af_netlink.c 41.0 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10
/*
 * NETLINK      Kernel-user communication protocol.
 *
 * 		Authors:	Alan Cox <alan@redhat.com>
 * 				Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
 *
 *		This program is free software; you can redistribute it and/or
 *		modify it under the terms of the GNU General Public License
 *		as published by the Free Software Foundation; either version
 *		2 of the License, or (at your option) any later version.
11
 *
L
Linus Torvalds 已提交
12 13 14 15
 * Tue Jun 26 14:36:48 MEST 2001 Herbert "herp" Rosmanith
 *                               added netlink_proto_exit
 * Tue Jan 22 18:32:44 BRST 2002 Arnaldo C. de Melo <acme@conectiva.com.br>
 * 				 use nlk_sk, as sk->protinfo is on a diet 8)
16 17 18 19 20 21
 * Fri Jul 22 19:51:12 MEST 2005 Harald Welte <laforge@gnumonks.org>
 * 				 - inc module use count of module that owns
 * 				   the kernel socket in case userspace opens
 * 				   socket of same protocol
 * 				 - remove all module support, since netlink is
 * 				   mandatory if CONFIG_NET=y these days
L
Linus Torvalds 已提交
22 23 24 25
 */

#include <linux/module.h>

26
#include <linux/capability.h>
L
Linus Torvalds 已提交
27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56
#include <linux/kernel.h>
#include <linux/init.h>
#include <linux/signal.h>
#include <linux/sched.h>
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/stat.h>
#include <linux/socket.h>
#include <linux/un.h>
#include <linux/fcntl.h>
#include <linux/termios.h>
#include <linux/sockios.h>
#include <linux/net.h>
#include <linux/fs.h>
#include <linux/slab.h>
#include <asm/uaccess.h>
#include <linux/skbuff.h>
#include <linux/netdevice.h>
#include <linux/rtnetlink.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <linux/smp_lock.h>
#include <linux/notifier.h>
#include <linux/security.h>
#include <linux/jhash.h>
#include <linux/jiffies.h>
#include <linux/random.h>
#include <linux/bitops.h>
#include <linux/mm.h>
#include <linux/types.h>
A
Andrew Morton 已提交
57
#include <linux/audit.h>
58
#include <linux/selinux.h>
59
#include <linux/mutex.h>
A
Andrew Morton 已提交
60

L
Linus Torvalds 已提交
61 62
#include <net/sock.h>
#include <net/scm.h>
63
#include <net/netlink.h>
L
Linus Torvalds 已提交
64

65
#define NLGRPSZ(x)	(ALIGN(x, sizeof(unsigned long) * 8) / 8)
L
Linus Torvalds 已提交
66 67 68 69 70 71

struct netlink_sock {
	/* struct sock has to be the first member of netlink_sock */
	struct sock		sk;
	u32			pid;
	u32			dst_pid;
72
	u32			dst_group;
73 74 75 76
	u32			flags;
	u32			subscriptions;
	u32			ngroups;
	unsigned long		*groups;
L
Linus Torvalds 已提交
77 78 79
	unsigned long		state;
	wait_queue_head_t	wait;
	struct netlink_callback	*cb;
80 81
	struct mutex		*cb_mutex;
	struct mutex		cb_def_mutex;
L
Linus Torvalds 已提交
82
	void			(*data_ready)(struct sock *sk, int bytes);
83
	struct module		*module;
L
Linus Torvalds 已提交
84 85
};

86
#define NETLINK_KERNEL_SOCKET	0x1
87
#define NETLINK_RECV_PKTINFO	0x2
88

L
Linus Torvalds 已提交
89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109
static inline struct netlink_sock *nlk_sk(struct sock *sk)
{
	return (struct netlink_sock *)sk;
}

struct nl_pid_hash {
	struct hlist_head *table;
	unsigned long rehash_time;

	unsigned int mask;
	unsigned int shift;

	unsigned int entries;
	unsigned int max_shift;

	u32 rnd;
};

struct netlink_table {
	struct nl_pid_hash hash;
	struct hlist_head mc_list;
110
	unsigned long *listeners;
L
Linus Torvalds 已提交
111
	unsigned int nl_nonroot;
112
	unsigned int groups;
113
	struct mutex *cb_mutex;
114
	struct module *module;
115
	int registered;
L
Linus Torvalds 已提交
116 117 118 119 120 121 122 123 124 125 126 127
};

static struct netlink_table *nl_table;

static DECLARE_WAIT_QUEUE_HEAD(nl_table_wait);

static int netlink_dump(struct sock *sk);
static void netlink_destroy_callback(struct netlink_callback *cb);

static DEFINE_RWLOCK(nl_table_lock);
static atomic_t nl_table_users = ATOMIC_INIT(0);

128
static ATOMIC_NOTIFIER_HEAD(netlink_chain);
L
Linus Torvalds 已提交
129

130 131 132 133 134
static u32 netlink_group_mask(u32 group)
{
	return group ? 1 << (group - 1) : 0;
}

L
Linus Torvalds 已提交
135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150
static struct hlist_head *nl_pid_hashfn(struct nl_pid_hash *hash, u32 pid)
{
	return &hash->table[jhash_1word(pid, hash->rnd) & hash->mask];
}

static void netlink_sock_destruct(struct sock *sk)
{
	skb_queue_purge(&sk->sk_receive_queue);

	if (!sock_flag(sk, SOCK_DEAD)) {
		printk("Freeing alive netlink socket %p\n", sk);
		return;
	}
	BUG_TRAP(!atomic_read(&sk->sk_rmem_alloc));
	BUG_TRAP(!atomic_read(&sk->sk_wmem_alloc));
	BUG_TRAP(!nlk_sk(sk)->cb);
151
	BUG_TRAP(!nlk_sk(sk)->groups);
L
Linus Torvalds 已提交
152 153 154 155 156 157 158 159 160 161
}

/* This lock without WQ_FLAG_EXCLUSIVE is good on UP and it is _very_ bad on SMP.
 * Look, when several writers sleep and reader wakes them up, all but one
 * immediately hit write lock and grab all the cpus. Exclusive sleep solves
 * this, _but_ remember, it adds useless work on UP machines.
 */

static void netlink_table_grab(void)
{
162
	write_lock_irq(&nl_table_lock);
L
Linus Torvalds 已提交
163 164 165 166 167 168 169 170 171

	if (atomic_read(&nl_table_users)) {
		DECLARE_WAITQUEUE(wait, current);

		add_wait_queue_exclusive(&nl_table_wait, &wait);
		for(;;) {
			set_current_state(TASK_UNINTERRUPTIBLE);
			if (atomic_read(&nl_table_users) == 0)
				break;
172
			write_unlock_irq(&nl_table_lock);
L
Linus Torvalds 已提交
173
			schedule();
174
			write_lock_irq(&nl_table_lock);
L
Linus Torvalds 已提交
175 176 177 178 179 180 181 182 183
		}

		__set_current_state(TASK_RUNNING);
		remove_wait_queue(&nl_table_wait, &wait);
	}
}

static __inline__ void netlink_table_ungrab(void)
{
184
	write_unlock_irq(&nl_table_lock);
L
Linus Torvalds 已提交
185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299
	wake_up(&nl_table_wait);
}

static __inline__ void
netlink_lock_table(void)
{
	/* read_lock() synchronizes us to netlink_table_grab */

	read_lock(&nl_table_lock);
	atomic_inc(&nl_table_users);
	read_unlock(&nl_table_lock);
}

static __inline__ void
netlink_unlock_table(void)
{
	if (atomic_dec_and_test(&nl_table_users))
		wake_up(&nl_table_wait);
}

static __inline__ struct sock *netlink_lookup(int protocol, u32 pid)
{
	struct nl_pid_hash *hash = &nl_table[protocol].hash;
	struct hlist_head *head;
	struct sock *sk;
	struct hlist_node *node;

	read_lock(&nl_table_lock);
	head = nl_pid_hashfn(hash, pid);
	sk_for_each(sk, node, head) {
		if (nlk_sk(sk)->pid == pid) {
			sock_hold(sk);
			goto found;
		}
	}
	sk = NULL;
found:
	read_unlock(&nl_table_lock);
	return sk;
}

static inline struct hlist_head *nl_pid_hash_alloc(size_t size)
{
	if (size <= PAGE_SIZE)
		return kmalloc(size, GFP_ATOMIC);
	else
		return (struct hlist_head *)
			__get_free_pages(GFP_ATOMIC, get_order(size));
}

static inline void nl_pid_hash_free(struct hlist_head *table, size_t size)
{
	if (size <= PAGE_SIZE)
		kfree(table);
	else
		free_pages((unsigned long)table, get_order(size));
}

static int nl_pid_hash_rehash(struct nl_pid_hash *hash, int grow)
{
	unsigned int omask, mask, shift;
	size_t osize, size;
	struct hlist_head *otable, *table;
	int i;

	omask = mask = hash->mask;
	osize = size = (mask + 1) * sizeof(*table);
	shift = hash->shift;

	if (grow) {
		if (++shift > hash->max_shift)
			return 0;
		mask = mask * 2 + 1;
		size *= 2;
	}

	table = nl_pid_hash_alloc(size);
	if (!table)
		return 0;

	memset(table, 0, size);
	otable = hash->table;
	hash->table = table;
	hash->mask = mask;
	hash->shift = shift;
	get_random_bytes(&hash->rnd, sizeof(hash->rnd));

	for (i = 0; i <= omask; i++) {
		struct sock *sk;
		struct hlist_node *node, *tmp;

		sk_for_each_safe(sk, node, tmp, &otable[i])
			__sk_add_node(sk, nl_pid_hashfn(hash, nlk_sk(sk)->pid));
	}

	nl_pid_hash_free(otable, osize);
	hash->rehash_time = jiffies + 10 * 60 * HZ;
	return 1;
}

static inline int nl_pid_hash_dilute(struct nl_pid_hash *hash, int len)
{
	int avg = hash->entries >> hash->shift;

	if (unlikely(avg > 1) && nl_pid_hash_rehash(hash, 1))
		return 1;

	if (unlikely(len > avg) && time_after(jiffies, hash->rehash_time)) {
		nl_pid_hash_rehash(hash, 0);
		return 1;
	}

	return 0;
}

300
static const struct proto_ops netlink_ops;
L
Linus Torvalds 已提交
301

302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319
static void
netlink_update_listeners(struct sock *sk)
{
	struct netlink_table *tbl = &nl_table[sk->sk_protocol];
	struct hlist_node *node;
	unsigned long mask;
	unsigned int i;

	for (i = 0; i < NLGRPSZ(tbl->groups)/sizeof(unsigned long); i++) {
		mask = 0;
		sk_for_each_bound(sk, node, &tbl->mc_list)
			mask |= nlk_sk(sk)->groups[i];
		tbl->listeners[i] = mask;
	}
	/* this function is only called with the netlink table "grabbed", which
	 * makes sure updates are visible before bind or setsockopt return. */
}

L
Linus Torvalds 已提交
320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362
static int netlink_insert(struct sock *sk, u32 pid)
{
	struct nl_pid_hash *hash = &nl_table[sk->sk_protocol].hash;
	struct hlist_head *head;
	int err = -EADDRINUSE;
	struct sock *osk;
	struct hlist_node *node;
	int len;

	netlink_table_grab();
	head = nl_pid_hashfn(hash, pid);
	len = 0;
	sk_for_each(osk, node, head) {
		if (nlk_sk(osk)->pid == pid)
			break;
		len++;
	}
	if (node)
		goto err;

	err = -EBUSY;
	if (nlk_sk(sk)->pid)
		goto err;

	err = -ENOMEM;
	if (BITS_PER_LONG > 32 && unlikely(hash->entries >= UINT_MAX))
		goto err;

	if (len && nl_pid_hash_dilute(hash, len))
		head = nl_pid_hashfn(hash, pid);
	hash->entries++;
	nlk_sk(sk)->pid = pid;
	sk_add_node(sk, head);
	err = 0;

err:
	netlink_table_ungrab();
	return err;
}

static void netlink_remove(struct sock *sk)
{
	netlink_table_grab();
363 364
	if (sk_del_node_init(sk))
		nl_table[sk->sk_protocol].hash.entries--;
365
	if (nlk_sk(sk)->subscriptions)
L
Linus Torvalds 已提交
366 367 368 369 370 371 372 373 374 375
		__sk_del_bind_node(sk);
	netlink_table_ungrab();
}

static struct proto netlink_proto = {
	.name	  = "NETLINK",
	.owner	  = THIS_MODULE,
	.obj_size = sizeof(struct netlink_sock),
};

376 377
static int __netlink_create(struct socket *sock, struct mutex *cb_mutex,
			    int protocol)
L
Linus Torvalds 已提交
378 379 380
{
	struct sock *sk;
	struct netlink_sock *nlk;
381 382 383 384 385 386 387 388 389 390

	sock->ops = &netlink_ops;

	sk = sk_alloc(PF_NETLINK, GFP_KERNEL, &netlink_proto, 1);
	if (!sk)
		return -ENOMEM;

	sock_init_data(sock, sk);

	nlk = nlk_sk(sk);
391 392 393 394 395 396
	if (cb_mutex)
		nlk->cb_mutex = cb_mutex;
	else {
		nlk->cb_mutex = &nlk->cb_def_mutex;
		mutex_init(nlk->cb_mutex);
	}
397 398 399 400 401 402 403 404 405 406
	init_waitqueue_head(&nlk->wait);

	sk->sk_destruct = netlink_sock_destruct;
	sk->sk_protocol = protocol;
	return 0;
}

static int netlink_create(struct socket *sock, int protocol)
{
	struct module *module = NULL;
407
	struct mutex *cb_mutex;
408
	struct netlink_sock *nlk;
409
	int err = 0;
L
Linus Torvalds 已提交
410 411 412 413 414 415 416 417 418

	sock->state = SS_UNCONNECTED;

	if (sock->type != SOCK_RAW && sock->type != SOCK_DGRAM)
		return -ESOCKTNOSUPPORT;

	if (protocol<0 || protocol >= MAX_LINKS)
		return -EPROTONOSUPPORT;

419
	netlink_lock_table();
420
#ifdef CONFIG_KMOD
421
	if (!nl_table[protocol].registered) {
422
		netlink_unlock_table();
423
		request_module("net-pf-%d-proto-%d", PF_NETLINK, protocol);
424
		netlink_lock_table();
425
	}
426 427 428 429
#endif
	if (nl_table[protocol].registered &&
	    try_module_get(nl_table[protocol].module))
		module = nl_table[protocol].module;
430
	cb_mutex = nl_table[protocol].cb_mutex;
431
	netlink_unlock_table();
432

433
	if ((err = __netlink_create(sock, cb_mutex, protocol)) < 0)
434 435 436 437
		goto out_module;

	nlk = nlk_sk(sock->sk);
	nlk->module = module;
438 439
out:
	return err;
L
Linus Torvalds 已提交
440

441 442 443
out_module:
	module_put(module);
	goto out;
L
Linus Torvalds 已提交
444 445 446 447 448 449 450 451 452 453 454
}

static int netlink_release(struct socket *sock)
{
	struct sock *sk = sock->sk;
	struct netlink_sock *nlk;

	if (!sk)
		return 0;

	netlink_remove(sk);
455
	sock_orphan(sk);
L
Linus Torvalds 已提交
456 457
	nlk = nlk_sk(sk);

458
	mutex_lock(nlk->cb_mutex);
L
Linus Torvalds 已提交
459
	if (nlk->cb) {
460 461
		if (nlk->cb->done)
			nlk->cb->done(nlk->cb);
L
Linus Torvalds 已提交
462 463 464
		netlink_destroy_callback(nlk->cb);
		nlk->cb = NULL;
	}
465
	mutex_unlock(nlk->cb_mutex);
L
Linus Torvalds 已提交
466 467 468 469 470 471 472 473 474

	/* OK. Socket is unlinked, and, therefore,
	   no new packets will arrive */

	sock->sk = NULL;
	wake_up_interruptible_all(&nlk->wait);

	skb_queue_purge(&sk->sk_write_queue);

475
	if (nlk->pid && !nlk->subscriptions) {
L
Linus Torvalds 已提交
476 477 478 479
		struct netlink_notify n = {
						.protocol = sk->sk_protocol,
						.pid = nlk->pid,
					  };
480 481
		atomic_notifier_call_chain(&netlink_chain,
				NETLINK_URELEASE, &n);
482
	}
483

484
	module_put(nlk->module);
485

486
	netlink_table_grab();
487
	if (nlk->flags & NETLINK_KERNEL_SOCKET) {
488
		kfree(nl_table[sk->sk_protocol].listeners);
489
		nl_table[sk->sk_protocol].module = NULL;
490
		nl_table[sk->sk_protocol].registered = 0;
491 492 493
	} else if (nlk->subscriptions)
		netlink_update_listeners(sk);
	netlink_table_ungrab();
494

495 496 497
	kfree(nlk->groups);
	nlk->groups = NULL;

L
Linus Torvalds 已提交
498 499 500 501 502 503 504 505 506 507 508
	sock_put(sk);
	return 0;
}

static int netlink_autobind(struct socket *sock)
{
	struct sock *sk = sock->sk;
	struct nl_pid_hash *hash = &nl_table[sk->sk_protocol].hash;
	struct hlist_head *head;
	struct sock *osk;
	struct hlist_node *node;
509
	s32 pid = current->tgid;
L
Linus Torvalds 已提交
510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531
	int err;
	static s32 rover = -4097;

retry:
	cond_resched();
	netlink_table_grab();
	head = nl_pid_hashfn(hash, pid);
	sk_for_each(osk, node, head) {
		if (nlk_sk(osk)->pid == pid) {
			/* Bind collision, search negative pid values. */
			pid = rover--;
			if (rover > -4097)
				rover = -4097;
			netlink_table_ungrab();
			goto retry;
		}
	}
	netlink_table_ungrab();

	err = netlink_insert(sk, pid);
	if (err == -EADDRINUSE)
		goto retry;
532 533 534 535 536 537

	/* If 2 threads race to autobind, that is fine.  */
	if (err == -EBUSY)
		err = 0;

	return err;
L
Linus Torvalds 已提交
538 539
}

540 541
static inline int netlink_capable(struct socket *sock, unsigned int flag)
{
L
Linus Torvalds 已提交
542 543
	return (nl_table[sock->sk->sk_protocol].nl_nonroot & flag) ||
	       capable(CAP_NET_ADMIN);
544
}
L
Linus Torvalds 已提交
545

546 547 548 549 550 551 552 553 554 555 556 557
static void
netlink_update_subscriptions(struct sock *sk, unsigned int subscriptions)
{
	struct netlink_sock *nlk = nlk_sk(sk);

	if (nlk->subscriptions && !subscriptions)
		__sk_del_bind_node(sk);
	else if (!nlk->subscriptions && subscriptions)
		sk_add_bind_node(sk, &nl_table[sk->sk_protocol].mc_list);
	nlk->subscriptions = subscriptions;
}

558 559 560 561 562 563 564 565 566 567 568 569 570 571 572
static int netlink_alloc_groups(struct sock *sk)
{
	struct netlink_sock *nlk = nlk_sk(sk);
	unsigned int groups;
	int err = 0;

	netlink_lock_table();
	groups = nl_table[sk->sk_protocol].groups;
	if (!nl_table[sk->sk_protocol].registered)
		err = -ENOENT;
	netlink_unlock_table();

	if (err)
		return err;

573
	nlk->groups = kzalloc(NLGRPSZ(groups), GFP_KERNEL);
574 575 576 577 578 579
	if (nlk->groups == NULL)
		return -ENOMEM;
	nlk->ngroups = groups;
	return 0;
}

L
Linus Torvalds 已提交
580 581 582 583 584 585
static int netlink_bind(struct socket *sock, struct sockaddr *addr, int addr_len)
{
	struct sock *sk = sock->sk;
	struct netlink_sock *nlk = nlk_sk(sk);
	struct sockaddr_nl *nladdr = (struct sockaddr_nl *)addr;
	int err;
586

L
Linus Torvalds 已提交
587 588 589 590
	if (nladdr->nl_family != AF_NETLINK)
		return -EINVAL;

	/* Only superuser is allowed to listen multicasts */
591 592 593 594 595 596 597 598 599
	if (nladdr->nl_groups) {
		if (!netlink_capable(sock, NL_NONROOT_RECV))
			return -EPERM;
		if (nlk->groups == NULL) {
			err = netlink_alloc_groups(sk);
			if (err)
				return err;
		}
	}
L
Linus Torvalds 已提交
600 601 602 603 604 605 606 607 608 609 610 611

	if (nlk->pid) {
		if (nladdr->nl_pid != nlk->pid)
			return -EINVAL;
	} else {
		err = nladdr->nl_pid ?
			netlink_insert(sk, nladdr->nl_pid) :
			netlink_autobind(sock);
		if (err)
			return err;
	}

612
	if (!nladdr->nl_groups && (nlk->groups == NULL || !(u32)nlk->groups[0]))
L
Linus Torvalds 已提交
613 614 615
		return 0;

	netlink_table_grab();
616
	netlink_update_subscriptions(sk, nlk->subscriptions +
617 618 619
					 hweight32(nladdr->nl_groups) -
					 hweight32(nlk->groups[0]));
	nlk->groups[0] = (nlk->groups[0] & ~0xffffffffUL) | nladdr->nl_groups;
620
	netlink_update_listeners(sk);
L
Linus Torvalds 已提交
621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636
	netlink_table_ungrab();

	return 0;
}

static int netlink_connect(struct socket *sock, struct sockaddr *addr,
			   int alen, int flags)
{
	int err = 0;
	struct sock *sk = sock->sk;
	struct netlink_sock *nlk = nlk_sk(sk);
	struct sockaddr_nl *nladdr=(struct sockaddr_nl*)addr;

	if (addr->sa_family == AF_UNSPEC) {
		sk->sk_state	= NETLINK_UNCONNECTED;
		nlk->dst_pid	= 0;
637
		nlk->dst_group  = 0;
L
Linus Torvalds 已提交
638 639 640 641 642 643 644 645 646 647 648 649 650 651 652
		return 0;
	}
	if (addr->sa_family != AF_NETLINK)
		return -EINVAL;

	/* Only superuser is allowed to send multicasts */
	if (nladdr->nl_groups && !netlink_capable(sock, NL_NONROOT_SEND))
		return -EPERM;

	if (!nlk->pid)
		err = netlink_autobind(sock);

	if (err == 0) {
		sk->sk_state	= NETLINK_CONNECTED;
		nlk->dst_pid 	= nladdr->nl_pid;
653
		nlk->dst_group  = ffs(nladdr->nl_groups);
L
Linus Torvalds 已提交
654 655 656 657 658 659 660 661 662 663
	}

	return err;
}

static int netlink_getname(struct socket *sock, struct sockaddr *addr, int *addr_len, int peer)
{
	struct sock *sk = sock->sk;
	struct netlink_sock *nlk = nlk_sk(sk);
	struct sockaddr_nl *nladdr=(struct sockaddr_nl *)addr;
664

L
Linus Torvalds 已提交
665 666 667 668 669 670
	nladdr->nl_family = AF_NETLINK;
	nladdr->nl_pad = 0;
	*addr_len = sizeof(*nladdr);

	if (peer) {
		nladdr->nl_pid = nlk->dst_pid;
671
		nladdr->nl_groups = netlink_group_mask(nlk->dst_group);
L
Linus Torvalds 已提交
672 673
	} else {
		nladdr->nl_pid = nlk->pid;
674
		nladdr->nl_groups = nlk->groups ? nlk->groups[0] : 0;
L
Linus Torvalds 已提交
675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709
	}
	return 0;
}

static void netlink_overrun(struct sock *sk)
{
	if (!test_and_set_bit(0, &nlk_sk(sk)->state)) {
		sk->sk_err = ENOBUFS;
		sk->sk_error_report(sk);
	}
}

static struct sock *netlink_getsockbypid(struct sock *ssk, u32 pid)
{
	int protocol = ssk->sk_protocol;
	struct sock *sock;
	struct netlink_sock *nlk;

	sock = netlink_lookup(protocol, pid);
	if (!sock)
		return ERR_PTR(-ECONNREFUSED);

	/* Don't bother queuing skb if kernel socket has no input function */
	nlk = nlk_sk(sock);
	if ((nlk->pid == 0 && !nlk->data_ready) ||
	    (sock->sk_state == NETLINK_CONNECTED &&
	     nlk->dst_pid != nlk_sk(ssk)->pid)) {
		sock_put(sock);
		return ERR_PTR(-ECONNREFUSED);
	}
	return sock;
}

struct sock *netlink_getsockbyfilp(struct file *filp)
{
J
Josef Sipek 已提交
710
	struct inode *inode = filp->f_path.dentry->d_inode;
L
Linus Torvalds 已提交
711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733
	struct sock *sock;

	if (!S_ISSOCK(inode->i_mode))
		return ERR_PTR(-ENOTSOCK);

	sock = SOCKET_I(inode)->sk;
	if (sock->sk_family != AF_NETLINK)
		return ERR_PTR(-EINVAL);

	sock_hold(sock);
	return sock;
}

/*
 * Attach a skb to a netlink socket.
 * The caller must hold a reference to the destination socket. On error, the
 * reference is dropped. The skb is not send to the destination, just all
 * all error checks are performed and memory in the queue is reserved.
 * Return values:
 * < 0: error. skb freed, reference to sock dropped.
 * 0: continue
 * 1: repeat lookup - reference dropped while waiting for socket memory.
 */
A
Alexey Kuznetsov 已提交
734 735
int netlink_attachskb(struct sock *sk, struct sk_buff *skb, int nonblock,
		long timeo, struct sock *ssk)
L
Linus Torvalds 已提交
736 737 738 739 740 741 742 743 744
{
	struct netlink_sock *nlk;

	nlk = nlk_sk(sk);

	if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf ||
	    test_bit(0, &nlk->state)) {
		DECLARE_WAITQUEUE(wait, current);
		if (!timeo) {
A
Alexey Kuznetsov 已提交
745
			if (!ssk || nlk_sk(ssk)->pid == 0)
L
Linus Torvalds 已提交
746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789
				netlink_overrun(sk);
			sock_put(sk);
			kfree_skb(skb);
			return -EAGAIN;
		}

		__set_current_state(TASK_INTERRUPTIBLE);
		add_wait_queue(&nlk->wait, &wait);

		if ((atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf ||
		     test_bit(0, &nlk->state)) &&
		    !sock_flag(sk, SOCK_DEAD))
			timeo = schedule_timeout(timeo);

		__set_current_state(TASK_RUNNING);
		remove_wait_queue(&nlk->wait, &wait);
		sock_put(sk);

		if (signal_pending(current)) {
			kfree_skb(skb);
			return sock_intr_errno(timeo);
		}
		return 1;
	}
	skb_set_owner_r(skb, sk);
	return 0;
}

int netlink_sendskb(struct sock *sk, struct sk_buff *skb, int protocol)
{
	int len = skb->len;

	skb_queue_tail(&sk->sk_receive_queue, skb);
	sk->sk_data_ready(sk, len);
	sock_put(sk);
	return len;
}

void netlink_detachskb(struct sock *sk, struct sk_buff *skb)
{
	kfree_skb(skb);
	sock_put(sk);
}

790
static inline struct sk_buff *netlink_trim(struct sk_buff *skb,
A
Al Viro 已提交
791
					   gfp_t allocation)
L
Linus Torvalds 已提交
792 793 794 795 796
{
	int delta;

	skb_orphan(skb);

797
	delta = skb->end - skb->tail;
L
Linus Torvalds 已提交
798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829
	if (delta * 2 < skb->truesize)
		return skb;

	if (skb_shared(skb)) {
		struct sk_buff *nskb = skb_clone(skb, allocation);
		if (!nskb)
			return skb;
		kfree_skb(skb);
		skb = nskb;
	}

	if (!pskb_expand_head(skb, 0, -delta, allocation))
		skb->truesize -= delta;

	return skb;
}

int netlink_unicast(struct sock *ssk, struct sk_buff *skb, u32 pid, int nonblock)
{
	struct sock *sk;
	int err;
	long timeo;

	skb = netlink_trim(skb, gfp_any());

	timeo = sock_sndtimeo(ssk, nonblock);
retry:
	sk = netlink_getsockbypid(ssk, pid);
	if (IS_ERR(sk)) {
		kfree_skb(skb);
		return PTR_ERR(sk);
	}
A
Alexey Kuznetsov 已提交
830
	err = netlink_attachskb(sk, skb, nonblock, timeo, ssk);
L
Linus Torvalds 已提交
831 832 833 834 835 836 837 838
	if (err == 1)
		goto retry;
	if (err)
		return err;

	return netlink_sendskb(sk, skb, ssk->sk_protocol);
}

839 840 841 842 843 844 845 846 847 848 849
int netlink_has_listeners(struct sock *sk, unsigned int group)
{
	int res = 0;

	BUG_ON(!(nlk_sk(sk)->flags & NETLINK_KERNEL_SOCKET));
	if (group - 1 < nl_table[sk->sk_protocol].groups)
		res = test_bit(group - 1, nl_table[sk->sk_protocol].listeners);
	return res;
}
EXPORT_SYMBOL_GPL(netlink_has_listeners);

L
Linus Torvalds 已提交
850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870
static __inline__ int netlink_broadcast_deliver(struct sock *sk, struct sk_buff *skb)
{
	struct netlink_sock *nlk = nlk_sk(sk);

	if (atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf &&
	    !test_bit(0, &nlk->state)) {
		skb_set_owner_r(skb, sk);
		skb_queue_tail(&sk->sk_receive_queue, skb);
		sk->sk_data_ready(sk, skb->len);
		return atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf;
	}
	return -1;
}

struct netlink_broadcast_data {
	struct sock *exclude_sk;
	u32 pid;
	u32 group;
	int failure;
	int congested;
	int delivered;
A
Al Viro 已提交
871
	gfp_t allocation;
L
Linus Torvalds 已提交
872 873 874 875 876 877 878 879 880 881 882 883
	struct sk_buff *skb, *skb2;
};

static inline int do_one_broadcast(struct sock *sk,
				   struct netlink_broadcast_data *p)
{
	struct netlink_sock *nlk = nlk_sk(sk);
	int val;

	if (p->exclude_sk == sk)
		goto out;

884 885
	if (nlk->pid == p->pid || p->group - 1 >= nlk->ngroups ||
	    !test_bit(p->group - 1, nlk->groups))
L
Linus Torvalds 已提交
886 887 888 889 890 891 892 893 894
		goto out;

	if (p->failure) {
		netlink_overrun(sk);
		goto out;
	}

	sock_hold(sk);
	if (p->skb2 == NULL) {
895
		if (skb_shared(p->skb)) {
L
Linus Torvalds 已提交
896 897
			p->skb2 = skb_clone(p->skb, p->allocation);
		} else {
898 899 900 901 902 903
			p->skb2 = skb_get(p->skb);
			/*
			 * skb ownership may have been set when
			 * delivered to a previous socket.
			 */
			skb_orphan(p->skb2);
L
Linus Torvalds 已提交
904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923
		}
	}
	if (p->skb2 == NULL) {
		netlink_overrun(sk);
		/* Clone failed. Notify ALL listeners. */
		p->failure = 1;
	} else if ((val = netlink_broadcast_deliver(sk, p->skb2)) < 0) {
		netlink_overrun(sk);
	} else {
		p->congested |= val;
		p->delivered = 1;
		p->skb2 = NULL;
	}
	sock_put(sk);

out:
	return 0;
}

int netlink_broadcast(struct sock *ssk, struct sk_buff *skb, u32 pid,
A
Al Viro 已提交
924
		      u32 group, gfp_t allocation)
L
Linus Torvalds 已提交
925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948
{
	struct netlink_broadcast_data info;
	struct hlist_node *node;
	struct sock *sk;

	skb = netlink_trim(skb, allocation);

	info.exclude_sk = ssk;
	info.pid = pid;
	info.group = group;
	info.failure = 0;
	info.congested = 0;
	info.delivered = 0;
	info.allocation = allocation;
	info.skb = skb;
	info.skb2 = NULL;

	/* While we sleep in clone, do not allow to change socket list */

	netlink_lock_table();

	sk_for_each_bound(sk, node, &nl_table[ssk->sk_protocol].mc_list)
		do_one_broadcast(sk, &info);

949 950
	kfree_skb(skb);

L
Linus Torvalds 已提交
951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980
	netlink_unlock_table();

	if (info.skb2)
		kfree_skb(info.skb2);

	if (info.delivered) {
		if (info.congested && (allocation & __GFP_WAIT))
			yield();
		return 0;
	}
	if (info.failure)
		return -ENOBUFS;
	return -ESRCH;
}

struct netlink_set_err_data {
	struct sock *exclude_sk;
	u32 pid;
	u32 group;
	int code;
};

static inline int do_one_set_err(struct sock *sk,
				 struct netlink_set_err_data *p)
{
	struct netlink_sock *nlk = nlk_sk(sk);

	if (sk == p->exclude_sk)
		goto out;

981 982
	if (nlk->pid == p->pid || p->group - 1 >= nlk->ngroups ||
	    !test_bit(p->group - 1, nlk->groups))
L
Linus Torvalds 已提交
983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009
		goto out;

	sk->sk_err = p->code;
	sk->sk_error_report(sk);
out:
	return 0;
}

void netlink_set_err(struct sock *ssk, u32 pid, u32 group, int code)
{
	struct netlink_set_err_data info;
	struct hlist_node *node;
	struct sock *sk;

	info.exclude_sk = ssk;
	info.pid = pid;
	info.group = group;
	info.code = code;

	read_lock(&nl_table_lock);

	sk_for_each_bound(sk, node, &nl_table[ssk->sk_protocol].mc_list)
		do_one_set_err(sk, &info);

	read_unlock(&nl_table_lock);
}

1010
static int netlink_setsockopt(struct socket *sock, int level, int optname,
1011
			      char __user *optval, int optlen)
1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038
{
	struct sock *sk = sock->sk;
	struct netlink_sock *nlk = nlk_sk(sk);
	int val = 0, err;

	if (level != SOL_NETLINK)
		return -ENOPROTOOPT;

	if (optlen >= sizeof(int) &&
	    get_user(val, (int __user *)optval))
		return -EFAULT;

	switch (optname) {
	case NETLINK_PKTINFO:
		if (val)
			nlk->flags |= NETLINK_RECV_PKTINFO;
		else
			nlk->flags &= ~NETLINK_RECV_PKTINFO;
		err = 0;
		break;
	case NETLINK_ADD_MEMBERSHIP:
	case NETLINK_DROP_MEMBERSHIP: {
		unsigned int subscriptions;
		int old, new = optname == NETLINK_ADD_MEMBERSHIP ? 1 : 0;

		if (!netlink_capable(sock, NL_NONROOT_RECV))
			return -EPERM;
1039 1040 1041 1042 1043
		if (nlk->groups == NULL) {
			err = netlink_alloc_groups(sk);
			if (err)
				return err;
		}
1044 1045 1046 1047 1048 1049 1050 1051 1052 1053
		if (!val || val - 1 >= nlk->ngroups)
			return -EINVAL;
		netlink_table_grab();
		old = test_bit(val - 1, nlk->groups);
		subscriptions = nlk->subscriptions - old + new;
		if (new)
			__set_bit(val - 1, nlk->groups);
		else
			__clear_bit(val - 1, nlk->groups);
		netlink_update_subscriptions(sk, subscriptions);
1054
		netlink_update_listeners(sk);
1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065
		netlink_table_ungrab();
		err = 0;
		break;
	}
	default:
		err = -ENOPROTOOPT;
	}
	return err;
}

static int netlink_getsockopt(struct socket *sock, int level, int optname,
1066
			      char __user *optval, int __user *optlen)
1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085
{
	struct sock *sk = sock->sk;
	struct netlink_sock *nlk = nlk_sk(sk);
	int len, val, err;

	if (level != SOL_NETLINK)
		return -ENOPROTOOPT;

	if (get_user(len, optlen))
		return -EFAULT;
	if (len < 0)
		return -EINVAL;

	switch (optname) {
	case NETLINK_PKTINFO:
		if (len < sizeof(int))
			return -EINVAL;
		len = sizeof(int);
		val = nlk->flags & NETLINK_RECV_PKTINFO ? 1 : 0;
H
Heiko Carstens 已提交
1086 1087 1088
		if (put_user(len, optlen) ||
		    put_user(val, optval))
			return -EFAULT;
1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104
		err = 0;
		break;
	default:
		err = -ENOPROTOOPT;
	}
	return err;
}

static void netlink_cmsg_recv_pktinfo(struct msghdr *msg, struct sk_buff *skb)
{
	struct nl_pktinfo info;

	info.group = NETLINK_CB(skb).dst_group;
	put_cmsg(msg, SOL_NETLINK, NETLINK_PKTINFO, sizeof(info), &info);
}

L
Linus Torvalds 已提交
1105 1106 1107 1108
static inline void netlink_rcv_wake(struct sock *sk)
{
	struct netlink_sock *nlk = nlk_sk(sk);

1109
	if (skb_queue_empty(&sk->sk_receive_queue))
L
Linus Torvalds 已提交
1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122
		clear_bit(0, &nlk->state);
	if (!test_bit(0, &nlk->state))
		wake_up_interruptible(&nlk->wait);
}

static int netlink_sendmsg(struct kiocb *kiocb, struct socket *sock,
			   struct msghdr *msg, size_t len)
{
	struct sock_iocb *siocb = kiocb_to_siocb(kiocb);
	struct sock *sk = sock->sk;
	struct netlink_sock *nlk = nlk_sk(sk);
	struct sockaddr_nl *addr=msg->msg_name;
	u32 dst_pid;
1123
	u32 dst_group;
L
Linus Torvalds 已提交
1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140
	struct sk_buff *skb;
	int err;
	struct scm_cookie scm;

	if (msg->msg_flags&MSG_OOB)
		return -EOPNOTSUPP;

	if (NULL == siocb->scm)
		siocb->scm = &scm;
	err = scm_send(sock, msg, siocb->scm);
	if (err < 0)
		return err;

	if (msg->msg_namelen) {
		if (addr->nl_family != AF_NETLINK)
			return -EINVAL;
		dst_pid = addr->nl_pid;
1141 1142
		dst_group = ffs(addr->nl_groups);
		if (dst_group && !netlink_capable(sock, NL_NONROOT_SEND))
L
Linus Torvalds 已提交
1143 1144 1145
			return -EPERM;
	} else {
		dst_pid = nlk->dst_pid;
1146
		dst_group = nlk->dst_group;
L
Linus Torvalds 已提交
1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
	}

	if (!nlk->pid) {
		err = netlink_autobind(sock);
		if (err)
			goto out;
	}

	err = -EMSGSIZE;
	if (len > sk->sk_sndbuf - 32)
		goto out;
	err = -ENOBUFS;
1159
	skb = alloc_skb(len, GFP_KERNEL);
L
Linus Torvalds 已提交
1160 1161 1162 1163
	if (skb==NULL)
		goto out;

	NETLINK_CB(skb).pid	= nlk->pid;
1164
	NETLINK_CB(skb).dst_group = dst_group;
1165
	NETLINK_CB(skb).loginuid = audit_get_loginuid(current->audit_context);
1166
	selinux_get_task_sid(current, &(NETLINK_CB(skb).sid));
L
Linus Torvalds 已提交
1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186
	memcpy(NETLINK_CREDS(skb), &siocb->scm->creds, sizeof(struct ucred));

	/* What can I do? Netlink is asynchronous, so that
	   we will have to save current capabilities to
	   check them, when this message will be delivered
	   to corresponding kernel module.   --ANK (980802)
	 */

	err = -EFAULT;
	if (memcpy_fromiovec(skb_put(skb,len), msg->msg_iov, len)) {
		kfree_skb(skb);
		goto out;
	}

	err = security_netlink_send(sk, skb);
	if (err) {
		kfree_skb(skb);
		goto out;
	}

1187
	if (dst_group) {
L
Linus Torvalds 已提交
1188
		atomic_inc(&skb->users);
1189
		netlink_broadcast(sk, skb, dst_pid, dst_group, GFP_KERNEL);
L
Linus Torvalds 已提交
1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226
	}
	err = netlink_unicast(sk, skb, dst_pid, msg->msg_flags&MSG_DONTWAIT);

out:
	return err;
}

static int netlink_recvmsg(struct kiocb *kiocb, struct socket *sock,
			   struct msghdr *msg, size_t len,
			   int flags)
{
	struct sock_iocb *siocb = kiocb_to_siocb(kiocb);
	struct scm_cookie scm;
	struct sock *sk = sock->sk;
	struct netlink_sock *nlk = nlk_sk(sk);
	int noblock = flags&MSG_DONTWAIT;
	size_t copied;
	struct sk_buff *skb;
	int err;

	if (flags&MSG_OOB)
		return -EOPNOTSUPP;

	copied = 0;

	skb = skb_recv_datagram(sk,flags,noblock,&err);
	if (skb==NULL)
		goto out;

	msg->msg_namelen = 0;

	copied = skb->len;
	if (len < copied) {
		msg->msg_flags |= MSG_TRUNC;
		copied = len;
	}

1227
	skb_reset_transport_header(skb);
L
Linus Torvalds 已提交
1228 1229 1230 1231 1232 1233 1234
	err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);

	if (msg->msg_name) {
		struct sockaddr_nl *addr = (struct sockaddr_nl*)msg->msg_name;
		addr->nl_family = AF_NETLINK;
		addr->nl_pad    = 0;
		addr->nl_pid	= NETLINK_CB(skb).pid;
1235
		addr->nl_groups	= netlink_group_mask(NETLINK_CB(skb).dst_group);
L
Linus Torvalds 已提交
1236 1237 1238
		msg->msg_namelen = sizeof(*addr);
	}

1239 1240 1241
	if (nlk->flags & NETLINK_RECV_PKTINFO)
		netlink_cmsg_recv_pktinfo(msg, skb);

L
Linus Torvalds 已提交
1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253
	if (NULL == siocb->scm) {
		memset(&scm, 0, sizeof(scm));
		siocb->scm = &scm;
	}
	siocb->scm->creds = *NETLINK_CREDS(skb);
	skb_free_datagram(sk, skb);

	if (nlk->cb && atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf / 2)
		netlink_dump(sk);

	scm_recv(sock, msg, siocb->scm, flags);

1254 1255 1256
	if (flags & MSG_TRUNC)
		copied = skb->len;

L
Linus Torvalds 已提交
1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271
out:
	netlink_rcv_wake(sk);
	return err ? : copied;
}

static void netlink_data_ready(struct sock *sk, int len)
{
	struct netlink_sock *nlk = nlk_sk(sk);

	if (nlk->data_ready)
		nlk->data_ready(sk, len);
	netlink_rcv_wake(sk);
}

/*
1272
 *	We export these functions to other modules. They provide a
L
Linus Torvalds 已提交
1273 1274 1275 1276 1277
 *	complete set of kernel non-blocking support for message
 *	queueing.
 */

struct sock *
1278
netlink_kernel_create(int unit, unsigned int groups,
1279
		      void (*input)(struct sock *sk, int len),
1280
		      struct mutex *cb_mutex, struct module *module)
L
Linus Torvalds 已提交
1281 1282 1283
{
	struct socket *sock;
	struct sock *sk;
1284
	struct netlink_sock *nlk;
1285
	unsigned long *listeners = NULL;
L
Linus Torvalds 已提交
1286

1287
	BUG_ON(!nl_table);
L
Linus Torvalds 已提交
1288 1289 1290 1291 1292 1293 1294

	if (unit<0 || unit>=MAX_LINKS)
		return NULL;

	if (sock_create_lite(PF_NETLINK, SOCK_DGRAM, unit, &sock))
		return NULL;

1295
	if (__netlink_create(sock, cb_mutex, unit) < 0)
1296
		goto out_sock_release;
1297

1298 1299 1300 1301 1302 1303 1304
	if (groups < 32)
		groups = 32;

	listeners = kzalloc(NLGRPSZ(groups), GFP_KERNEL);
	if (!listeners)
		goto out_sock_release;

L
Linus Torvalds 已提交
1305 1306 1307 1308 1309
	sk = sock->sk;
	sk->sk_data_ready = netlink_data_ready;
	if (input)
		nlk_sk(sk)->data_ready = input;

1310 1311
	if (netlink_insert(sk, 0))
		goto out_sock_release;
1312

1313 1314
	nlk = nlk_sk(sk);
	nlk->flags |= NETLINK_KERNEL_SOCKET;
1315 1316

	netlink_table_grab();
1317 1318
	nl_table[unit].groups = groups;
	nl_table[unit].listeners = listeners;
1319
	nl_table[unit].cb_mutex = cb_mutex;
1320
	nl_table[unit].module = module;
1321
	nl_table[unit].registered = 1;
1322
	netlink_table_ungrab();
1323 1324 1325

	return sk;

1326
out_sock_release:
1327
	kfree(listeners);
1328
	sock_release(sock);
1329
	return NULL;
L
Linus Torvalds 已提交
1330 1331 1332
}

void netlink_set_nonroot(int protocol, unsigned int flags)
1333 1334
{
	if ((unsigned int)protocol < MAX_LINKS)
L
Linus Torvalds 已提交
1335
		nl_table[protocol].nl_nonroot = flags;
1336
}
L
Linus Torvalds 已提交
1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355

static void netlink_destroy_callback(struct netlink_callback *cb)
{
	if (cb->skb)
		kfree_skb(cb->skb);
	kfree(cb);
}

/*
 * It looks a bit ugly.
 * It would be better to create kernel thread.
 */

static int netlink_dump(struct sock *sk)
{
	struct netlink_sock *nlk = nlk_sk(sk);
	struct netlink_callback *cb;
	struct sk_buff *skb;
	struct nlmsghdr *nlh;
1356
	int len, err = -ENOBUFS;
1357

L
Linus Torvalds 已提交
1358 1359
	skb = sock_rmalloc(sk, NLMSG_GOODSIZE, 0, GFP_KERNEL);
	if (!skb)
1360
		goto errout;
L
Linus Torvalds 已提交
1361

1362
	mutex_lock(nlk->cb_mutex);
L
Linus Torvalds 已提交
1363 1364 1365

	cb = nlk->cb;
	if (cb == NULL) {
1366 1367
		err = -EINVAL;
		goto errout_skb;
L
Linus Torvalds 已提交
1368 1369 1370 1371 1372
	}

	len = cb->dump(skb, cb);

	if (len > 0) {
1373
		mutex_unlock(nlk->cb_mutex);
L
Linus Torvalds 已提交
1374 1375 1376 1377 1378
		skb_queue_tail(&sk->sk_receive_queue, skb);
		sk->sk_data_ready(sk, len);
		return 0;
	}

1379 1380 1381 1382 1383 1384
	nlh = nlmsg_put_answer(skb, cb, NLMSG_DONE, sizeof(len), NLM_F_MULTI);
	if (!nlh)
		goto errout_skb;

	memcpy(nlmsg_data(nlh), &len, sizeof(len));

L
Linus Torvalds 已提交
1385 1386 1387
	skb_queue_tail(&sk->sk_receive_queue, skb);
	sk->sk_data_ready(sk, skb->len);

1388 1389
	if (cb->done)
		cb->done(cb);
L
Linus Torvalds 已提交
1390
	nlk->cb = NULL;
1391
	mutex_unlock(nlk->cb_mutex);
L
Linus Torvalds 已提交
1392 1393 1394

	netlink_destroy_callback(cb);
	return 0;
1395

1396
errout_skb:
1397
	mutex_unlock(nlk->cb_mutex);
1398 1399 1400
	kfree_skb(skb);
errout:
	return err;
L
Linus Torvalds 已提交
1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411
}

int netlink_dump_start(struct sock *ssk, struct sk_buff *skb,
		       struct nlmsghdr *nlh,
		       int (*dump)(struct sk_buff *skb, struct netlink_callback*),
		       int (*done)(struct netlink_callback*))
{
	struct netlink_callback *cb;
	struct sock *sk;
	struct netlink_sock *nlk;

1412
	cb = kzalloc(sizeof(*cb), GFP_KERNEL);
L
Linus Torvalds 已提交
1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427
	if (cb == NULL)
		return -ENOBUFS;

	cb->dump = dump;
	cb->done = done;
	cb->nlh = nlh;
	atomic_inc(&skb->users);
	cb->skb = skb;

	sk = netlink_lookup(ssk->sk_protocol, NETLINK_CB(skb).pid);
	if (sk == NULL) {
		netlink_destroy_callback(cb);
		return -ECONNREFUSED;
	}
	nlk = nlk_sk(sk);
1428
	/* A dump or destruction is in progress... */
1429
	mutex_lock(nlk->cb_mutex);
1430
	if (nlk->cb || sock_flag(sk, SOCK_DEAD)) {
1431
		mutex_unlock(nlk->cb_mutex);
L
Linus Torvalds 已提交
1432 1433 1434 1435 1436
		netlink_destroy_callback(cb);
		sock_put(sk);
		return -EBUSY;
	}
	nlk->cb = cb;
1437
	mutex_unlock(nlk->cb_mutex);
L
Linus Torvalds 已提交
1438 1439 1440

	netlink_dump(sk);
	sock_put(sk);
1441 1442 1443 1444 1445 1446

	/* We successfully started a dump, by returning -EINTR we
	 * signal the queue mangement to interrupt processing of
	 * any netlink messages so userspace gets a chance to read
	 * the results. */
	return -EINTR;
L
Linus Torvalds 已提交
1447 1448 1449 1450 1451 1452 1453
}

void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err)
{
	struct sk_buff *skb;
	struct nlmsghdr *rep;
	struct nlmsgerr *errmsg;
1454
	size_t payload = sizeof(*errmsg);
L
Linus Torvalds 已提交
1455

1456 1457 1458
	/* error messages get the original request appened */
	if (err)
		payload += nlmsg_len(nlh);
L
Linus Torvalds 已提交
1459

1460
	skb = nlmsg_new(payload, GFP_KERNEL);
L
Linus Torvalds 已提交
1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474
	if (!skb) {
		struct sock *sk;

		sk = netlink_lookup(in_skb->sk->sk_protocol,
				    NETLINK_CB(in_skb).pid);
		if (sk) {
			sk->sk_err = ENOBUFS;
			sk->sk_error_report(sk);
			sock_put(sk);
		}
		return;
	}

	rep = __nlmsg_put(skb, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq,
1475
			  NLMSG_ERROR, sizeof(struct nlmsgerr), 0);
1476
	errmsg = nlmsg_data(rep);
L
Linus Torvalds 已提交
1477
	errmsg->error = err;
1478
	memcpy(&errmsg->msg, nlh, err ? nlh->nlmsg_len : sizeof(*nlh));
L
Linus Torvalds 已提交
1479 1480 1481
	netlink_unicast(in_skb->sk, skb, NETLINK_CB(in_skb).pid, MSG_DONTWAIT);
}

1482
static int netlink_rcv_skb(struct sk_buff *skb, int (*cb)(struct sk_buff *,
1483
						     struct nlmsghdr *))
1484 1485 1486 1487 1488
{
	struct nlmsghdr *nlh;
	int err;

	while (skb->len >= nlmsg_total_size(0)) {
1489
		nlh = nlmsg_hdr(skb);
1490
		err = 0;
1491

1492
		if (nlh->nlmsg_len < NLMSG_HDRLEN || skb->len < nlh->nlmsg_len)
1493 1494
			return 0;

1495 1496 1497
		/* Only requests are handled by the kernel */
		if (!(nlh->nlmsg_flags & NLM_F_REQUEST))
			goto skip;
1498 1499 1500 1501

		/* Skip control messages */
		if (nlh->nlmsg_type < NLMSG_MIN_TYPE)
			goto skip;
1502

1503 1504 1505 1506 1507
		err = cb(skb, nlh);
		if (err == -EINTR) {
			/* Not an error, but we interrupt processing */
			netlink_queue_skip(nlh, skb);
			return err;
1508 1509 1510
		}
skip:
		if (nlh->nlmsg_flags & NLM_F_ACK || err)
1511 1512
			netlink_ack(skb, nlh, err);

1513
		netlink_queue_skip(nlh, skb);
1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532
	}

	return 0;
}

/**
 * nelink_run_queue - Process netlink receive queue.
 * @sk: Netlink socket containing the queue
 * @qlen: Place to store queue length upon entry
 * @cb: Callback function invoked for each netlink message found
 *
 * Processes as much as there was in the queue upon entry and invokes
 * a callback function for each netlink message found. The callback
 * function may refuse a message by returning a negative error code
 * but setting the error pointer to 0 in which case this function
 * returns with a qlen != 0.
 *
 * qlen must be initialized to 0 before the initial entry, afterwards
 * the function may be called repeatedly until qlen reaches 0.
1533 1534 1535 1536 1537
 *
 * The callback function may return -EINTR to signal that processing
 * of netlink messages shall be interrupted. In this case the message
 * currently being processed will NOT be requeued onto the receive
 * queue.
1538 1539
 */
void netlink_run_queue(struct sock *sk, unsigned int *qlen,
1540
		       int (*cb)(struct sk_buff *, struct nlmsghdr *))
1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579
{
	struct sk_buff *skb;

	if (!*qlen || *qlen > skb_queue_len(&sk->sk_receive_queue))
		*qlen = skb_queue_len(&sk->sk_receive_queue);

	for (; *qlen; (*qlen)--) {
		skb = skb_dequeue(&sk->sk_receive_queue);
		if (netlink_rcv_skb(skb, cb)) {
			if (skb->len)
				skb_queue_head(&sk->sk_receive_queue, skb);
			else {
				kfree_skb(skb);
				(*qlen)--;
			}
			break;
		}

		kfree_skb(skb);
	}
}

/**
 * netlink_queue_skip - Skip netlink message while processing queue.
 * @nlh: Netlink message to be skipped
 * @skb: Socket buffer containing the netlink messages.
 *
 * Pulls the given netlink message off the socket buffer so the next
 * call to netlink_queue_run() will not reconsider the message.
 */
void netlink_queue_skip(struct nlmsghdr *nlh, struct sk_buff *skb)
{
	int msglen = NLMSG_ALIGN(nlh->nlmsg_len);

	if (msglen > skb->len)
		msglen = skb->len;

	skb_pull(skb, msglen);
}
L
Linus Torvalds 已提交
1580

1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612
/**
 * nlmsg_notify - send a notification netlink message
 * @sk: netlink socket to use
 * @skb: notification message
 * @pid: destination netlink pid for reports or 0
 * @group: destination multicast group or 0
 * @report: 1 to report back, 0 to disable
 * @flags: allocation flags
 */
int nlmsg_notify(struct sock *sk, struct sk_buff *skb, u32 pid,
		 unsigned int group, int report, gfp_t flags)
{
	int err = 0;

	if (group) {
		int exclude_pid = 0;

		if (report) {
			atomic_inc(&skb->users);
			exclude_pid = pid;
		}

		/* errors reported via destination sk->sk_err */
		nlmsg_multicast(sk, skb, exclude_pid, group, flags);
	}

	if (report)
		err = nlmsg_unicast(sk, skb, pid);

	return err;
}

L
Linus Torvalds 已提交
1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659
#ifdef CONFIG_PROC_FS
struct nl_seq_iter {
	int link;
	int hash_idx;
};

static struct sock *netlink_seq_socket_idx(struct seq_file *seq, loff_t pos)
{
	struct nl_seq_iter *iter = seq->private;
	int i, j;
	struct sock *s;
	struct hlist_node *node;
	loff_t off = 0;

	for (i=0; i<MAX_LINKS; i++) {
		struct nl_pid_hash *hash = &nl_table[i].hash;

		for (j = 0; j <= hash->mask; j++) {
			sk_for_each(s, node, &hash->table[j]) {
				if (off == pos) {
					iter->link = i;
					iter->hash_idx = j;
					return s;
				}
				++off;
			}
		}
	}
	return NULL;
}

static void *netlink_seq_start(struct seq_file *seq, loff_t *pos)
{
	read_lock(&nl_table_lock);
	return *pos ? netlink_seq_socket_idx(seq, *pos - 1) : SEQ_START_TOKEN;
}

static void *netlink_seq_next(struct seq_file *seq, void *v, loff_t *pos)
{
	struct sock *s;
	struct nl_seq_iter *iter;
	int i, j;

	++*pos;

	if (v == SEQ_START_TOKEN)
		return netlink_seq_socket_idx(seq, 0);
1660

L
Linus Torvalds 已提交
1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706
	s = sk_next(v);
	if (s)
		return s;

	iter = seq->private;
	i = iter->link;
	j = iter->hash_idx + 1;

	do {
		struct nl_pid_hash *hash = &nl_table[i].hash;

		for (; j <= hash->mask; j++) {
			s = sk_head(&hash->table[j]);
			if (s) {
				iter->link = i;
				iter->hash_idx = j;
				return s;
			}
		}

		j = 0;
	} while (++i < MAX_LINKS);

	return NULL;
}

static void netlink_seq_stop(struct seq_file *seq, void *v)
{
	read_unlock(&nl_table_lock);
}


static int netlink_seq_show(struct seq_file *seq, void *v)
{
	if (v == SEQ_START_TOKEN)
		seq_puts(seq,
			 "sk       Eth Pid    Groups   "
			 "Rmem     Wmem     Dump     Locks\n");
	else {
		struct sock *s = v;
		struct netlink_sock *nlk = nlk_sk(s);

		seq_printf(seq, "%p %-3d %-6d %08x %-8d %-8d %p %d\n",
			   s,
			   s->sk_protocol,
			   nlk->pid,
1707
			   nlk->groups ? (u32)nlk->groups[0] : 0,
L
Linus Torvalds 已提交
1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731
			   atomic_read(&s->sk_rmem_alloc),
			   atomic_read(&s->sk_wmem_alloc),
			   nlk->cb,
			   atomic_read(&s->sk_refcnt)
			);

	}
	return 0;
}

static struct seq_operations netlink_seq_ops = {
	.start  = netlink_seq_start,
	.next   = netlink_seq_next,
	.stop   = netlink_seq_stop,
	.show   = netlink_seq_show,
};


static int netlink_seq_open(struct inode *inode, struct file *file)
{
	struct seq_file *seq;
	struct nl_seq_iter *iter;
	int err;

1732
	iter = kzalloc(sizeof(*iter), GFP_KERNEL);
L
Linus Torvalds 已提交
1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746
	if (!iter)
		return -ENOMEM;

	err = seq_open(file, &netlink_seq_ops);
	if (err) {
		kfree(iter);
		return err;
	}

	seq = file->private_data;
	seq->private = iter;
	return 0;
}

1747
static const struct file_operations netlink_seq_fops = {
L
Linus Torvalds 已提交
1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758
	.owner		= THIS_MODULE,
	.open		= netlink_seq_open,
	.read		= seq_read,
	.llseek		= seq_lseek,
	.release	= seq_release_private,
};

#endif

int netlink_register_notifier(struct notifier_block *nb)
{
1759
	return atomic_notifier_chain_register(&netlink_chain, nb);
L
Linus Torvalds 已提交
1760 1761 1762 1763
}

int netlink_unregister_notifier(struct notifier_block *nb)
{
1764
	return atomic_notifier_chain_unregister(&netlink_chain, nb);
L
Linus Torvalds 已提交
1765
}
1766

1767
static const struct proto_ops netlink_ops = {
L
Linus Torvalds 已提交
1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779
	.family =	PF_NETLINK,
	.owner =	THIS_MODULE,
	.release =	netlink_release,
	.bind =		netlink_bind,
	.connect =	netlink_connect,
	.socketpair =	sock_no_socketpair,
	.accept =	sock_no_accept,
	.getname =	netlink_getname,
	.poll =		datagram_poll,
	.ioctl =	sock_no_ioctl,
	.listen =	sock_no_listen,
	.shutdown =	sock_no_shutdown,
1780 1781
	.setsockopt =	netlink_setsockopt,
	.getsockopt =	netlink_getsockopt,
L
Linus Torvalds 已提交
1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804
	.sendmsg =	netlink_sendmsg,
	.recvmsg =	netlink_recvmsg,
	.mmap =		sock_no_mmap,
	.sendpage =	sock_no_sendpage,
};

static struct net_proto_family netlink_family_ops = {
	.family = PF_NETLINK,
	.create = netlink_create,
	.owner	= THIS_MODULE,	/* for consistency 8) */
};

static int __init netlink_proto_init(void)
{
	struct sk_buff *dummy_skb;
	int i;
	unsigned long max;
	unsigned int order;
	int err = proto_register(&netlink_proto, 0);

	if (err != 0)
		goto out;

1805
	BUILD_BUG_ON(sizeof(struct netlink_skb_parms) > sizeof(dummy_skb->cb));
L
Linus Torvalds 已提交
1806

1807
	nl_table = kcalloc(MAX_LINKS, sizeof(*nl_table), GFP_KERNEL);
1808 1809
	if (!nl_table)
		goto panic;
L
Linus Torvalds 已提交
1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828

	if (num_physpages >= (128 * 1024))
		max = num_physpages >> (21 - PAGE_SHIFT);
	else
		max = num_physpages >> (23 - PAGE_SHIFT);

	order = get_bitmask_order(max) - 1 + PAGE_SHIFT;
	max = (1UL << order) / sizeof(struct hlist_head);
	order = get_bitmask_order(max > UINT_MAX ? UINT_MAX : max) - 1;

	for (i = 0; i < MAX_LINKS; i++) {
		struct nl_pid_hash *hash = &nl_table[i].hash;

		hash->table = nl_pid_hash_alloc(1 * sizeof(*hash->table));
		if (!hash->table) {
			while (i-- > 0)
				nl_pid_hash_free(nl_table[i].hash.table,
						 1 * sizeof(*hash->table));
			kfree(nl_table);
1829
			goto panic;
L
Linus Torvalds 已提交
1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841
		}
		memset(hash->table, 0, 1 * sizeof(*hash->table));
		hash->max_shift = order;
		hash->shift = 0;
		hash->mask = 0;
		hash->rehash_time = jiffies;
	}

	sock_register(&netlink_family_ops);
#ifdef CONFIG_PROC_FS
	proc_net_fops_create("netlink", 0, &netlink_seq_fops);
#endif
1842
	/* The netlink device handler may be needed early. */
L
Linus Torvalds 已提交
1843 1844 1845
	rtnetlink_init();
out:
	return err;
1846 1847
panic:
	panic("netlink_init: Cannot allocate nl_table\n");
L
Linus Torvalds 已提交
1848 1849 1850 1851 1852
}

core_initcall(netlink_proto_init);

EXPORT_SYMBOL(netlink_ack);
1853 1854
EXPORT_SYMBOL(netlink_run_queue);
EXPORT_SYMBOL(netlink_queue_skip);
L
Linus Torvalds 已提交
1855 1856 1857 1858 1859 1860 1861
EXPORT_SYMBOL(netlink_broadcast);
EXPORT_SYMBOL(netlink_dump_start);
EXPORT_SYMBOL(netlink_kernel_create);
EXPORT_SYMBOL(netlink_register_notifier);
EXPORT_SYMBOL(netlink_set_nonroot);
EXPORT_SYMBOL(netlink_unicast);
EXPORT_SYMBOL(netlink_unregister_notifier);
1862
EXPORT_SYMBOL(nlmsg_notify);