);

1341
			__l2tp_session_unhash(session);
1342 1343
			l2tp_session_queue_purge(session);

1344 1345 1346 1347 1348 1349
			if (session->session_close != NULL)
				(*session->session_close)(session);

			if (session->deref != NULL)
				(*session->deref)(session);

1350 1351
			l2tp_session_dec_refcount(session);

1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363
			write_lock_bh(&tunnel->hlist_lock);

			/* Now restart from the beginning of this hash
			 * chain.  We always remove a session from the
			 * list so we are guaranteed to make forward
			 * progress.
			 */
			goto again;
		}
	}
	write_unlock_bh(&tunnel->hlist_lock);
}
1364
EXPORT_SYMBOL_GPL(l2tp_tunnel_closeall);
1365

1366 1367 1368 1369 1370 1371 1372 1373 1374 1375
/* Tunnel socket destroy hook for UDP encapsulation */
static void l2tp_udp_encap_destroy(struct sock *sk)
{
	struct l2tp_tunnel *tunnel = l2tp_sock_to_tunnel(sk);
	if (tunnel) {
		l2tp_tunnel_closeall(tunnel);
		sock_put(sk);
	}
}

1376 1377 1378 1379 1380 1381 1382 1383
/* Workqueue tunnel deletion function */
static void l2tp_tunnel_del_work(struct work_struct *work)
{
	struct l2tp_tunnel *tunnel = NULL;
	struct socket *sock = NULL;
	struct sock *sk = NULL;

	tunnel = container_of(work, struct l2tp_tunnel, del_work);
1384 1385 1386

	l2tp_tunnel_closeall(tunnel);

1387 1388
	sk = l2tp_tunnel_sock_lookup(tunnel);
	if (!sk)
1389
		goto out;
1390 1391 1392

	sock = sk->sk_socket;

1393 1394 1395 1396
	/* If the tunnel socket was created by userspace, then go through the
	 * inet layer to shut the socket down, and let userspace close it.
	 * Otherwise, if we created the socket directly within the kernel, use
	 * the sk API to release it here.
1397 1398
	 * In either case the tunnel resources are freed in the socket
	 * destructor when the tunnel socket goes away.
1399
	 */
1400 1401 1402
	if (tunnel->fd >= 0) {
		if (sock)
			inet_shutdown(sock, 2);
1403
	} else {
1404
		if (sock) {
1405
			kernel_sock_shutdown(sock, SHUT_RDWR);
1406 1407
			sock_release(sock);
		}
1408
	}
1409 1410

	l2tp_tunnel_sock_put(sk);
1411 1412
out:
	l2tp_tunnel_dec_refcount(tunnel);
1413 1414
}

1415 1416 1417
/* Create a socket for the tunnel, if one isn't set up by
 * userspace. This is used for static tunnels where there is no
 * managing L2TP daemon.
1418 1419 1420 1421 1422
 *
 * Since we don't want these sockets to keep a namespace alive by
 * themselves, we drop the socket's namespace refcount after creation.
 * These sockets are freed when the namespace exits using the pernet
 * exit hook.
1423
 */
1424 1425 1426 1427 1428
static int l2tp_tunnel_sock_create(struct net *net,
				u32 tunnel_id,
				u32 peer_tunnel_id,
				struct l2tp_tunnel_cfg *cfg,
				struct socket **sockp)
1429 1430
{
	int err = -EINVAL;
1431
	struct socket *sock = NULL;
Tom Herbert's avatar
Tom Herbert committed
1432
	struct udp_port_cfg udp_conf;
1433 1434 1435

	switch (cfg->encap) {
	case L2TP_ENCAPTYPE_UDP:
Tom Herbert's avatar
Tom Herbert committed
1436 1437
		memset(&udp_conf, 0, sizeof(udp_conf));

1438 1439
#if IS_ENABLED(CONFIG_IPV6)
		if (cfg->local_ip6 && cfg->peer_ip6) {
Tom Herbert's avatar
Tom Herbert committed
1440 1441 1442 1443 1444 1445
			udp_conf.family = AF_INET6;
			memcpy(&udp_conf.local_ip6, cfg->local_ip6,
			       sizeof(udp_conf.local_ip6));
			memcpy(&udp_conf.peer_ip6, cfg->peer_ip6,
			       sizeof(udp_conf.peer_ip6));
			udp_conf.use_udp6_tx_checksums =
1446
			  ! cfg->udp6_zero_tx_checksums;
Tom Herbert's avatar
Tom Herbert committed
1447
			udp_conf.use_udp6_rx_checksums =
1448
			  ! cfg->udp6_zero_rx_checksums;
1449 1450 1451
		} else
#endif
		{
Tom Herbert's avatar
Tom Herbert committed
1452 1453 1454 1455
			udp_conf.family = AF_INET;
			udp_conf.local_ip = cfg->local_ip;
			udp_conf.peer_ip = cfg->peer_ip;
			udp_conf.use_udp_checksums = cfg->use_udp_checksums;
1456
		}
1457

Tom Herbert's avatar
Tom Herbert committed
1458 1459 1460 1461 1462 1463
		udp_conf.local_udp_port = htons(cfg->local_udp_port);
		udp_conf.peer_udp_port = htons(cfg->peer_udp_port);

		err = udp_sock_create(net, &udp_conf, &sock);
		if (err < 0)
			goto out;
1464 1465 1466 1467

		break;

	case L2TP_ENCAPTYPE_IP:
1468 1469
#if IS_ENABLED(CONFIG_IPV6)
		if (cfg->local_ip6 && cfg->peer_ip6) {
Tom Herbert's avatar
Tom Herbert committed
1470 1471
			struct sockaddr_l2tpip6 ip6_addr = {0};

1472
			err = sock_create_kern(net, AF_INET6, SOCK_DGRAM,
1473
					  IPPROTO_L2TP, &sock);
1474 1475
			if (err < 0)
				goto out;