rtnetlink.c 104.0 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
/*
 * INET		An implementation of the TCP/IP protocol suite for the LINUX
 *		operating system.  INET is implemented using the  BSD Socket
 *		interface as the means of communication with the user level.
 *
 *		Routing netlink socket interface: protocol independent part.
 *
 * Authors:	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.
 *
 *	Fixes:
 *	Vitaly E. Lavrov		RTA_OK arithmetics was wrong.
 */

19
#include <linux/bitops.h>
L
Linus Torvalds 已提交
20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
#include <linux/errno.h>
#include <linux/module.h>
#include <linux/types.h>
#include <linux/socket.h>
#include <linux/kernel.h>
#include <linux/timer.h>
#include <linux/string.h>
#include <linux/sockios.h>
#include <linux/net.h>
#include <linux/fcntl.h>
#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/interrupt.h>
#include <linux/capability.h>
#include <linux/skbuff.h>
#include <linux/init.h>
#include <linux/security.h>
37
#include <linux/mutex.h>
38
#include <linux/if_addr.h>
39
#include <linux/if_bridge.h>
40
#include <linux/if_vlan.h>
41
#include <linux/pci.h>
42
#include <linux/etherdevice.h>
M
Martin KaFai Lau 已提交
43
#include <linux/bpf.h>
L
Linus Torvalds 已提交
44

45
#include <linux/uaccess.h>
L
Linus Torvalds 已提交
46 47 48

#include <linux/inet.h>
#include <linux/netdevice.h>
49
#include <net/switchdev.h>
L
Linus Torvalds 已提交
50 51 52 53 54
#include <net/ip.h>
#include <net/protocol.h>
#include <net/arp.h>
#include <net/route.h>
#include <net/udp.h>
55
#include <net/tcp.h>
L
Linus Torvalds 已提交
56 57
#include <net/sock.h>
#include <net/pkt_sched.h>
58
#include <net/fib_rules.h>
59
#include <net/rtnetlink.h>
60
#include <net/net_namespace.h>
L
Linus Torvalds 已提交
61

E
Eric Dumazet 已提交
62
struct rtnl_link {
63 64
	rtnl_doit_func		doit;
	rtnl_dumpit_func	dumpit;
65
	rtnl_calcit_func 	calcit;
66 67
};

68
static DEFINE_MUTEX(rtnl_mutex);
L
Linus Torvalds 已提交
69 70 71

void rtnl_lock(void)
{
72
	mutex_lock(&rtnl_mutex);
L
Linus Torvalds 已提交
73
}
E
Eric Dumazet 已提交
74
EXPORT_SYMBOL(rtnl_lock);
L
Linus Torvalds 已提交
75

76 77 78 79 80 81 82 83 84 85
static struct sk_buff *defer_kfree_skb_list;
void rtnl_kfree_skbs(struct sk_buff *head, struct sk_buff *tail)
{
	if (head && tail) {
		tail->next = defer_kfree_skb_list;
		defer_kfree_skb_list = head;
	}
}
EXPORT_SYMBOL(rtnl_kfree_skbs);

86
void __rtnl_unlock(void)
L
Linus Torvalds 已提交
87
{
88 89 90 91
	struct sk_buff *head = defer_kfree_skb_list;

	defer_kfree_skb_list = NULL;

92
	mutex_unlock(&rtnl_mutex);
93 94 95 96 97 98 99 100

	while (head) {
		struct sk_buff *next = head->next;

		kfree_skb(head);
		cond_resched();
		head = next;
	}
L
Linus Torvalds 已提交
101
}
102

L
Linus Torvalds 已提交
103 104
void rtnl_unlock(void)
{
H
Herbert Xu 已提交
105
	/* This fellow will unlock it for us. */
L
Linus Torvalds 已提交
106 107
	netdev_run_todo();
}
E
Eric Dumazet 已提交
108
EXPORT_SYMBOL(rtnl_unlock);
L
Linus Torvalds 已提交
109

110 111 112 113
int rtnl_trylock(void)
{
	return mutex_trylock(&rtnl_mutex);
}
E
Eric Dumazet 已提交
114
EXPORT_SYMBOL(rtnl_trylock);
115

116 117 118 119
int rtnl_is_locked(void)
{
	return mutex_is_locked(&rtnl_mutex);
}
E
Eric Dumazet 已提交
120
EXPORT_SYMBOL(rtnl_is_locked);
121

122
#ifdef CONFIG_PROVE_LOCKING
123
bool lockdep_rtnl_is_held(void)
124 125 126 127 128 129
{
	return lockdep_is_held(&rtnl_mutex);
}
EXPORT_SYMBOL(lockdep_rtnl_is_held);
#endif /* #ifdef CONFIG_PROVE_LOCKING */

130
static struct rtnl_link *rtnl_msg_handlers[RTNL_FAMILY_MAX + 1];
131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149

static inline int rtm_msgindex(int msgtype)
{
	int msgindex = msgtype - RTM_BASE;

	/*
	 * msgindex < 0 implies someone tried to register a netlink
	 * control code. msgindex >= RTM_NR_MSGTYPES may indicate that
	 * the message type has not been added to linux/rtnetlink.h
	 */
	BUG_ON(msgindex < 0 || msgindex >= RTM_NR_MSGTYPES);

	return msgindex;
}

static rtnl_doit_func rtnl_get_doit(int protocol, int msgindex)
{
	struct rtnl_link *tab;

150
	if (protocol <= RTNL_FAMILY_MAX)
151 152 153 154
		tab = rtnl_msg_handlers[protocol];
	else
		tab = NULL;

155
	if (tab == NULL || tab[msgindex].doit == NULL)
156 157
		tab = rtnl_msg_handlers[PF_UNSPEC];

158
	return tab[msgindex].doit;
159 160 161 162 163 164
}

static rtnl_dumpit_func rtnl_get_dumpit(int protocol, int msgindex)
{
	struct rtnl_link *tab;

165
	if (protocol <= RTNL_FAMILY_MAX)
166 167 168 169
		tab = rtnl_msg_handlers[protocol];
	else
		tab = NULL;

170
	if (tab == NULL || tab[msgindex].dumpit == NULL)
171 172
		tab = rtnl_msg_handlers[PF_UNSPEC];

173
	return tab[msgindex].dumpit;
174 175
}

176 177 178 179 180 181 182 183 184 185 186 187
static rtnl_calcit_func rtnl_get_calcit(int protocol, int msgindex)
{
	struct rtnl_link *tab;

	if (protocol <= RTNL_FAMILY_MAX)
		tab = rtnl_msg_handlers[protocol];
	else
		tab = NULL;

	if (tab == NULL || tab[msgindex].calcit == NULL)
		tab = rtnl_msg_handlers[PF_UNSPEC];

188
	return tab[msgindex].calcit;
189 190
}

191 192 193 194 195 196
/**
 * __rtnl_register - Register a rtnetlink message type
 * @protocol: Protocol family or PF_UNSPEC
 * @msgtype: rtnetlink message type
 * @doit: Function pointer called for each request message
 * @dumpit: Function pointer called for each dump request (NLM_F_DUMP) message
197
 * @calcit: Function pointer to calc size of dump message
198 199 200 201 202 203 204 205 206 207 208 209
 *
 * Registers the specified function pointers (at least one of them has
 * to be non-NULL) to be called whenever a request message for the
 * specified protocol family and message type is received.
 *
 * The special protocol family PF_UNSPEC may be used to define fallback
 * function pointers for the case when no entry for the specific protocol
 * family exists.
 *
 * Returns 0 on success or a negative error code.
 */
int __rtnl_register(int protocol, int msgtype,
210 211
		    rtnl_doit_func doit, rtnl_dumpit_func dumpit,
		    rtnl_calcit_func calcit)
212 213 214 215
{
	struct rtnl_link *tab;
	int msgindex;

216
	BUG_ON(protocol < 0 || protocol > RTNL_FAMILY_MAX);
217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233
	msgindex = rtm_msgindex(msgtype);

	tab = rtnl_msg_handlers[protocol];
	if (tab == NULL) {
		tab = kcalloc(RTM_NR_MSGTYPES, sizeof(*tab), GFP_KERNEL);
		if (tab == NULL)
			return -ENOBUFS;

		rtnl_msg_handlers[protocol] = tab;
	}

	if (doit)
		tab[msgindex].doit = doit;

	if (dumpit)
		tab[msgindex].dumpit = dumpit;

234 235 236
	if (calcit)
		tab[msgindex].calcit = calcit;

237 238 239 240 241 242 243 244 245 246 247
	return 0;
}
EXPORT_SYMBOL_GPL(__rtnl_register);

/**
 * rtnl_register - Register a rtnetlink message type
 *
 * Identical to __rtnl_register() but panics on failure. This is useful
 * as failure of this function is very unlikely, it can only happen due
 * to lack of memory when allocating the chain to store all message
 * handlers for a protocol. Meant for use in init functions where lack
L
Lucas De Marchi 已提交
248
 * of memory implies no sense in continuing.
249 250
 */
void rtnl_register(int protocol, int msgtype,
251 252
		   rtnl_doit_func doit, rtnl_dumpit_func dumpit,
		   rtnl_calcit_func calcit)
253
{
254
	if (__rtnl_register(protocol, msgtype, doit, dumpit, calcit) < 0)
255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271
		panic("Unable to register rtnetlink message handler, "
		      "protocol = %d, message type = %d\n",
		      protocol, msgtype);
}
EXPORT_SYMBOL_GPL(rtnl_register);

/**
 * rtnl_unregister - Unregister a rtnetlink message type
 * @protocol: Protocol family or PF_UNSPEC
 * @msgtype: rtnetlink message type
 *
 * Returns 0 on success or a negative error code.
 */
int rtnl_unregister(int protocol, int msgtype)
{
	int msgindex;

272
	BUG_ON(protocol < 0 || protocol > RTNL_FAMILY_MAX);
273 274 275 276 277 278 279
	msgindex = rtm_msgindex(msgtype);

	if (rtnl_msg_handlers[protocol] == NULL)
		return -ENOENT;

	rtnl_msg_handlers[protocol][msgindex].doit = NULL;
	rtnl_msg_handlers[protocol][msgindex].dumpit = NULL;
280
	rtnl_msg_handlers[protocol][msgindex].calcit = NULL;
281 282 283 284 285 286 287 288 289 290 291 292 293 294

	return 0;
}
EXPORT_SYMBOL_GPL(rtnl_unregister);

/**
 * rtnl_unregister_all - Unregister all rtnetlink message type of a protocol
 * @protocol : Protocol family or PF_UNSPEC
 *
 * Identical to calling rtnl_unregster() for all registered message types
 * of a certain protocol family.
 */
void rtnl_unregister_all(int protocol)
{
295
	BUG_ON(protocol < 0 || protocol > RTNL_FAMILY_MAX);
296 297 298 299 300

	kfree(rtnl_msg_handlers[protocol]);
	rtnl_msg_handlers[protocol] = NULL;
}
EXPORT_SYMBOL_GPL(rtnl_unregister_all);
L
Linus Torvalds 已提交
301

P
Patrick McHardy 已提交
302 303
static LIST_HEAD(link_ops);

304 305 306 307 308 309 310 311 312 313 314
static const struct rtnl_link_ops *rtnl_link_ops_get(const char *kind)
{
	const struct rtnl_link_ops *ops;

	list_for_each_entry(ops, &link_ops, list) {
		if (!strcmp(ops->kind, kind))
			return ops;
	}
	return NULL;
}

P
Patrick McHardy 已提交
315 316 317 318 319 320 321 322 323 324 325 326
/**
 * __rtnl_link_register - Register rtnl_link_ops with rtnetlink.
 * @ops: struct rtnl_link_ops * to register
 *
 * The caller must hold the rtnl_mutex. This function should be used
 * by drivers that create devices during module initialization. It
 * must be called before registering the devices.
 *
 * Returns 0 on success or a negative error code.
 */
int __rtnl_link_register(struct rtnl_link_ops *ops)
{
327 328 329
	if (rtnl_link_ops_get(ops->kind))
		return -EEXIST;

330 331 332 333 334 335
	/* The check for setup is here because if ops
	 * does not have that filled up, it is not possible
	 * to use the ops for creating device. So do not
	 * fill up dellink as well. That disables rtnl_dellink.
	 */
	if (ops->setup && !ops->dellink)
336
		ops->dellink = unregister_netdevice_queue;
337

P
Patrick McHardy 已提交
338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359
	list_add_tail(&ops->list, &link_ops);
	return 0;
}
EXPORT_SYMBOL_GPL(__rtnl_link_register);

/**
 * rtnl_link_register - Register rtnl_link_ops with rtnetlink.
 * @ops: struct rtnl_link_ops * to register
 *
 * Returns 0 on success or a negative error code.
 */
int rtnl_link_register(struct rtnl_link_ops *ops)
{
	int err;

	rtnl_lock();
	err = __rtnl_link_register(ops);
	rtnl_unlock();
	return err;
}
EXPORT_SYMBOL_GPL(rtnl_link_register);

360 361 362
static void __rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops)
{
	struct net_device *dev;
363 364
	LIST_HEAD(list_kill);

365
	for_each_netdev(net, dev) {
366 367
		if (dev->rtnl_link_ops == ops)
			ops->dellink(dev, &list_kill);
368
	}
369
	unregister_netdevice_many(&list_kill);
370 371
}

P
Patrick McHardy 已提交
372 373 374 375
/**
 * __rtnl_link_unregister - Unregister rtnl_link_ops from rtnetlink.
 * @ops: struct rtnl_link_ops * to unregister
 *
376
 * The caller must hold the rtnl_mutex.
P
Patrick McHardy 已提交
377 378 379
 */
void __rtnl_link_unregister(struct rtnl_link_ops *ops)
{
380
	struct net *net;
381

382
	for_each_net(net) {
383
		__rtnl_kill_links(net, ops);
384
	}
P
Patrick McHardy 已提交
385 386 387 388
	list_del(&ops->list);
}
EXPORT_SYMBOL_GPL(__rtnl_link_unregister);

389 390 391 392 393 394 395
/* Return with the rtnl_lock held when there are no network
 * devices unregistering in any network namespace.
 */
static void rtnl_lock_unregistering_all(void)
{
	struct net *net;
	bool unregistering;
396
	DEFINE_WAIT_FUNC(wait, woken_wake_function);
397

398
	add_wait_queue(&netdev_unregistering_wq, &wait);
399 400 401 402 403 404 405 406 407 408 409 410
	for (;;) {
		unregistering = false;
		rtnl_lock();
		for_each_net(net) {
			if (net->dev_unreg_count > 0) {
				unregistering = true;
				break;
			}
		}
		if (!unregistering)
			break;
		__rtnl_unlock();
411 412

		wait_woken(&wait, TASK_UNINTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT);
413
	}
414
	remove_wait_queue(&netdev_unregistering_wq, &wait);
415 416
}

P
Patrick McHardy 已提交
417 418 419 420 421 422
/**
 * rtnl_link_unregister - Unregister rtnl_link_ops from rtnetlink.
 * @ops: struct rtnl_link_ops * to unregister
 */
void rtnl_link_unregister(struct rtnl_link_ops *ops)
{
423 424 425
	/* Close the race with cleanup_net() */
	mutex_lock(&net_mutex);
	rtnl_lock_unregistering_all();
P
Patrick McHardy 已提交
426 427
	__rtnl_link_unregister(ops);
	rtnl_unlock();
428
	mutex_unlock(&net_mutex);
P
Patrick McHardy 已提交
429 430 431
}
EXPORT_SYMBOL_GPL(rtnl_link_unregister);

432 433 434 435 436 437 438 439 440
static size_t rtnl_link_get_slave_info_data_size(const struct net_device *dev)
{
	struct net_device *master_dev;
	const struct rtnl_link_ops *ops;

	master_dev = netdev_master_upper_dev_get((struct net_device *) dev);
	if (!master_dev)
		return 0;
	ops = master_dev->rtnl_link_ops;
441
	if (!ops || !ops->get_slave_size)
442 443 444 445 446 447
		return 0;
	/* IFLA_INFO_SLAVE_DATA + nested data */
	return nla_total_size(sizeof(struct nlattr)) +
	       ops->get_slave_size(master_dev, dev);
}

P
Patrick McHardy 已提交
448 449 450 451 452 453 454 455
static size_t rtnl_link_get_size(const struct net_device *dev)
{
	const struct rtnl_link_ops *ops = dev->rtnl_link_ops;
	size_t size;

	if (!ops)
		return 0;

456 457
	size = nla_total_size(sizeof(struct nlattr)) + /* IFLA_LINKINFO */
	       nla_total_size(strlen(ops->kind) + 1);  /* IFLA_INFO_KIND */
P
Patrick McHardy 已提交
458 459 460

	if (ops->get_size)
		/* IFLA_INFO_DATA + nested data */
461
		size += nla_total_size(sizeof(struct nlattr)) +
P
Patrick McHardy 已提交
462 463 464
			ops->get_size(dev);

	if (ops->get_xstats_size)
465 466
		/* IFLA_INFO_XSTATS */
		size += nla_total_size(ops->get_xstats_size(dev));
P
Patrick McHardy 已提交
467

468 469
	size += rtnl_link_get_slave_info_data_size(dev);

P
Patrick McHardy 已提交
470 471 472
	return size;
}

T
Thomas Graf 已提交
473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492
static LIST_HEAD(rtnl_af_ops);

static const struct rtnl_af_ops *rtnl_af_lookup(const int family)
{
	const struct rtnl_af_ops *ops;

	list_for_each_entry(ops, &rtnl_af_ops, list) {
		if (ops->family == family)
			return ops;
	}

	return NULL;
}

/**
 * rtnl_af_register - Register rtnl_af_ops with rtnetlink.
 * @ops: struct rtnl_af_ops * to register
 *
 * Returns 0 on success or a negative error code.
 */
493
void rtnl_af_register(struct rtnl_af_ops *ops)
T
Thomas Graf 已提交
494 495
{
	rtnl_lock();
496
	list_add_tail(&ops->list, &rtnl_af_ops);
T
Thomas Graf 已提交
497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524
	rtnl_unlock();
}
EXPORT_SYMBOL_GPL(rtnl_af_register);

/**
 * __rtnl_af_unregister - Unregister rtnl_af_ops from rtnetlink.
 * @ops: struct rtnl_af_ops * to unregister
 *
 * The caller must hold the rtnl_mutex.
 */
void __rtnl_af_unregister(struct rtnl_af_ops *ops)
{
	list_del(&ops->list);
}
EXPORT_SYMBOL_GPL(__rtnl_af_unregister);

/**
 * rtnl_af_unregister - Unregister rtnl_af_ops from rtnetlink.
 * @ops: struct rtnl_af_ops * to unregister
 */
void rtnl_af_unregister(struct rtnl_af_ops *ops)
{
	rtnl_lock();
	__rtnl_af_unregister(ops);
	rtnl_unlock();
}
EXPORT_SYMBOL_GPL(rtnl_af_unregister);

525 526
static size_t rtnl_link_get_af_size(const struct net_device *dev,
				    u32 ext_filter_mask)
T
Thomas Graf 已提交
527 528 529 530 531 532 533 534 535 536 537
{
	struct rtnl_af_ops *af_ops;
	size_t size;

	/* IFLA_AF_SPEC */
	size = nla_total_size(sizeof(struct nlattr));

	list_for_each_entry(af_ops, &rtnl_af_ops, list) {
		if (af_ops->get_link_af_size) {
			/* AF_* + nested data */
			size += nla_total_size(sizeof(struct nlattr)) +
538
				af_ops->get_link_af_size(dev, ext_filter_mask);
T
Thomas Graf 已提交
539 540 541 542 543 544
		}
	}

	return size;
}

545
static bool rtnl_have_link_slave_info(const struct net_device *dev)
P
Patrick McHardy 已提交
546
{
547
	struct net_device *master_dev;
P
Patrick McHardy 已提交
548

549
	master_dev = netdev_master_upper_dev_get((struct net_device *) dev);
550
	if (master_dev && master_dev->rtnl_link_ops)
551 552 553 554 555 556 557 558 559 560 561
		return true;
	return false;
}

static int rtnl_link_slave_info_fill(struct sk_buff *skb,
				     const struct net_device *dev)
{
	struct net_device *master_dev;
	const struct rtnl_link_ops *ops;
	struct nlattr *slave_data;
	int err;
P
Patrick McHardy 已提交
562

563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595
	master_dev = netdev_master_upper_dev_get((struct net_device *) dev);
	if (!master_dev)
		return 0;
	ops = master_dev->rtnl_link_ops;
	if (!ops)
		return 0;
	if (nla_put_string(skb, IFLA_INFO_SLAVE_KIND, ops->kind) < 0)
		return -EMSGSIZE;
	if (ops->fill_slave_info) {
		slave_data = nla_nest_start(skb, IFLA_INFO_SLAVE_DATA);
		if (!slave_data)
			return -EMSGSIZE;
		err = ops->fill_slave_info(skb, master_dev, dev);
		if (err < 0)
			goto err_cancel_slave_data;
		nla_nest_end(skb, slave_data);
	}
	return 0;

err_cancel_slave_data:
	nla_nest_cancel(skb, slave_data);
	return err;
}

static int rtnl_link_info_fill(struct sk_buff *skb,
			       const struct net_device *dev)
{
	const struct rtnl_link_ops *ops = dev->rtnl_link_ops;
	struct nlattr *data;
	int err;

	if (!ops)
		return 0;
P
Patrick McHardy 已提交
596
	if (nla_put_string(skb, IFLA_INFO_KIND, ops->kind) < 0)
597
		return -EMSGSIZE;
P
Patrick McHardy 已提交
598 599 600
	if (ops->fill_xstats) {
		err = ops->fill_xstats(skb, dev);
		if (err < 0)
601
			return err;
P
Patrick McHardy 已提交
602 603 604
	}
	if (ops->fill_info) {
		data = nla_nest_start(skb, IFLA_INFO_DATA);
605 606
		if (data == NULL)
			return -EMSGSIZE;
P
Patrick McHardy 已提交
607 608 609 610 611 612 613 614 615
		err = ops->fill_info(skb, dev);
		if (err < 0)
			goto err_cancel_data;
		nla_nest_end(skb, data);
	}
	return 0;

err_cancel_data:
	nla_nest_cancel(skb, data);
616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638
	return err;
}

static int rtnl_link_fill(struct sk_buff *skb, const struct net_device *dev)
{
	struct nlattr *linkinfo;
	int err = -EMSGSIZE;

	linkinfo = nla_nest_start(skb, IFLA_LINKINFO);
	if (linkinfo == NULL)
		goto out;

	err = rtnl_link_info_fill(skb, dev);
	if (err < 0)
		goto err_cancel_link;

	err = rtnl_link_slave_info_fill(skb, dev);
	if (err < 0)
		goto err_cancel_link;

	nla_nest_end(skb, linkinfo);
	return 0;

P
Patrick McHardy 已提交
639 640 641 642 643 644
err_cancel_link:
	nla_nest_cancel(skb, linkinfo);
out:
	return err;
}

645
int rtnetlink_send(struct sk_buff *skb, struct net *net, u32 pid, unsigned int group, int echo)
L
Linus Torvalds 已提交
646
{
647
	struct sock *rtnl = net->rtnl;
L
Linus Torvalds 已提交
648 649
	int err = 0;

650
	NETLINK_CB(skb).dst_group = group;
L
Linus Torvalds 已提交
651 652 653 654 655 656 657 658
	if (echo)
		atomic_inc(&skb->users);
	netlink_broadcast(rtnl, skb, pid, group, GFP_KERNEL);
	if (echo)
		err = netlink_unicast(rtnl, skb, pid, MSG_DONTWAIT);
	return err;
}

659
int rtnl_unicast(struct sk_buff *skb, struct net *net, u32 pid)
660
{
661 662
	struct sock *rtnl = net->rtnl;

663 664
	return nlmsg_unicast(rtnl, skb, pid);
}
E
Eric Dumazet 已提交
665
EXPORT_SYMBOL(rtnl_unicast);
666

667 668
void rtnl_notify(struct sk_buff *skb, struct net *net, u32 pid, u32 group,
		 struct nlmsghdr *nlh, gfp_t flags)
669
{
670
	struct sock *rtnl = net->rtnl;
671 672 673 674 675
	int report = 0;

	if (nlh)
		report = nlmsg_report(nlh);

676
	nlmsg_notify(rtnl, skb, pid, group, report, flags);
677
}
E
Eric Dumazet 已提交
678
EXPORT_SYMBOL(rtnl_notify);
679

680
void rtnl_set_sk_err(struct net *net, u32 group, int error)
681
{
682 683
	struct sock *rtnl = net->rtnl;

684 685
	netlink_set_err(rtnl, 0, group, error);
}
E
Eric Dumazet 已提交
686
EXPORT_SYMBOL(rtnl_set_sk_err);
687

L
Linus Torvalds 已提交
688 689
int rtnetlink_put_metrics(struct sk_buff *skb, u32 *metrics)
{
690 691 692 693 694 695 696 697 698
	struct nlattr *mx;
	int i, valid = 0;

	mx = nla_nest_start(skb, RTA_METRICS);
	if (mx == NULL)
		return -ENOBUFS;

	for (i = 0; i < RTAX_MAX; i++) {
		if (metrics[i]) {
699 700 701 702 703 704 705 706
			if (i == RTAX_CC_ALGO - 1) {
				char tmp[TCP_CA_NAME_MAX], *name;

				name = tcp_ca_get_name_by_key(metrics[i], tmp);
				if (!name)
					continue;
				if (nla_put_string(skb, i + 1, name))
					goto nla_put_failure;
707 708 709
			} else if (i == RTAX_FEATURES - 1) {
				u32 user_features = metrics[i] & RTAX_FEATURE_MASK;

710 711
				if (!user_features)
					continue;
712 713 714
				BUILD_BUG_ON(RTAX_FEATURE_MASK & DST_FEATURE_MASK);
				if (nla_put_u32(skb, i + 1, user_features))
					goto nla_put_failure;
715 716 717 718
			} else {
				if (nla_put_u32(skb, i + 1, metrics[i]))
					goto nla_put_failure;
			}
719 720
			valid++;
		}
L
Linus Torvalds 已提交
721 722
	}

723 724 725 726
	if (!valid) {
		nla_nest_cancel(skb, mx);
		return 0;
	}
727 728 729 730

	return nla_nest_end(skb, mx);

nla_put_failure:
731 732
	nla_nest_cancel(skb, mx);
	return -EMSGSIZE;
L
Linus Torvalds 已提交
733
}
E
Eric Dumazet 已提交
734
EXPORT_SYMBOL(rtnetlink_put_metrics);
L
Linus Torvalds 已提交
735

736
int rtnl_put_cacheinfo(struct sk_buff *skb, struct dst_entry *dst, u32 id,
737
		       long expires, u32 error)
738 739
{
	struct rta_cacheinfo ci = {
740
		.rta_lastuse = jiffies_delta_to_clock_t(jiffies - dst->lastuse),
741 742 743 744 745 746
		.rta_used = dst->__use,
		.rta_clntref = atomic_read(&(dst->__refcnt)),
		.rta_error = error,
		.rta_id =  id,
	};

747 748
	if (expires) {
		unsigned long clock;
749

750 751 752 753
		clock = jiffies_to_clock_t(abs(expires));
		clock = min_t(unsigned long, clock, INT_MAX);
		ci.rta_expires = (expires > 0) ? clock : -clock;
	}
754 755 756
	return nla_put(skb, RTA_CACHEINFO, sizeof(ci), &ci);
}
EXPORT_SYMBOL_GPL(rtnl_put_cacheinfo);
L
Linus Torvalds 已提交
757

758
static void set_operstate(struct net_device *dev, unsigned char transition)
S
Stefan Rompf 已提交
759 760 761
{
	unsigned char operstate = dev->operstate;

E
Eric Dumazet 已提交
762
	switch (transition) {
S
Stefan Rompf 已提交
763 764 765 766 767 768 769 770 771 772 773 774
	case IF_OPER_UP:
		if ((operstate == IF_OPER_DORMANT ||
		     operstate == IF_OPER_UNKNOWN) &&
		    !netif_dormant(dev))
			operstate = IF_OPER_UP;
		break;

	case IF_OPER_DORMANT:
		if (operstate == IF_OPER_UP ||
		    operstate == IF_OPER_UNKNOWN)
			operstate = IF_OPER_DORMANT;
		break;
775
	}
S
Stefan Rompf 已提交
776 777 778 779 780

	if (dev->operstate != operstate) {
		write_lock_bh(&dev_base_lock);
		dev->operstate = operstate;
		write_unlock_bh(&dev_base_lock);
781 782
		netdev_state_change(dev);
	}
S
Stefan Rompf 已提交
783 784
}

785 786 787 788 789 790
static unsigned int rtnl_dev_get_flags(const struct net_device *dev)
{
	return (dev->flags & ~(IFF_PROMISC | IFF_ALLMULTI)) |
	       (dev->gflags & (IFF_PROMISC | IFF_ALLMULTI));
}

791 792 793 794 795 796 797 798
static unsigned int rtnl_dev_combine_flags(const struct net_device *dev,
					   const struct ifinfomsg *ifm)
{
	unsigned int flags = ifm->ifi_flags;

	/* bugwards compatibility: ifi_change == 0 is treated as ~0 */
	if (ifm->ifi_change)
		flags = (flags & ifm->ifi_change) |
799
			(rtnl_dev_get_flags(dev) & ~ifm->ifi_change);
800 801 802 803

	return flags;
}

804
static void copy_rtnl_link_stats(struct rtnl_link_stats *a,
805
				 const struct rtnl_link_stats64 *b)
L
Linus Torvalds 已提交
806
{
807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833
	a->rx_packets = b->rx_packets;
	a->tx_packets = b->tx_packets;
	a->rx_bytes = b->rx_bytes;
	a->tx_bytes = b->tx_bytes;
	a->rx_errors = b->rx_errors;
	a->tx_errors = b->tx_errors;
	a->rx_dropped = b->rx_dropped;
	a->tx_dropped = b->tx_dropped;

	a->multicast = b->multicast;
	a->collisions = b->collisions;

	a->rx_length_errors = b->rx_length_errors;
	a->rx_over_errors = b->rx_over_errors;
	a->rx_crc_errors = b->rx_crc_errors;
	a->rx_frame_errors = b->rx_frame_errors;
	a->rx_fifo_errors = b->rx_fifo_errors;
	a->rx_missed_errors = b->rx_missed_errors;

	a->tx_aborted_errors = b->tx_aborted_errors;
	a->tx_carrier_errors = b->tx_carrier_errors;
	a->tx_fifo_errors = b->tx_fifo_errors;
	a->tx_heartbeat_errors = b->tx_heartbeat_errors;
	a->tx_window_errors = b->tx_window_errors;

	a->rx_compressed = b->rx_compressed;
	a->tx_compressed = b->tx_compressed;
834 835

	a->rx_nohandler = b->rx_nohandler;
836 837
}

838
/* All VF info */
839 840
static inline int rtnl_vfinfo_size(const struct net_device *dev,
				   u32 ext_filter_mask)
841
{
842
	if (dev->dev.parent && (ext_filter_mask & RTEXT_FILTER_VF)) {
843
		int num_vfs = dev_num_vf(dev->dev.parent);
S
Sabrina Dubroca 已提交
844
		size_t size = nla_total_size(0);
845
		size += num_vfs *
S
Sabrina Dubroca 已提交
846 847 848 849
			(nla_total_size(0) +
			 nla_total_size(sizeof(struct ifla_vf_mac)) +
			 nla_total_size(sizeof(struct ifla_vf_vlan)) +
			 nla_total_size(0) + /* nest IFLA_VF_VLAN_LIST */
850 851
			 nla_total_size(MAX_VLAN_LIST_LEN *
					sizeof(struct ifla_vf_vlan_info)) +
852
			 nla_total_size(sizeof(struct ifla_vf_spoofchk)) +
S
Sabrina Dubroca 已提交
853
			 nla_total_size(sizeof(struct ifla_vf_tx_rate)) +
J
Jiri Benc 已提交
854
			 nla_total_size(sizeof(struct ifla_vf_rate)) +
855
			 nla_total_size(sizeof(struct ifla_vf_link_state)) +
856
			 nla_total_size(sizeof(struct ifla_vf_rss_query_en)) +
S
Sabrina Dubroca 已提交
857
			 nla_total_size(0) + /* nest IFLA_VF_STATS */
858
			 /* IFLA_VF_STATS_RX_PACKETS */
N
Nicolas Dichtel 已提交
859
			 nla_total_size_64bit(sizeof(__u64)) +
860
			 /* IFLA_VF_STATS_TX_PACKETS */
N
Nicolas Dichtel 已提交
861
			 nla_total_size_64bit(sizeof(__u64)) +
862
			 /* IFLA_VF_STATS_RX_BYTES */
N
Nicolas Dichtel 已提交
863
			 nla_total_size_64bit(sizeof(__u64)) +
864
			 /* IFLA_VF_STATS_TX_BYTES */
N
Nicolas Dichtel 已提交
865
			 nla_total_size_64bit(sizeof(__u64)) +
866
			 /* IFLA_VF_STATS_BROADCAST */
N
Nicolas Dichtel 已提交
867
			 nla_total_size_64bit(sizeof(__u64)) +
868
			 /* IFLA_VF_STATS_MULTICAST */
N
Nicolas Dichtel 已提交
869
			 nla_total_size_64bit(sizeof(__u64)) +
H
Hiroshi Shimamoto 已提交
870
			 nla_total_size(sizeof(struct ifla_vf_trust)));
871 872
		return size;
	} else
873 874 875
		return 0;
}

876 877
static size_t rtnl_port_size(const struct net_device *dev,
			     u32 ext_filter_mask)
878 879 880 881 882 883 884 885 886 887 888 889 890
{
	size_t port_size = nla_total_size(4)		/* PORT_VF */
		+ nla_total_size(PORT_PROFILE_MAX)	/* PORT_PROFILE */
		+ nla_total_size(PORT_UUID_MAX)		/* PORT_INSTANCE_UUID */
		+ nla_total_size(PORT_UUID_MAX)		/* PORT_HOST_UUID */
		+ nla_total_size(1)			/* PROT_VDP_REQUEST */
		+ nla_total_size(2);			/* PORT_VDP_RESPONSE */
	size_t vf_ports_size = nla_total_size(sizeof(struct nlattr));
	size_t vf_port_size = nla_total_size(sizeof(struct nlattr))
		+ port_size;
	size_t port_self_size = nla_total_size(sizeof(struct nlattr))
		+ port_size;

891 892
	if (!dev->netdev_ops->ndo_get_vf_port || !dev->dev.parent ||
	    !(ext_filter_mask & RTEXT_FILTER_VF))
893 894 895 896 897 898 899 900
		return 0;
	if (dev_num_vf(dev->dev.parent))
		return port_self_size + vf_ports_size +
			vf_port_size * dev_num_vf(dev->dev.parent);
	else
		return port_self_size;
}

D
David S. Miller 已提交
901
static size_t rtnl_xdp_size(void)
902
{
903
	size_t xdp_size = nla_total_size(0) +	/* nest IFLA_XDP */
M
Martin KaFai Lau 已提交
904 905
			  nla_total_size(1) +	/* XDP_ATTACHED */
			  nla_total_size(4);	/* XDP_PROG_ID */
906

D
David S. Miller 已提交
907
	return xdp_size;
908 909
}

910 911
static noinline size_t if_nlmsg_size(const struct net_device *dev,
				     u32 ext_filter_mask)
912 913 914
{
	return NLMSG_ALIGN(sizeof(struct ifinfomsg))
	       + nla_total_size(IFNAMSIZ) /* IFLA_IFNAME */
915
	       + nla_total_size(IFALIASZ) /* IFLA_IFALIAS */
916
	       + nla_total_size(IFNAMSIZ) /* IFLA_QDISC */
917
	       + nla_total_size_64bit(sizeof(struct rtnl_link_ifmap))
918
	       + nla_total_size(sizeof(struct rtnl_link_stats))
919
	       + nla_total_size_64bit(sizeof(struct rtnl_link_stats64))
920 921 922 923 924 925 926
	       + nla_total_size(MAX_ADDR_LEN) /* IFLA_ADDRESS */
	       + nla_total_size(MAX_ADDR_LEN) /* IFLA_BROADCAST */
	       + nla_total_size(4) /* IFLA_TXQLEN */
	       + nla_total_size(4) /* IFLA_WEIGHT */
	       + nla_total_size(4) /* IFLA_MTU */
	       + nla_total_size(4) /* IFLA_LINK */
	       + nla_total_size(4) /* IFLA_MASTER */
927
	       + nla_total_size(1) /* IFLA_CARRIER */
928
	       + nla_total_size(4) /* IFLA_PROMISCUITY */
929 930
	       + nla_total_size(4) /* IFLA_NUM_TX_QUEUES */
	       + nla_total_size(4) /* IFLA_NUM_RX_QUEUES */
931 932
	       + nla_total_size(4) /* IFLA_GSO_MAX_SEGS */
	       + nla_total_size(4) /* IFLA_GSO_MAX_SIZE */
933
	       + nla_total_size(1) /* IFLA_OPERSTATE */
P
Patrick McHardy 已提交
934
	       + nla_total_size(1) /* IFLA_LINKMODE */
935
	       + nla_total_size(4) /* IFLA_CARRIER_CHANGES */
936
	       + nla_total_size(4) /* IFLA_LINK_NETNSID */
937
	       + nla_total_size(4) /* IFLA_GROUP */
938 939 940
	       + nla_total_size(ext_filter_mask
			        & RTEXT_FILTER_VF ? 4 : 0) /* IFLA_NUM_VF */
	       + rtnl_vfinfo_size(dev, ext_filter_mask) /* IFLA_VFINFO_LIST */
941
	       + rtnl_port_size(dev, ext_filter_mask) /* IFLA_VF_PORTS + IFLA_PORT_SELF */
T
Thomas Graf 已提交
942
	       + rtnl_link_get_size(dev) /* IFLA_LINKINFO */
943
	       + rtnl_link_get_af_size(dev, ext_filter_mask) /* IFLA_AF_SPEC */
944
	       + nla_total_size(MAX_PHYS_ITEM_ID_LEN) /* IFLA_PHYS_PORT_ID */
945
	       + nla_total_size(MAX_PHYS_ITEM_ID_LEN) /* IFLA_PHYS_SWITCH_ID */
946
	       + nla_total_size(IFNAMSIZ) /* IFLA_PHYS_PORT_NAME */
D
David S. Miller 已提交
947
	       + rtnl_xdp_size() /* IFLA_XDP */
948
	       + nla_total_size(4)  /* IFLA_EVENT */
949 950
	       + nla_total_size(1); /* IFLA_PROTO_DOWN */

951 952
}

953 954 955 956 957 958 959 960 961 962 963 964 965
static int rtnl_vf_ports_fill(struct sk_buff *skb, struct net_device *dev)
{
	struct nlattr *vf_ports;
	struct nlattr *vf_port;
	int vf;
	int err;

	vf_ports = nla_nest_start(skb, IFLA_VF_PORTS);
	if (!vf_ports)
		return -EMSGSIZE;

	for (vf = 0; vf < dev_num_vf(dev->dev.parent); vf++) {
		vf_port = nla_nest_start(skb, IFLA_VF_PORT);
966 967
		if (!vf_port)
			goto nla_put_failure;
968 969
		if (nla_put_u32(skb, IFLA_PORT_VF, vf))
			goto nla_put_failure;
970
		err = dev->netdev_ops->ndo_get_vf_port(dev, vf, skb);
971 972
		if (err == -EMSGSIZE)
			goto nla_put_failure;
973 974 975 976 977 978 979 980 981 982
		if (err) {
			nla_nest_cancel(skb, vf_port);
			continue;
		}
		nla_nest_end(skb, vf_port);
	}

	nla_nest_end(skb, vf_ports);

	return 0;
983 984 985 986

nla_put_failure:
	nla_nest_cancel(skb, vf_ports);
	return -EMSGSIZE;
987 988 989 990 991 992 993 994 995 996 997 998 999 1000
}

static int rtnl_port_self_fill(struct sk_buff *skb, struct net_device *dev)
{
	struct nlattr *port_self;
	int err;

	port_self = nla_nest_start(skb, IFLA_PORT_SELF);
	if (!port_self)
		return -EMSGSIZE;

	err = dev->netdev_ops->ndo_get_vf_port(dev, PORT_SELF_VF, skb);
	if (err) {
		nla_nest_cancel(skb, port_self);
1001
		return (err == -EMSGSIZE) ? err : 0;
1002 1003 1004 1005 1006 1007 1008
	}

	nla_nest_end(skb, port_self);

	return 0;
}

1009 1010
static int rtnl_port_fill(struct sk_buff *skb, struct net_device *dev,
			  u32 ext_filter_mask)
1011 1012 1013
{
	int err;

1014 1015
	if (!dev->netdev_ops->ndo_get_vf_port || !dev->dev.parent ||
	    !(ext_filter_mask & RTEXT_FILTER_VF))
1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030
		return 0;

	err = rtnl_port_self_fill(skb, dev);
	if (err)
		return err;

	if (dev_num_vf(dev->dev.parent)) {
		err = rtnl_vf_ports_fill(skb, dev);
		if (err)
			return err;
	}

	return 0;
}

1031 1032 1033
static int rtnl_phys_port_id_fill(struct sk_buff *skb, struct net_device *dev)
{
	int err;
1034
	struct netdev_phys_item_id ppid;
1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048

	err = dev_get_phys_port_id(dev, &ppid);
	if (err) {
		if (err == -EOPNOTSUPP)
			return 0;
		return err;
	}

	if (nla_put(skb, IFLA_PHYS_PORT_ID, ppid.id_len, ppid.id))
		return -EMSGSIZE;

	return 0;
}

1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060
static int rtnl_phys_port_name_fill(struct sk_buff *skb, struct net_device *dev)
{
	char name[IFNAMSIZ];
	int err;

	err = dev_get_phys_port_name(dev, name, sizeof(name));
	if (err) {
		if (err == -EOPNOTSUPP)
			return 0;
		return err;
	}

1061
	if (nla_put_string(skb, IFLA_PHYS_PORT_NAME, name))
1062 1063 1064 1065 1066
		return -EMSGSIZE;

	return 0;
}

1067 1068 1069
static int rtnl_phys_switch_id_fill(struct sk_buff *skb, struct net_device *dev)
{
	int err;
1070
	struct switchdev_attr attr = {
1071
		.orig_dev = dev,
1072
		.id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID,
1073 1074
		.flags = SWITCHDEV_F_NO_RECURSE,
	};
1075

1076
	err = switchdev_port_attr_get(dev, &attr);
1077 1078 1079 1080 1081 1082
	if (err) {
		if (err == -EOPNOTSUPP)
			return 0;
		return err;
	}

1083 1084
	if (nla_put(skb, IFLA_PHYS_SWITCH_ID, attr.u.ppid.id_len,
		    attr.u.ppid.id))
1085 1086 1087 1088 1089
		return -EMSGSIZE;

	return 0;
}

1090 1091 1092
static noinline_for_stack int rtnl_fill_stats(struct sk_buff *skb,
					      struct net_device *dev)
{
1093
	struct rtnl_link_stats64 *sp;
1094
	struct nlattr *attr;
1095

1096 1097
	attr = nla_reserve_64bit(skb, IFLA_STATS64,
				 sizeof(struct rtnl_link_stats64), IFLA_PAD);
1098 1099 1100
	if (!attr)
		return -EMSGSIZE;

1101 1102
	sp = nla_data(attr);
	dev_get_stats(dev, sp);
1103

1104 1105
	attr = nla_reserve(skb, IFLA_STATS,
			   sizeof(struct rtnl_link_stats));
1106 1107 1108
	if (!attr)
		return -EMSGSIZE;

1109
	copy_rtnl_link_stats(nla_data(attr), sp);
1110 1111 1112 1113 1114 1115 1116 1117 1118 1119

	return 0;
}

static noinline_for_stack int rtnl_fill_vfinfo(struct sk_buff *skb,
					       struct net_device *dev,
					       int vfs_num,
					       struct nlattr *vfinfo)
{
	struct ifla_vf_rss_query_en vf_rss_query_en;
1120
	struct nlattr *vf, *vfstats, *vfvlanlist;
1121
	struct ifla_vf_link_state vf_linkstate;
1122
	struct ifla_vf_vlan_info vf_vlan_info;
1123 1124 1125 1126 1127 1128 1129 1130 1131
	struct ifla_vf_spoofchk vf_spoofchk;
	struct ifla_vf_tx_rate vf_tx_rate;
	struct ifla_vf_stats vf_stats;
	struct ifla_vf_trust vf_trust;
	struct ifla_vf_vlan vf_vlan;
	struct ifla_vf_rate vf_rate;
	struct ifla_vf_mac vf_mac;
	struct ifla_vf_info ivi;

1132 1133
	memset(&ivi, 0, sizeof(ivi));

1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145
	/* Not all SR-IOV capable drivers support the
	 * spoofcheck and "RSS query enable" query.  Preset to
	 * -1 so the user space tool can detect that the driver
	 * didn't report anything.
	 */
	ivi.spoofchk = -1;
	ivi.rss_query_en = -1;
	ivi.trusted = -1;
	/* The default value for VF link state is "auto"
	 * IFLA_VF_LINK_STATE_AUTO which equals zero
	 */
	ivi.linkstate = 0;
1146 1147
	/* VLAN Protocol by default is 802.1Q */
	ivi.vlan_proto = htons(ETH_P_8021Q);
1148 1149 1150
	if (dev->netdev_ops->ndo_get_vf_config(dev, vfs_num, &ivi))
		return 0;

1151 1152
	memset(&vf_vlan_info, 0, sizeof(vf_vlan_info));

1153 1154
	vf_mac.vf =
		vf_vlan.vf =
1155
		vf_vlan_info.vf =
1156 1157 1158 1159 1160 1161 1162 1163 1164 1165
		vf_rate.vf =
		vf_tx_rate.vf =
		vf_spoofchk.vf =
		vf_linkstate.vf =
		vf_rss_query_en.vf =
		vf_trust.vf = ivi.vf;

	memcpy(vf_mac.mac, ivi.mac, sizeof(ivi.mac));
	vf_vlan.vlan = ivi.vlan;
	vf_vlan.qos = ivi.qos;
1166 1167 1168
	vf_vlan_info.vlan = ivi.vlan;
	vf_vlan_info.qos = ivi.qos;
	vf_vlan_info.vlan_proto = ivi.vlan_proto;
1169 1170 1171 1172 1173 1174 1175 1176
	vf_tx_rate.rate = ivi.max_tx_rate;
	vf_rate.min_tx_rate = ivi.min_tx_rate;
	vf_rate.max_tx_rate = ivi.max_tx_rate;
	vf_spoofchk.setting = ivi.spoofchk;
	vf_linkstate.link_state = ivi.linkstate;
	vf_rss_query_en.setting = ivi.rss_query_en;
	vf_trust.setting = ivi.trusted;
	vf = nla_nest_start(skb, IFLA_VF_INFO);
1177 1178
	if (!vf)
		goto nla_put_vfinfo_failure;
1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193
	if (nla_put(skb, IFLA_VF_MAC, sizeof(vf_mac), &vf_mac) ||
	    nla_put(skb, IFLA_VF_VLAN, sizeof(vf_vlan), &vf_vlan) ||
	    nla_put(skb, IFLA_VF_RATE, sizeof(vf_rate),
		    &vf_rate) ||
	    nla_put(skb, IFLA_VF_TX_RATE, sizeof(vf_tx_rate),
		    &vf_tx_rate) ||
	    nla_put(skb, IFLA_VF_SPOOFCHK, sizeof(vf_spoofchk),
		    &vf_spoofchk) ||
	    nla_put(skb, IFLA_VF_LINK_STATE, sizeof(vf_linkstate),
		    &vf_linkstate) ||
	    nla_put(skb, IFLA_VF_RSS_QUERY_EN,
		    sizeof(vf_rss_query_en),
		    &vf_rss_query_en) ||
	    nla_put(skb, IFLA_VF_TRUST,
		    sizeof(vf_trust), &vf_trust))
1194 1195 1196 1197 1198 1199 1200 1201 1202 1203
		goto nla_put_vf_failure;
	vfvlanlist = nla_nest_start(skb, IFLA_VF_VLAN_LIST);
	if (!vfvlanlist)
		goto nla_put_vf_failure;
	if (nla_put(skb, IFLA_VF_VLAN_INFO, sizeof(vf_vlan_info),
		    &vf_vlan_info)) {
		nla_nest_cancel(skb, vfvlanlist);
		goto nla_put_vf_failure;
	}
	nla_nest_end(skb, vfvlanlist);
1204 1205 1206 1207 1208
	memset(&vf_stats, 0, sizeof(vf_stats));
	if (dev->netdev_ops->ndo_get_vf_stats)
		dev->netdev_ops->ndo_get_vf_stats(dev, vfs_num,
						&vf_stats);
	vfstats = nla_nest_start(skb, IFLA_VF_STATS);
1209 1210
	if (!vfstats)
		goto nla_put_vf_failure;
N
Nicolas Dichtel 已提交
1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221
	if (nla_put_u64_64bit(skb, IFLA_VF_STATS_RX_PACKETS,
			      vf_stats.rx_packets, IFLA_VF_STATS_PAD) ||
	    nla_put_u64_64bit(skb, IFLA_VF_STATS_TX_PACKETS,
			      vf_stats.tx_packets, IFLA_VF_STATS_PAD) ||
	    nla_put_u64_64bit(skb, IFLA_VF_STATS_RX_BYTES,
			      vf_stats.rx_bytes, IFLA_VF_STATS_PAD) ||
	    nla_put_u64_64bit(skb, IFLA_VF_STATS_TX_BYTES,
			      vf_stats.tx_bytes, IFLA_VF_STATS_PAD) ||
	    nla_put_u64_64bit(skb, IFLA_VF_STATS_BROADCAST,
			      vf_stats.broadcast, IFLA_VF_STATS_PAD) ||
	    nla_put_u64_64bit(skb, IFLA_VF_STATS_MULTICAST,
1222 1223 1224 1225
			      vf_stats.multicast, IFLA_VF_STATS_PAD)) {
		nla_nest_cancel(skb, vfstats);
		goto nla_put_vf_failure;
	}
1226 1227 1228
	nla_nest_end(skb, vfstats);
	nla_nest_end(skb, vf);
	return 0;
1229 1230 1231 1232 1233 1234

nla_put_vf_failure:
	nla_nest_cancel(skb, vf);
nla_put_vfinfo_failure:
	nla_nest_cancel(skb, vfinfo);
	return -EMSGSIZE;
1235 1236 1237 1238
}

static int rtnl_fill_link_ifmap(struct sk_buff *skb, struct net_device *dev)
{
K
Kangjie Lu 已提交
1239 1240 1241 1242 1243 1244 1245 1246 1247 1248
	struct rtnl_link_ifmap map;

	memset(&map, 0, sizeof(map));
	map.mem_start   = dev->mem_start;
	map.mem_end     = dev->mem_end;
	map.base_addr   = dev->base_addr;
	map.irq         = dev->irq;
	map.dma         = dev->dma;
	map.port        = dev->if_port;

1249
	if (nla_put_64bit(skb, IFLA_MAP, sizeof(map), &map, IFLA_PAD))
1250 1251 1252 1253 1254
		return -EMSGSIZE;

	return 0;
}

M
Martin KaFai Lau 已提交
1255
static u8 rtnl_xdp_attached_mode(struct net_device *dev, u32 *prog_id)
1256 1257
{
	const struct net_device_ops *ops = dev->netdev_ops;
M
Martin KaFai Lau 已提交
1258
	const struct bpf_prog *generic_xdp_prog;
1259 1260 1261

	ASSERT_RTNL();

M
Martin KaFai Lau 已提交
1262 1263 1264 1265
	*prog_id = 0;
	generic_xdp_prog = rtnl_dereference(dev->xdp_prog);
	if (generic_xdp_prog) {
		*prog_id = generic_xdp_prog->aux->id;
1266
		return XDP_ATTACHED_SKB;
M
Martin KaFai Lau 已提交
1267
	}
1268 1269
	if (!ops->ndo_xdp)
		return XDP_ATTACHED_NONE;
1270

1271
	return __dev_xdp_attached(dev, ops->ndo_xdp, prog_id);
1272 1273
}

1274 1275 1276
static int rtnl_xdp_fill(struct sk_buff *skb, struct net_device *dev)
{
	struct nlattr *xdp;
M
Martin KaFai Lau 已提交
1277
	u32 prog_id;
1278 1279 1280 1281 1282
	int err;

	xdp = nla_nest_start(skb, IFLA_XDP);
	if (!xdp)
		return -EMSGSIZE;
1283 1284

	err = nla_put_u8(skb, IFLA_XDP_ATTACHED,
M
Martin KaFai Lau 已提交
1285
			 rtnl_xdp_attached_mode(dev, &prog_id));
1286 1287 1288
	if (err)
		goto err_cancel;

M
Martin KaFai Lau 已提交
1289 1290 1291 1292 1293 1294
	if (prog_id) {
		err = nla_put_u32(skb, IFLA_XDP_PROG_ID, prog_id);
		if (err)
			goto err_cancel;
	}

1295 1296 1297 1298 1299 1300 1301 1302
	nla_nest_end(skb, xdp);
	return 0;

err_cancel:
	nla_nest_cancel(skb, xdp);
	return err;
}

1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332
static u32 rtnl_get_event(unsigned long event)
{
	u32 rtnl_event_type = IFLA_EVENT_NONE;

	switch (event) {
	case NETDEV_REBOOT:
		rtnl_event_type = IFLA_EVENT_REBOOT;
		break;
	case NETDEV_FEAT_CHANGE:
		rtnl_event_type = IFLA_EVENT_FEATURES;
		break;
	case NETDEV_BONDING_FAILOVER:
		rtnl_event_type = IFLA_EVENT_BONDING_FAILOVER;
		break;
	case NETDEV_NOTIFY_PEERS:
		rtnl_event_type = IFLA_EVENT_NOTIFY_PEERS;
		break;
	case NETDEV_RESEND_IGMP:
		rtnl_event_type = IFLA_EVENT_IGMP_RESEND;
		break;
	case NETDEV_CHANGEINFODATA:
		rtnl_event_type = IFLA_EVENT_BONDING_OPTIONS;
		break;
	default:
		break;
	}

	return rtnl_event_type;
}

1333
static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
1334
			    int type, u32 pid, u32 seq, u32 change,
1335 1336
			    unsigned int flags, u32 ext_filter_mask,
			    u32 event)
1337 1338 1339
{
	struct ifinfomsg *ifm;
	struct nlmsghdr *nlh;
1340
	struct nlattr *af_spec;
T
Thomas Graf 已提交
1341
	struct rtnl_af_ops *af_ops;
1342
	struct net_device *upper_dev = netdev_master_upper_dev_get(dev);
L
Linus Torvalds 已提交
1343

1344
	ASSERT_RTNL();
1345 1346
	nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ifm), flags);
	if (nlh == NULL)
1347
		return -EMSGSIZE;
L
Linus Torvalds 已提交
1348

1349 1350 1351 1352 1353 1354 1355 1356
	ifm = nlmsg_data(nlh);
	ifm->ifi_family = AF_UNSPEC;
	ifm->__ifi_pad = 0;
	ifm->ifi_type = dev->type;
	ifm->ifi_index = dev->ifindex;
	ifm->ifi_flags = dev_get_flags(dev);
	ifm->ifi_change = change;

1357 1358 1359 1360 1361 1362 1363
	if (nla_put_string(skb, IFLA_IFNAME, dev->name) ||
	    nla_put_u32(skb, IFLA_TXQLEN, dev->tx_queue_len) ||
	    nla_put_u8(skb, IFLA_OPERSTATE,
		       netif_running(dev) ? dev->operstate : IF_OPER_DOWN) ||
	    nla_put_u8(skb, IFLA_LINKMODE, dev->link_mode) ||
	    nla_put_u32(skb, IFLA_MTU, dev->mtu) ||
	    nla_put_u32(skb, IFLA_GROUP, dev->group) ||
1364
	    nla_put_u32(skb, IFLA_PROMISCUITY, dev->promiscuity) ||
1365
	    nla_put_u32(skb, IFLA_NUM_TX_QUEUES, dev->num_tx_queues) ||
1366 1367
	    nla_put_u32(skb, IFLA_GSO_MAX_SEGS, dev->gso_max_segs) ||
	    nla_put_u32(skb, IFLA_GSO_MAX_SIZE, dev->gso_max_size) ||
1368
#ifdef CONFIG_RPS
1369
	    nla_put_u32(skb, IFLA_NUM_RX_QUEUES, dev->num_rx_queues) ||
1370
#endif
1371 1372
	    (dev->ifindex != dev_get_iflink(dev) &&
	     nla_put_u32(skb, IFLA_LINK, dev_get_iflink(dev))) ||
1373 1374
	    (upper_dev &&
	     nla_put_u32(skb, IFLA_MASTER, upper_dev->ifindex)) ||
1375
	    nla_put_u8(skb, IFLA_CARRIER, netif_carrier_ok(dev)) ||
1376 1377 1378
	    (dev->qdisc &&
	     nla_put_string(skb, IFLA_QDISC, dev->qdisc->ops->id)) ||
	    (dev->ifalias &&
1379 1380
	     nla_put_string(skb, IFLA_IFALIAS, dev->ifalias)) ||
	    nla_put_u32(skb, IFLA_CARRIER_CHANGES,
1381 1382
			atomic_read(&dev->carrier_changes)) ||
	    nla_put_u8(skb, IFLA_PROTO_DOWN, dev->proto_down))
1383
		goto nla_put_failure;
1384

1385 1386 1387 1388 1389
	if (event != IFLA_EVENT_NONE) {
		if (nla_put_u32(skb, IFLA_EVENT, event))
			goto nla_put_failure;
	}

1390 1391
	if (rtnl_fill_link_ifmap(skb, dev))
		goto nla_put_failure;
L
Linus Torvalds 已提交
1392 1393

	if (dev->addr_len) {
1394 1395 1396
		if (nla_put(skb, IFLA_ADDRESS, dev->addr_len, dev->dev_addr) ||
		    nla_put(skb, IFLA_BROADCAST, dev->addr_len, dev->broadcast))
			goto nla_put_failure;
L
Linus Torvalds 已提交
1397 1398
	}

1399 1400 1401
	if (rtnl_phys_port_id_fill(skb, dev))
		goto nla_put_failure;

1402 1403 1404
	if (rtnl_phys_port_name_fill(skb, dev))
		goto nla_put_failure;

1405 1406 1407
	if (rtnl_phys_switch_id_fill(skb, dev))
		goto nla_put_failure;

1408
	if (rtnl_fill_stats(skb, dev))
1409 1410
		goto nla_put_failure;

1411 1412 1413
	if (dev->dev.parent && (ext_filter_mask & RTEXT_FILTER_VF) &&
	    nla_put_u32(skb, IFLA_NUM_VF, dev_num_vf(dev->dev.parent)))
		goto nla_put_failure;
1414

1415 1416
	if (dev->netdev_ops->ndo_get_vf_config && dev->dev.parent &&
	    ext_filter_mask & RTEXT_FILTER_VF) {
1417
		int i;
1418
		struct nlattr *vfinfo;
1419 1420 1421 1422 1423 1424
		int num_vfs = dev_num_vf(dev->dev.parent);

		vfinfo = nla_nest_start(skb, IFLA_VFINFO_LIST);
		if (!vfinfo)
			goto nla_put_failure;
		for (i = 0; i < num_vfs; i++) {
1425
			if (rtnl_fill_vfinfo(skb, dev, i, vfinfo))
1426
				goto nla_put_failure;
1427
		}
1428

1429
		nla_nest_end(skb, vfinfo);
1430
	}
1431

1432
	if (rtnl_port_fill(skb, dev, ext_filter_mask))
1433 1434
		goto nla_put_failure;

1435 1436 1437
	if (rtnl_xdp_fill(skb, dev))
		goto nla_put_failure;

1438
	if (dev->rtnl_link_ops || rtnl_have_link_slave_info(dev)) {
P
Patrick McHardy 已提交
1439 1440 1441 1442
		if (rtnl_link_fill(skb, dev) < 0)
			goto nla_put_failure;
	}

1443 1444 1445 1446 1447
	if (dev->rtnl_link_ops &&
	    dev->rtnl_link_ops->get_link_net) {
		struct net *link_net = dev->rtnl_link_ops->get_link_net(dev);

		if (!net_eq(dev_net(dev), link_net)) {
1448
			int id = peernet2id_alloc(dev_net(dev), link_net);
1449 1450 1451 1452 1453 1454

			if (nla_put_s32(skb, IFLA_LINK_NETNSID, id))
				goto nla_put_failure;
		}
	}

T
Thomas Graf 已提交
1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465
	if (!(af_spec = nla_nest_start(skb, IFLA_AF_SPEC)))
		goto nla_put_failure;

	list_for_each_entry(af_ops, &rtnl_af_ops, list) {
		if (af_ops->fill_link_af) {
			struct nlattr *af;
			int err;

			if (!(af = nla_nest_start(skb, af_ops->family)))
				goto nla_put_failure;

1466
			err = af_ops->fill_link_af(skb, dev, ext_filter_mask);
T
Thomas Graf 已提交
1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484

			/*
			 * Caller may return ENODATA to indicate that there
			 * was no data to be dumped. This is not an error, it
			 * means we should trim the attribute header and
			 * continue.
			 */
			if (err == -ENODATA)
				nla_nest_cancel(skb, af);
			else if (err < 0)
				goto nla_put_failure;

			nla_nest_end(skb, af);
		}
	}

	nla_nest_end(skb, af_spec);

1485 1486
	nlmsg_end(skb, nlh);
	return 0;
1487 1488

nla_put_failure:
1489 1490
	nlmsg_cancel(skb, nlh);
	return -EMSGSIZE;
L
Linus Torvalds 已提交
1491 1492
}

J
Jiri Pirko 已提交
1493
static const struct nla_policy ifla_policy[IFLA_MAX+1] = {
1494
	[IFLA_IFNAME]		= { .type = NLA_STRING, .len = IFNAMSIZ-1 },
P
Patrick McHardy 已提交
1495 1496
	[IFLA_ADDRESS]		= { .type = NLA_BINARY, .len = MAX_ADDR_LEN },
	[IFLA_BROADCAST]	= { .type = NLA_BINARY, .len = MAX_ADDR_LEN },
1497
	[IFLA_MAP]		= { .len = sizeof(struct rtnl_link_ifmap) },
1498
	[IFLA_MTU]		= { .type = NLA_U32 },
1499
	[IFLA_LINK]		= { .type = NLA_U32 },
1500
	[IFLA_MASTER]		= { .type = NLA_U32 },
1501
	[IFLA_CARRIER]		= { .type = NLA_U8 },
1502 1503 1504 1505
	[IFLA_TXQLEN]		= { .type = NLA_U32 },
	[IFLA_WEIGHT]		= { .type = NLA_U32 },
	[IFLA_OPERSTATE]	= { .type = NLA_U8 },
	[IFLA_LINKMODE]		= { .type = NLA_U8 },
1506
	[IFLA_LINKINFO]		= { .type = NLA_NESTED },
1507
	[IFLA_NET_NS_PID]	= { .type = NLA_U32 },
1508
	[IFLA_NET_NS_FD]	= { .type = NLA_U32 },
1509
	[IFLA_IFALIAS]	        = { .type = NLA_STRING, .len = IFALIASZ-1 },
1510
	[IFLA_VFINFO_LIST]	= {. type = NLA_NESTED },
1511 1512
	[IFLA_VF_PORTS]		= { .type = NLA_NESTED },
	[IFLA_PORT_SELF]	= { .type = NLA_NESTED },
T
Thomas Graf 已提交
1513
	[IFLA_AF_SPEC]		= { .type = NLA_NESTED },
1514
	[IFLA_EXT_MASK]		= { .type = NLA_U32 },
1515
	[IFLA_PROMISCUITY]	= { .type = NLA_U32 },
1516 1517
	[IFLA_NUM_TX_QUEUES]	= { .type = NLA_U32 },
	[IFLA_NUM_RX_QUEUES]	= { .type = NLA_U32 },
1518
	[IFLA_PHYS_PORT_ID]	= { .type = NLA_BINARY, .len = MAX_PHYS_ITEM_ID_LEN },
1519
	[IFLA_CARRIER_CHANGES]	= { .type = NLA_U32 },  /* ignored */
1520
	[IFLA_PHYS_SWITCH_ID]	= { .type = NLA_BINARY, .len = MAX_PHYS_ITEM_ID_LEN },
1521
	[IFLA_LINK_NETNSID]	= { .type = NLA_S32 },
1522
	[IFLA_PROTO_DOWN]	= { .type = NLA_U8 },
1523
	[IFLA_XDP]		= { .type = NLA_NESTED },
1524
	[IFLA_EVENT]		= { .type = NLA_U32 },
1525
	[IFLA_GROUP]		= { .type = NLA_U32 },
1526 1527
};

P
Patrick McHardy 已提交
1528 1529 1530
static const struct nla_policy ifla_info_policy[IFLA_INFO_MAX+1] = {
	[IFLA_INFO_KIND]	= { .type = NLA_STRING },
	[IFLA_INFO_DATA]	= { .type = NLA_NESTED },
1531 1532
	[IFLA_INFO_SLAVE_KIND]	= { .type = NLA_STRING },
	[IFLA_INFO_SLAVE_DATA]	= { .type = NLA_NESTED },
P
Patrick McHardy 已提交
1533 1534
};

1535
static const struct nla_policy ifla_vf_policy[IFLA_VF_MAX+1] = {
1536 1537
	[IFLA_VF_MAC]		= { .len = sizeof(struct ifla_vf_mac) },
	[IFLA_VF_VLAN]		= { .len = sizeof(struct ifla_vf_vlan) },
1538
	[IFLA_VF_VLAN_LIST]     = { .type = NLA_NESTED },
1539 1540 1541 1542
	[IFLA_VF_TX_RATE]	= { .len = sizeof(struct ifla_vf_tx_rate) },
	[IFLA_VF_SPOOFCHK]	= { .len = sizeof(struct ifla_vf_spoofchk) },
	[IFLA_VF_RATE]		= { .len = sizeof(struct ifla_vf_rate) },
	[IFLA_VF_LINK_STATE]	= { .len = sizeof(struct ifla_vf_link_state) },
1543
	[IFLA_VF_RSS_QUERY_EN]	= { .len = sizeof(struct ifla_vf_rss_query_en) },
1544
	[IFLA_VF_STATS]		= { .type = NLA_NESTED },
H
Hiroshi Shimamoto 已提交
1545
	[IFLA_VF_TRUST]		= { .len = sizeof(struct ifla_vf_trust) },
1546 1547
	[IFLA_VF_IB_NODE_GUID]	= { .len = sizeof(struct ifla_vf_guid) },
	[IFLA_VF_IB_PORT_GUID]	= { .len = sizeof(struct ifla_vf_guid) },
1548 1549
};

1550 1551 1552 1553 1554 1555 1556 1557 1558 1559
static const struct nla_policy ifla_port_policy[IFLA_PORT_MAX+1] = {
	[IFLA_PORT_VF]		= { .type = NLA_U32 },
	[IFLA_PORT_PROFILE]	= { .type = NLA_STRING,
				    .len = PORT_PROFILE_MAX },
	[IFLA_PORT_INSTANCE_UUID] = { .type = NLA_BINARY,
				      .len = PORT_UUID_MAX },
	[IFLA_PORT_HOST_UUID]	= { .type = NLA_STRING,
				    .len = PORT_UUID_MAX },
	[IFLA_PORT_REQUEST]	= { .type = NLA_U8, },
	[IFLA_PORT_RESPONSE]	= { .type = NLA_U16, },
1560 1561 1562 1563 1564 1565 1566

	/* Unused, but we need to keep it here since user space could
	 * fill it. It's also broken with regard to NLA_BINARY use in
	 * combination with structs.
	 */
	[IFLA_PORT_VSI_TYPE]	= { .type = NLA_BINARY,
				    .len = sizeof(struct ifla_port_vsi) },
1567 1568
};

1569 1570 1571
static const struct nla_policy ifla_xdp_policy[IFLA_XDP_MAX + 1] = {
	[IFLA_XDP_FD]		= { .type = NLA_S32 },
	[IFLA_XDP_ATTACHED]	= { .type = NLA_U8 },
1572
	[IFLA_XDP_FLAGS]	= { .type = NLA_U32 },
M
Martin KaFai Lau 已提交
1573
	[IFLA_XDP_PROG_ID]	= { .type = NLA_U32 },
1574 1575
};

1576 1577 1578 1579 1580
static const struct rtnl_link_ops *linkinfo_to_kind_ops(const struct nlattr *nla)
{
	const struct rtnl_link_ops *ops = NULL;
	struct nlattr *linfo[IFLA_INFO_MAX + 1];

1581 1582
	if (nla_parse_nested(linfo, IFLA_INFO_MAX, nla,
			     ifla_info_policy, NULL) < 0)
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 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628
		return NULL;

	if (linfo[IFLA_INFO_KIND]) {
		char kind[MODULE_NAME_LEN];

		nla_strlcpy(kind, linfo[IFLA_INFO_KIND], sizeof(kind));
		ops = rtnl_link_ops_get(kind);
	}

	return ops;
}

static bool link_master_filtered(struct net_device *dev, int master_idx)
{
	struct net_device *master;

	if (!master_idx)
		return false;

	master = netdev_master_upper_dev_get(dev);
	if (!master || master->ifindex != master_idx)
		return true;

	return false;
}

static bool link_kind_filtered(const struct net_device *dev,
			       const struct rtnl_link_ops *kind_ops)
{
	if (kind_ops && dev->rtnl_link_ops != kind_ops)
		return true;

	return false;
}

static bool link_dump_filtered(struct net_device *dev,
			       int master_idx,
			       const struct rtnl_link_ops *kind_ops)
{
	if (link_master_filtered(dev, master_idx) ||
	    link_kind_filtered(dev, kind_ops))
		return true;

	return false;
}

J
Jiri Pirko 已提交
1629 1630 1631 1632 1633 1634 1635 1636 1637
static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
{
	struct net *net = sock_net(skb->sk);
	int h, s_h;
	int idx = 0, s_idx;
	struct net_device *dev;
	struct hlist_head *head;
	struct nlattr *tb[IFLA_MAX+1];
	u32 ext_filter_mask = 0;
1638 1639 1640
	const struct rtnl_link_ops *kind_ops = NULL;
	unsigned int flags = NLM_F_MULTI;
	int master_idx = 0;
1641
	int err;
1642
	int hdrlen;
J
Jiri Pirko 已提交
1643 1644 1645 1646 1647 1648

	s_h = cb->args[0];
	s_idx = cb->args[1];

	cb->seq = net->dev_base_seq;

1649 1650 1651 1652 1653 1654 1655 1656 1657 1658
	/* A hack to preserve kernel<->userspace interface.
	 * The correct header is ifinfomsg. It is consistent with rtnl_getlink.
	 * However, before Linux v3.9 the code here assumed rtgenmsg and that's
	 * what iproute2 < v3.9.0 used.
	 * We can detect the old iproute2. Even including the IFLA_EXT_MASK
	 * attribute, its netlink message is shorter than struct ifinfomsg.
	 */
	hdrlen = nlmsg_len(cb->nlh) < sizeof(struct ifinfomsg) ?
		 sizeof(struct rtgenmsg) : sizeof(struct ifinfomsg);

1659 1660
	if (nlmsg_parse(cb->nlh, hdrlen, tb, IFLA_MAX,
			ifla_policy, NULL) >= 0) {
J
Jiri Pirko 已提交
1661 1662
		if (tb[IFLA_EXT_MASK])
			ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);
1663 1664 1665 1666 1667 1668 1669 1670 1671

		if (tb[IFLA_MASTER])
			master_idx = nla_get_u32(tb[IFLA_MASTER]);

		if (tb[IFLA_LINKINFO])
			kind_ops = linkinfo_to_kind_ops(tb[IFLA_LINKINFO]);

		if (master_idx || kind_ops)
			flags |= NLM_F_DUMP_FILTERED;
J
Jiri Pirko 已提交
1672 1673 1674 1675 1676
	}

	for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) {
		idx = 0;
		head = &net->dev_index_head[h];
1677
		hlist_for_each_entry(dev, head, index_hlist) {
1678
			if (link_dump_filtered(dev, master_idx, kind_ops))
1679
				goto cont;
J
Jiri Pirko 已提交
1680 1681
			if (idx < s_idx)
				goto cont;
1682 1683 1684
			err = rtnl_fill_ifinfo(skb, dev, RTM_NEWLINK,
					       NETLINK_CB(cb->skb).portid,
					       cb->nlh->nlmsg_seq, 0,
1685
					       flags,
1686
					       ext_filter_mask, 0);
1687

1688 1689 1690 1691 1692 1693
			if (err < 0) {
				if (likely(skb->len))
					goto out;

				goto out_err;
			}
J
Jiri Pirko 已提交
1694 1695 1696 1697 1698 1699 1700

			nl_dump_check_consistent(cb, nlmsg_hdr(skb));
cont:
			idx++;
		}
	}
out:
1701 1702
	err = skb->len;
out_err:
J
Jiri Pirko 已提交
1703 1704 1705
	cb->args[1] = idx;
	cb->args[0] = h;

1706
	return err;
J
Jiri Pirko 已提交
1707 1708
}

1709 1710
int rtnl_nla_parse_ifla(struct nlattr **tb, const struct nlattr *head, int len,
			struct netlink_ext_ack *exterr)
J
Jiri Pirko 已提交
1711
{
1712
	return nla_parse(tb, IFLA_MAX, head, len, ifla_policy, exterr);
J
Jiri Pirko 已提交
1713 1714 1715
}
EXPORT_SYMBOL(rtnl_nla_parse_ifla);

1716 1717 1718 1719 1720 1721 1722 1723
struct net *rtnl_link_get_net(struct net *src_net, struct nlattr *tb[])
{
	struct net *net;
	/* Examine the link attributes and figure out which
	 * network namespace we are talking about.
	 */
	if (tb[IFLA_NET_NS_PID])
		net = get_net_ns_by_pid(nla_get_u32(tb[IFLA_NET_NS_PID]));
1724 1725
	else if (tb[IFLA_NET_NS_FD])
		net = get_net_ns_by_fd(nla_get_u32(tb[IFLA_NET_NS_FD]));
1726 1727 1728 1729 1730 1731
	else
		net = get_net(src_net);
	return net;
}
EXPORT_SYMBOL(rtnl_link_get_net);

1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743
static int validate_linkmsg(struct net_device *dev, struct nlattr *tb[])
{
	if (dev) {
		if (tb[IFLA_ADDRESS] &&
		    nla_len(tb[IFLA_ADDRESS]) < dev->addr_len)
			return -EINVAL;

		if (tb[IFLA_BROADCAST] &&
		    nla_len(tb[IFLA_BROADCAST]) < dev->addr_len)
			return -EINVAL;
	}

1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757
	if (tb[IFLA_AF_SPEC]) {
		struct nlattr *af;
		int rem, err;

		nla_for_each_nested(af, tb[IFLA_AF_SPEC], rem) {
			const struct rtnl_af_ops *af_ops;

			if (!(af_ops = rtnl_af_lookup(nla_type(af))))
				return -EAFNOSUPPORT;

			if (!af_ops->set_link_af)
				return -EOPNOTSUPP;

			if (af_ops->validate_link_af) {
1758
				err = af_ops->validate_link_af(dev, af);
1759 1760 1761 1762 1763 1764
				if (err < 0)
					return err;
			}
		}
	}

1765 1766 1767
	return 0;
}

1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783
static int handle_infiniband_guid(struct net_device *dev, struct ifla_vf_guid *ivt,
				  int guid_type)
{
	const struct net_device_ops *ops = dev->netdev_ops;

	return ops->ndo_set_vf_guid(dev, ivt->vf, ivt->guid, guid_type);
}

static int handle_vf_guid(struct net_device *dev, struct ifla_vf_guid *ivt, int guid_type)
{
	if (dev->type != ARPHRD_INFINIBAND)
		return -EOPNOTSUPP;

	return handle_infiniband_guid(dev, ivt, guid_type);
}

1784
static int do_setvfinfo(struct net_device *dev, struct nlattr **tb)
1785 1786
{
	const struct net_device_ops *ops = dev->netdev_ops;
1787
	int err = -EINVAL;
1788

1789 1790
	if (tb[IFLA_VF_MAC]) {
		struct ifla_vf_mac *ivm = nla_data(tb[IFLA_VF_MAC]);
1791

1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805
		err = -EOPNOTSUPP;
		if (ops->ndo_set_vf_mac)
			err = ops->ndo_set_vf_mac(dev, ivm->vf,
						  ivm->mac);
		if (err < 0)
			return err;
	}

	if (tb[IFLA_VF_VLAN]) {
		struct ifla_vf_vlan *ivv = nla_data(tb[IFLA_VF_VLAN]);

		err = -EOPNOTSUPP;
		if (ops->ndo_set_vf_vlan)
			err = ops->ndo_set_vf_vlan(dev, ivv->vf, ivv->vlan,
1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831
						   ivv->qos,
						   htons(ETH_P_8021Q));
		if (err < 0)
			return err;
	}

	if (tb[IFLA_VF_VLAN_LIST]) {
		struct ifla_vf_vlan_info *ivvl[MAX_VLAN_LIST_LEN];
		struct nlattr *attr;
		int rem, len = 0;

		err = -EOPNOTSUPP;
		if (!ops->ndo_set_vf_vlan)
			return err;

		nla_for_each_nested(attr, tb[IFLA_VF_VLAN_LIST], rem) {
			if (nla_type(attr) != IFLA_VF_VLAN_INFO ||
			    nla_len(attr) < NLA_HDRLEN) {
				return -EINVAL;
			}
			if (len >= MAX_VLAN_LIST_LEN)
				return -EOPNOTSUPP;
			ivvl[len] = nla_data(attr);

			len++;
		}
1832 1833 1834
		if (len == 0)
			return -EINVAL;

1835 1836
		err = ops->ndo_set_vf_vlan(dev, ivvl[0]->vf, ivvl[0]->vlan,
					   ivvl[0]->qos, ivvl[0]->vlan_proto);
1837 1838
		if (err < 0)
			return err;
1839
	}
1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905

	if (tb[IFLA_VF_TX_RATE]) {
		struct ifla_vf_tx_rate *ivt = nla_data(tb[IFLA_VF_TX_RATE]);
		struct ifla_vf_info ivf;

		err = -EOPNOTSUPP;
		if (ops->ndo_get_vf_config)
			err = ops->ndo_get_vf_config(dev, ivt->vf, &ivf);
		if (err < 0)
			return err;

		err = -EOPNOTSUPP;
		if (ops->ndo_set_vf_rate)
			err = ops->ndo_set_vf_rate(dev, ivt->vf,
						   ivf.min_tx_rate,
						   ivt->rate);
		if (err < 0)
			return err;
	}

	if (tb[IFLA_VF_RATE]) {
		struct ifla_vf_rate *ivt = nla_data(tb[IFLA_VF_RATE]);

		err = -EOPNOTSUPP;
		if (ops->ndo_set_vf_rate)
			err = ops->ndo_set_vf_rate(dev, ivt->vf,
						   ivt->min_tx_rate,
						   ivt->max_tx_rate);
		if (err < 0)
			return err;
	}

	if (tb[IFLA_VF_SPOOFCHK]) {
		struct ifla_vf_spoofchk *ivs = nla_data(tb[IFLA_VF_SPOOFCHK]);

		err = -EOPNOTSUPP;
		if (ops->ndo_set_vf_spoofchk)
			err = ops->ndo_set_vf_spoofchk(dev, ivs->vf,
						       ivs->setting);
		if (err < 0)
			return err;
	}

	if (tb[IFLA_VF_LINK_STATE]) {
		struct ifla_vf_link_state *ivl = nla_data(tb[IFLA_VF_LINK_STATE]);

		err = -EOPNOTSUPP;
		if (ops->ndo_set_vf_link_state)
			err = ops->ndo_set_vf_link_state(dev, ivl->vf,
							 ivl->link_state);
		if (err < 0)
			return err;
	}

	if (tb[IFLA_VF_RSS_QUERY_EN]) {
		struct ifla_vf_rss_query_en *ivrssq_en;

		err = -EOPNOTSUPP;
		ivrssq_en = nla_data(tb[IFLA_VF_RSS_QUERY_EN]);
		if (ops->ndo_set_vf_rss_query_en)
			err = ops->ndo_set_vf_rss_query_en(dev, ivrssq_en->vf,
							   ivrssq_en->setting);
		if (err < 0)
			return err;
	}

H
Hiroshi Shimamoto 已提交
1906 1907 1908 1909 1910 1911 1912 1913 1914 1915
	if (tb[IFLA_VF_TRUST]) {
		struct ifla_vf_trust *ivt = nla_data(tb[IFLA_VF_TRUST]);

		err = -EOPNOTSUPP;
		if (ops->ndo_set_vf_trust)
			err = ops->ndo_set_vf_trust(dev, ivt->vf, ivt->setting);
		if (err < 0)
			return err;
	}

1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933
	if (tb[IFLA_VF_IB_NODE_GUID]) {
		struct ifla_vf_guid *ivt = nla_data(tb[IFLA_VF_IB_NODE_GUID]);

		if (!ops->ndo_set_vf_guid)
			return -EOPNOTSUPP;

		return handle_vf_guid(dev, ivt, IFLA_VF_IB_NODE_GUID);
	}

	if (tb[IFLA_VF_IB_PORT_GUID]) {
		struct ifla_vf_guid *ivt = nla_data(tb[IFLA_VF_IB_PORT_GUID]);

		if (!ops->ndo_set_vf_guid)
			return -EOPNOTSUPP;

		return handle_vf_guid(dev, ivt, IFLA_VF_IB_PORT_GUID);
	}

1934 1935 1936
	return err;
}

1937 1938
static int do_set_master(struct net_device *dev, int ifindex)
{
1939
	struct net_device *upper_dev = netdev_master_upper_dev_get(dev);
1940 1941 1942
	const struct net_device_ops *ops;
	int err;

1943 1944
	if (upper_dev) {
		if (upper_dev->ifindex == ifindex)
1945
			return 0;
1946
		ops = upper_dev->netdev_ops;
1947
		if (ops->ndo_del_slave) {
1948
			err = ops->ndo_del_slave(upper_dev, dev);
1949 1950 1951 1952 1953 1954 1955 1956
			if (err)
				return err;
		} else {
			return -EOPNOTSUPP;
		}
	}

	if (ifindex) {
1957 1958
		upper_dev = __dev_get_by_index(dev_net(dev), ifindex);
		if (!upper_dev)
1959
			return -EINVAL;
1960
		ops = upper_dev->netdev_ops;
1961
		if (ops->ndo_add_slave) {
1962
			err = ops->ndo_add_slave(upper_dev, dev);
1963 1964 1965 1966 1967 1968 1969 1970 1971
			if (err)
				return err;
		} else {
			return -EOPNOTSUPP;
		}
	}
	return 0;
}

1972
#define DO_SETLINK_MODIFIED	0x01
1973 1974
/* notify flag means notify + modified. */
#define DO_SETLINK_NOTIFY	0x03
1975 1976
static int do_setlink(const struct sk_buff *skb,
		      struct net_device *dev, struct ifinfomsg *ifm,
1977
		      struct netlink_ext_ack *extack,
1978
		      struct nlattr **tb, char *ifname, int status)
L
Linus Torvalds 已提交
1979
{
1980
	const struct net_device_ops *ops = dev->netdev_ops;
1981
	int err;
L
Linus Torvalds 已提交
1982

1983
	if (tb[IFLA_NET_NS_PID] || tb[IFLA_NET_NS_FD]) {
1984
		struct net *net = rtnl_link_get_net(dev_net(dev), tb);
1985 1986 1987 1988
		if (IS_ERR(net)) {
			err = PTR_ERR(net);
			goto errout;
		}
1989
		if (!netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) {
1990
			put_net(net);
1991 1992 1993
			err = -EPERM;
			goto errout;
		}
1994 1995 1996 1997
		err = dev_change_net_namespace(dev, net, ifname);
		put_net(net);
		if (err)
			goto errout;
1998
		status |= DO_SETLINK_MODIFIED;
1999 2000
	}

2001
	if (tb[IFLA_MAP]) {
L
Linus Torvalds 已提交
2002 2003 2004
		struct rtnl_link_ifmap *u_map;
		struct ifmap k_map;

2005
		if (!ops->ndo_set_config) {
L
Linus Torvalds 已提交
2006
			err = -EOPNOTSUPP;
2007
			goto errout;
L
Linus Torvalds 已提交
2008 2009 2010 2011
		}

		if (!netif_device_present(dev)) {
			err = -ENODEV;
2012
			goto errout;
L
Linus Torvalds 已提交
2013 2014
		}

2015
		u_map = nla_data(tb[IFLA_MAP]);
L
Linus Torvalds 已提交
2016 2017 2018 2019 2020 2021 2022
		k_map.mem_start = (unsigned long) u_map->mem_start;
		k_map.mem_end = (unsigned long) u_map->mem_end;
		k_map.base_addr = (unsigned short) u_map->base_addr;
		k_map.irq = (unsigned char) u_map->irq;
		k_map.dma = (unsigned char) u_map->dma;
		k_map.port = (unsigned char) u_map->port;

2023
		err = ops->ndo_set_config(dev, &k_map);
2024
		if (err < 0)
2025
			goto errout;
L
Linus Torvalds 已提交
2026

2027
		status |= DO_SETLINK_NOTIFY;
L
Linus Torvalds 已提交
2028 2029
	}

2030
	if (tb[IFLA_ADDRESS]) {
2031 2032 2033 2034 2035 2036 2037
		struct sockaddr *sa;
		int len;

		len = sizeof(sa_family_t) + dev->addr_len;
		sa = kmalloc(len, GFP_KERNEL);
		if (!sa) {
			err = -ENOMEM;
2038
			goto errout;
2039 2040
		}
		sa->sa_family = dev->type;
2041
		memcpy(sa->sa_data, nla_data(tb[IFLA_ADDRESS]),
2042
		       dev->addr_len);
2043
		err = dev_set_mac_address(dev, sa);
2044
		kfree(sa);
L
Linus Torvalds 已提交
2045
		if (err)
2046
			goto errout;
2047
		status |= DO_SETLINK_MODIFIED;
L
Linus Torvalds 已提交
2048 2049
	}

2050 2051 2052
	if (tb[IFLA_MTU]) {
		err = dev_set_mtu(dev, nla_get_u32(tb[IFLA_MTU]));
		if (err < 0)
2053
			goto errout;
2054
		status |= DO_SETLINK_MODIFIED;
L
Linus Torvalds 已提交
2055 2056
	}

2057 2058
	if (tb[IFLA_GROUP]) {
		dev_set_group(dev, nla_get_u32(tb[IFLA_GROUP]));
2059
		status |= DO_SETLINK_NOTIFY;
2060 2061
	}

2062 2063 2064 2065 2066
	/*
	 * Interface selected by interface index but interface
	 * name provided implies that a name change has been
	 * requested.
	 */
2067
	if (ifm->ifi_index > 0 && ifname[0]) {
2068 2069
		err = dev_change_name(dev, ifname);
		if (err < 0)
2070
			goto errout;
2071
		status |= DO_SETLINK_MODIFIED;
L
Linus Torvalds 已提交
2072 2073
	}

2074 2075 2076 2077 2078
	if (tb[IFLA_IFALIAS]) {
		err = dev_set_alias(dev, nla_data(tb[IFLA_IFALIAS]),
				    nla_len(tb[IFLA_IFALIAS]));
		if (err < 0)
			goto errout;
2079
		status |= DO_SETLINK_NOTIFY;
2080 2081
	}

2082 2083
	if (tb[IFLA_BROADCAST]) {
		nla_memcpy(dev->broadcast, tb[IFLA_BROADCAST], dev->addr_len);
2084
		call_netdevice_notifiers(NETDEV_CHANGEADDR, dev);
L
Linus Torvalds 已提交
2085 2086
	}

2087
	if (ifm->ifi_flags || ifm->ifi_change) {
2088
		err = dev_change_flags(dev, rtnl_dev_combine_flags(dev, ifm));
2089 2090
		if (err < 0)
			goto errout;
2091
	}
L
Linus Torvalds 已提交
2092

2093 2094 2095 2096
	if (tb[IFLA_MASTER]) {
		err = do_set_master(dev, nla_get_u32(tb[IFLA_MASTER]));
		if (err)
			goto errout;
2097
		status |= DO_SETLINK_MODIFIED;
2098 2099
	}

2100 2101 2102 2103
	if (tb[IFLA_CARRIER]) {
		err = dev_change_carrier(dev, nla_get_u8(tb[IFLA_CARRIER]));
		if (err)
			goto errout;
2104
		status |= DO_SETLINK_MODIFIED;
2105 2106
	}

2107
	if (tb[IFLA_TXQLEN]) {
2108 2109
		unsigned int value = nla_get_u32(tb[IFLA_TXQLEN]);
		unsigned int orig_len = dev->tx_queue_len;
2110 2111 2112 2113 2114 2115 2116 2117 2118 2119

		if (dev->tx_queue_len ^ value) {
			dev->tx_queue_len = value;
			err = call_netdevice_notifiers(
			      NETDEV_CHANGE_TX_QUEUE_LEN, dev);
			err = notifier_to_errno(err);
			if (err) {
				dev->tx_queue_len = orig_len;
				goto errout;
			}
2120
			status |= DO_SETLINK_NOTIFY;
2121
		}
2122
	}
S
Stefan Rompf 已提交
2123

2124
	if (tb[IFLA_OPERSTATE])
2125
		set_operstate(dev, nla_get_u8(tb[IFLA_OPERSTATE]));
S
Stefan Rompf 已提交
2126

2127
	if (tb[IFLA_LINKMODE]) {
2128 2129
		unsigned char value = nla_get_u8(tb[IFLA_LINKMODE]);

2130
		write_lock_bh(&dev_base_lock);
2131
		if (dev->link_mode ^ value)
2132
			status |= DO_SETLINK_NOTIFY;
2133
		dev->link_mode = value;
2134
		write_unlock_bh(&dev_base_lock);
S
Stefan Rompf 已提交
2135 2136
	}

2137
	if (tb[IFLA_VFINFO_LIST]) {
2138
		struct nlattr *vfinfo[IFLA_VF_MAX + 1];
2139 2140
		struct nlattr *attr;
		int rem;
2141

2142
		nla_for_each_nested(attr, tb[IFLA_VFINFO_LIST], rem) {
2143 2144
			if (nla_type(attr) != IFLA_VF_INFO ||
			    nla_len(attr) < NLA_HDRLEN) {
2145
				err = -EINVAL;
2146
				goto errout;
2147
			}
2148
			err = nla_parse_nested(vfinfo, IFLA_VF_MAX, attr,
2149
					       ifla_vf_policy, NULL);
2150 2151 2152
			if (err < 0)
				goto errout;
			err = do_setvfinfo(dev, vfinfo);
2153 2154
			if (err < 0)
				goto errout;
2155
			status |= DO_SETLINK_NOTIFY;
2156
		}
2157
	}
L
Linus Torvalds 已提交
2158 2159
	err = 0;

2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170
	if (tb[IFLA_VF_PORTS]) {
		struct nlattr *port[IFLA_PORT_MAX+1];
		struct nlattr *attr;
		int vf;
		int rem;

		err = -EOPNOTSUPP;
		if (!ops->ndo_set_vf_port)
			goto errout;

		nla_for_each_nested(attr, tb[IFLA_VF_PORTS], rem) {
2171 2172 2173 2174 2175 2176
			if (nla_type(attr) != IFLA_VF_PORT ||
			    nla_len(attr) < NLA_HDRLEN) {
				err = -EINVAL;
				goto errout;
			}
			err = nla_parse_nested(port, IFLA_PORT_MAX, attr,
2177
					       ifla_port_policy, NULL);
2178 2179 2180 2181 2182 2183 2184 2185 2186 2187
			if (err < 0)
				goto errout;
			if (!port[IFLA_PORT_VF]) {
				err = -EOPNOTSUPP;
				goto errout;
			}
			vf = nla_get_u32(port[IFLA_PORT_VF]);
			err = ops->ndo_set_vf_port(dev, vf, port);
			if (err < 0)
				goto errout;
2188
			status |= DO_SETLINK_NOTIFY;
2189 2190 2191 2192 2193 2194 2195 2196
		}
	}
	err = 0;

	if (tb[IFLA_PORT_SELF]) {
		struct nlattr *port[IFLA_PORT_MAX+1];

		err = nla_parse_nested(port, IFLA_PORT_MAX,
2197 2198
				       tb[IFLA_PORT_SELF], ifla_port_policy,
				       NULL);
2199 2200 2201 2202 2203 2204 2205 2206
		if (err < 0)
			goto errout;

		err = -EOPNOTSUPP;
		if (ops->ndo_set_vf_port)
			err = ops->ndo_set_vf_port(dev, PORT_SELF_VF, port);
		if (err < 0)
			goto errout;
2207
		status |= DO_SETLINK_NOTIFY;
2208
	}
T
Thomas Graf 已提交
2209 2210 2211 2212 2213 2214 2215 2216 2217

	if (tb[IFLA_AF_SPEC]) {
		struct nlattr *af;
		int rem;

		nla_for_each_nested(af, tb[IFLA_AF_SPEC], rem) {
			const struct rtnl_af_ops *af_ops;

			if (!(af_ops = rtnl_af_lookup(nla_type(af))))
2218
				BUG();
T
Thomas Graf 已提交
2219

2220
			err = af_ops->set_link_af(dev, af);
T
Thomas Graf 已提交
2221 2222 2223
			if (err < 0)
				goto errout;

2224
			status |= DO_SETLINK_NOTIFY;
T
Thomas Graf 已提交
2225 2226
		}
	}
2227 2228
	err = 0;

2229 2230 2231 2232 2233 2234 2235 2236
	if (tb[IFLA_PROTO_DOWN]) {
		err = dev_change_proto_down(dev,
					    nla_get_u8(tb[IFLA_PROTO_DOWN]));
		if (err)
			goto errout;
		status |= DO_SETLINK_NOTIFY;
	}

2237 2238
	if (tb[IFLA_XDP]) {
		struct nlattr *xdp[IFLA_XDP_MAX + 1];
2239
		u32 xdp_flags = 0;
2240 2241

		err = nla_parse_nested(xdp, IFLA_XDP_MAX, tb[IFLA_XDP],
2242
				       ifla_xdp_policy, NULL);
2243 2244 2245
		if (err < 0)
			goto errout;

M
Martin KaFai Lau 已提交
2246
		if (xdp[IFLA_XDP_ATTACHED] || xdp[IFLA_XDP_PROG_ID]) {
2247 2248 2249
			err = -EINVAL;
			goto errout;
		}
2250 2251 2252 2253 2254 2255 2256

		if (xdp[IFLA_XDP_FLAGS]) {
			xdp_flags = nla_get_u32(xdp[IFLA_XDP_FLAGS]);
			if (xdp_flags & ~XDP_FLAGS_MASK) {
				err = -EINVAL;
				goto errout;
			}
2257
			if (hweight32(xdp_flags & XDP_FLAGS_MODES) > 1) {
2258 2259 2260
				err = -EINVAL;
				goto errout;
			}
2261 2262
		}

2263
		if (xdp[IFLA_XDP_FD]) {
2264
			err = dev_change_xdp_fd(dev, extack,
2265 2266
						nla_get_s32(xdp[IFLA_XDP_FD]),
						xdp_flags);
2267 2268 2269 2270 2271 2272
			if (err)
				goto errout;
			status |= DO_SETLINK_NOTIFY;
		}
	}

2273
errout:
2274 2275 2276 2277 2278 2279 2280 2281
	if (status & DO_SETLINK_MODIFIED) {
		if (status & DO_SETLINK_NOTIFY)
			netdev_state_change(dev);

		if (err < 0)
			net_warn_ratelimited("A link change request failed with some changes committed already. Interface %s may have been left with an inconsistent configuration, please check.\n",
					     dev->name);
	}
2282

2283 2284
	return err;
}
L
Linus Torvalds 已提交
2285

2286 2287
static int rtnl_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
2288
{
2289
	struct net *net = sock_net(skb->sk);
2290 2291 2292 2293 2294 2295
	struct ifinfomsg *ifm;
	struct net_device *dev;
	int err;
	struct nlattr *tb[IFLA_MAX+1];
	char ifname[IFNAMSIZ];

2296 2297
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy,
			  extack);
2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308
	if (err < 0)
		goto errout;

	if (tb[IFLA_IFNAME])
		nla_strlcpy(ifname, tb[IFLA_IFNAME], IFNAMSIZ);
	else
		ifname[0] = '\0';

	err = -EINVAL;
	ifm = nlmsg_data(nlh);
	if (ifm->ifi_index > 0)
2309
		dev = __dev_get_by_index(net, ifm->ifi_index);
2310
	else if (tb[IFLA_IFNAME])
2311
		dev = __dev_get_by_name(net, ifname);
2312 2313 2314 2315 2316 2317 2318 2319
	else
		goto errout;

	if (dev == NULL) {
		err = -ENODEV;
		goto errout;
	}

E
Eric Dumazet 已提交
2320 2321
	err = validate_linkmsg(dev, tb);
	if (err < 0)
2322
		goto errout;
2323

2324
	err = do_setlink(skb, dev, ifm, extack, tb, ifname, 0);
2325
errout:
L
Linus Torvalds 已提交
2326 2327 2328
	return err;
}

2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364
static int rtnl_group_dellink(const struct net *net, int group)
{
	struct net_device *dev, *aux;
	LIST_HEAD(list_kill);
	bool found = false;

	if (!group)
		return -EPERM;

	for_each_netdev(net, dev) {
		if (dev->group == group) {
			const struct rtnl_link_ops *ops;

			found = true;
			ops = dev->rtnl_link_ops;
			if (!ops || !ops->dellink)
				return -EOPNOTSUPP;
		}
	}

	if (!found)
		return -ENODEV;

	for_each_netdev_safe(net, dev, aux) {
		if (dev->group == group) {
			const struct rtnl_link_ops *ops;

			ops = dev->rtnl_link_ops;
			ops->dellink(dev, &list_kill);
		}
	}
	unregister_netdevice_many(&list_kill);

	return 0;
}

2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380
int rtnl_delete_link(struct net_device *dev)
{
	const struct rtnl_link_ops *ops;
	LIST_HEAD(list_kill);

	ops = dev->rtnl_link_ops;
	if (!ops || !ops->dellink)
		return -EOPNOTSUPP;

	ops->dellink(dev, &list_kill);
	unregister_netdevice_many(&list_kill);

	return 0;
}
EXPORT_SYMBOL_GPL(rtnl_delete_link);

2381 2382
static int rtnl_dellink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
P
Patrick McHardy 已提交
2383
{
2384
	struct net *net = sock_net(skb->sk);
P
Patrick McHardy 已提交
2385 2386 2387 2388 2389 2390
	struct net_device *dev;
	struct ifinfomsg *ifm;
	char ifname[IFNAMSIZ];
	struct nlattr *tb[IFLA_MAX+1];
	int err;

2391
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy, extack);
P
Patrick McHardy 已提交
2392 2393 2394 2395 2396 2397 2398 2399
	if (err < 0)
		return err;

	if (tb[IFLA_IFNAME])
		nla_strlcpy(ifname, tb[IFLA_IFNAME], IFNAMSIZ);

	ifm = nlmsg_data(nlh);
	if (ifm->ifi_index > 0)
2400
		dev = __dev_get_by_index(net, ifm->ifi_index);
P
Patrick McHardy 已提交
2401
	else if (tb[IFLA_IFNAME])
2402
		dev = __dev_get_by_name(net, ifname);
2403 2404
	else if (tb[IFLA_GROUP])
		return rtnl_group_dellink(net, nla_get_u32(tb[IFLA_GROUP]));
P
Patrick McHardy 已提交
2405 2406 2407 2408 2409 2410
	else
		return -EINVAL;

	if (!dev)
		return -ENODEV;

2411
	return rtnl_delete_link(dev);
P
Patrick McHardy 已提交
2412 2413
}

2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427
int rtnl_configure_link(struct net_device *dev, const struct ifinfomsg *ifm)
{
	unsigned int old_flags;
	int err;

	old_flags = dev->flags;
	if (ifm && (ifm->ifi_flags || ifm->ifi_change)) {
		err = __dev_change_flags(dev, rtnl_dev_combine_flags(dev, ifm));
		if (err < 0)
			return err;
	}

	dev->rtnl_link_state = RTNL_LINK_INITIALIZED;

2428
	__dev_notify_flags(dev, old_flags, ~0U);
2429 2430 2431 2432
	return 0;
}
EXPORT_SYMBOL(rtnl_configure_link);

2433
struct net_device *rtnl_create_link(struct net *net,
2434
	const char *ifname, unsigned char name_assign_type,
2435
	const struct rtnl_link_ops *ops, struct nlattr *tb[])
2436 2437
{
	struct net_device *dev;
2438 2439
	unsigned int num_tx_queues = 1;
	unsigned int num_rx_queues = 1;
2440

2441 2442 2443
	if (tb[IFLA_NUM_TX_QUEUES])
		num_tx_queues = nla_get_u32(tb[IFLA_NUM_TX_QUEUES]);
	else if (ops->get_num_tx_queues)
2444
		num_tx_queues = ops->get_num_tx_queues();
2445 2446 2447 2448

	if (tb[IFLA_NUM_RX_QUEUES])
		num_rx_queues = nla_get_u32(tb[IFLA_NUM_RX_QUEUES]);
	else if (ops->get_num_rx_queues)
2449
		num_rx_queues = ops->get_num_rx_queues();
2450

2451
	dev = alloc_netdev_mqs(ops->priv_size, ifname, name_assign_type,
2452
			       ops->setup, num_tx_queues, num_rx_queues);
2453
	if (!dev)
2454
		return ERR_PTR(-ENOMEM);
2455

2456 2457
	dev_net_set(dev, net);
	dev->rtnl_link_ops = ops;
2458
	dev->rtnl_link_state = RTNL_LINK_INITIALIZING;
2459

2460 2461
	if (tb[IFLA_MTU])
		dev->mtu = nla_get_u32(tb[IFLA_MTU]);
2462
	if (tb[IFLA_ADDRESS]) {
2463 2464
		memcpy(dev->dev_addr, nla_data(tb[IFLA_ADDRESS]),
				nla_len(tb[IFLA_ADDRESS]));
2465 2466
		dev->addr_assign_type = NET_ADDR_SET;
	}
2467 2468 2469 2470 2471 2472
	if (tb[IFLA_BROADCAST])
		memcpy(dev->broadcast, nla_data(tb[IFLA_BROADCAST]),
				nla_len(tb[IFLA_BROADCAST]));
	if (tb[IFLA_TXQLEN])
		dev->tx_queue_len = nla_get_u32(tb[IFLA_TXQLEN]);
	if (tb[IFLA_OPERSTATE])
2473
		set_operstate(dev, nla_get_u8(tb[IFLA_OPERSTATE]));
2474 2475
	if (tb[IFLA_LINKMODE])
		dev->link_mode = nla_get_u8(tb[IFLA_LINKMODE]);
2476 2477
	if (tb[IFLA_GROUP])
		dev_set_group(dev, nla_get_u32(tb[IFLA_GROUP]));
2478 2479 2480

	return dev;
}
E
Eric Dumazet 已提交
2481
EXPORT_SYMBOL(rtnl_create_link);
2482

2483 2484
static int rtnl_group_changelink(const struct sk_buff *skb,
		struct net *net, int group,
2485
		struct ifinfomsg *ifm,
2486
		struct netlink_ext_ack *extack,
2487 2488
		struct nlattr **tb)
{
2489
	struct net_device *dev, *aux;
2490 2491
	int err;

2492
	for_each_netdev_safe(net, dev, aux) {
2493
		if (dev->group == group) {
2494
			err = do_setlink(skb, dev, ifm, extack, tb, NULL, 0);
2495 2496 2497 2498 2499 2500 2501 2502
			if (err < 0)
				return err;
		}
	}

	return 0;
}

2503 2504
static int rtnl_newlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
P
Patrick McHardy 已提交
2505
{
2506
	struct net *net = sock_net(skb->sk);
P
Patrick McHardy 已提交
2507
	const struct rtnl_link_ops *ops;
2508
	const struct rtnl_link_ops *m_ops = NULL;
P
Patrick McHardy 已提交
2509
	struct net_device *dev;
2510
	struct net_device *master_dev = NULL;
P
Patrick McHardy 已提交
2511 2512 2513 2514 2515
	struct ifinfomsg *ifm;
	char kind[MODULE_NAME_LEN];
	char ifname[IFNAMSIZ];
	struct nlattr *tb[IFLA_MAX+1];
	struct nlattr *linkinfo[IFLA_INFO_MAX+1];
2516
	unsigned char name_assign_type = NET_NAME_USER;
P
Patrick McHardy 已提交
2517 2518
	int err;

2519
#ifdef CONFIG_MODULES
P
Patrick McHardy 已提交
2520
replay:
2521
#endif
2522
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy, extack);
P
Patrick McHardy 已提交
2523 2524 2525 2526 2527 2528 2529 2530 2531 2532
	if (err < 0)
		return err;

	if (tb[IFLA_IFNAME])
		nla_strlcpy(ifname, tb[IFLA_IFNAME], IFNAMSIZ);
	else
		ifname[0] = '\0';

	ifm = nlmsg_data(nlh);
	if (ifm->ifi_index > 0)
2533
		dev = __dev_get_by_index(net, ifm->ifi_index);
2534 2535 2536 2537 2538 2539
	else {
		if (ifname[0])
			dev = __dev_get_by_name(net, ifname);
		else
			dev = NULL;
	}
P
Patrick McHardy 已提交
2540

2541 2542 2543 2544 2545 2546
	if (dev) {
		master_dev = netdev_master_upper_dev_get(dev);
		if (master_dev)
			m_ops = master_dev->rtnl_link_ops;
	}

E
Eric Dumazet 已提交
2547 2548
	err = validate_linkmsg(dev, tb);
	if (err < 0)
2549 2550
		return err;

P
Patrick McHardy 已提交
2551 2552
	if (tb[IFLA_LINKINFO]) {
		err = nla_parse_nested(linkinfo, IFLA_INFO_MAX,
2553 2554
				       tb[IFLA_LINKINFO], ifla_info_policy,
				       NULL);
P
Patrick McHardy 已提交
2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568
		if (err < 0)
			return err;
	} else
		memset(linkinfo, 0, sizeof(linkinfo));

	if (linkinfo[IFLA_INFO_KIND]) {
		nla_strlcpy(kind, linkinfo[IFLA_INFO_KIND], sizeof(kind));
		ops = rtnl_link_ops_get(kind);
	} else {
		kind[0] = '\0';
		ops = NULL;
	}

	if (1) {
S
Sasha Levin 已提交
2569 2570
		struct nlattr *attr[ops ? ops->maxtype + 1 : 1];
		struct nlattr *slave_attr[m_ops ? m_ops->slave_maxtype + 1 : 1];
2571 2572
		struct nlattr **data = NULL;
		struct nlattr **slave_data = NULL;
2573
		struct net *dest_net, *link_net = NULL;
P
Patrick McHardy 已提交
2574 2575 2576 2577 2578

		if (ops) {
			if (ops->maxtype && linkinfo[IFLA_INFO_DATA]) {
				err = nla_parse_nested(attr, ops->maxtype,
						       linkinfo[IFLA_INFO_DATA],
2579
						       ops->policy, NULL);
P
Patrick McHardy 已提交
2580 2581 2582 2583 2584
				if (err < 0)
					return err;
				data = attr;
			}
			if (ops->validate) {
2585
				err = ops->validate(tb, data, extack);
P
Patrick McHardy 已提交
2586 2587 2588 2589 2590
				if (err < 0)
					return err;
			}
		}

2591 2592 2593 2594 2595 2596
		if (m_ops) {
			if (m_ops->slave_maxtype &&
			    linkinfo[IFLA_INFO_SLAVE_DATA]) {
				err = nla_parse_nested(slave_attr,
						       m_ops->slave_maxtype,
						       linkinfo[IFLA_INFO_SLAVE_DATA],
2597 2598
						       m_ops->slave_policy,
						       NULL);
2599 2600 2601 2602 2603
				if (err < 0)
					return err;
				slave_data = slave_attr;
			}
			if (m_ops->slave_validate) {
2604 2605
				err = m_ops->slave_validate(tb, slave_data,
							    extack);
2606 2607 2608 2609 2610
				if (err < 0)
					return err;
			}
		}

P
Patrick McHardy 已提交
2611
		if (dev) {
2612
			int status = 0;
P
Patrick McHardy 已提交
2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623

			if (nlh->nlmsg_flags & NLM_F_EXCL)
				return -EEXIST;
			if (nlh->nlmsg_flags & NLM_F_REPLACE)
				return -EOPNOTSUPP;

			if (linkinfo[IFLA_INFO_DATA]) {
				if (!ops || ops != dev->rtnl_link_ops ||
				    !ops->changelink)
					return -EOPNOTSUPP;

2624
				err = ops->changelink(dev, tb, data, extack);
P
Patrick McHardy 已提交
2625 2626
				if (err < 0)
					return err;
2627
				status |= DO_SETLINK_NOTIFY;
P
Patrick McHardy 已提交
2628 2629
			}

2630 2631 2632 2633 2634
			if (linkinfo[IFLA_INFO_SLAVE_DATA]) {
				if (!m_ops || !m_ops->slave_changelink)
					return -EOPNOTSUPP;

				err = m_ops->slave_changelink(master_dev, dev,
2635 2636
							      tb, slave_data,
							      extack);
2637 2638
				if (err < 0)
					return err;
2639
				status |= DO_SETLINK_NOTIFY;
2640 2641
			}

2642 2643
			return do_setlink(skb, dev, ifm, extack, tb, ifname,
					  status);
P
Patrick McHardy 已提交
2644 2645
		}

2646 2647
		if (!(nlh->nlmsg_flags & NLM_F_CREATE)) {
			if (ifm->ifi_index == 0 && tb[IFLA_GROUP])
2648
				return rtnl_group_changelink(skb, net,
2649
						nla_get_u32(tb[IFLA_GROUP]),
2650
						ifm, extack, tb);
P
Patrick McHardy 已提交
2651
			return -ENODEV;
2652
		}
P
Patrick McHardy 已提交
2653

2654
		if (tb[IFLA_MAP] || tb[IFLA_PROTINFO])
P
Patrick McHardy 已提交
2655 2656 2657
			return -EOPNOTSUPP;

		if (!ops) {
2658
#ifdef CONFIG_MODULES
P
Patrick McHardy 已提交
2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670
			if (kind[0]) {
				__rtnl_unlock();
				request_module("rtnl-link-%s", kind);
				rtnl_lock();
				ops = rtnl_link_ops_get(kind);
				if (ops)
					goto replay;
			}
#endif
			return -EOPNOTSUPP;
		}

2671 2672 2673
		if (!ops->setup)
			return -EOPNOTSUPP;

2674
		if (!ifname[0]) {
P
Patrick McHardy 已提交
2675
			snprintf(ifname, IFNAMSIZ, "%s%%d", ops->kind);
2676 2677
			name_assign_type = NET_NAME_ENUM;
		}
2678

2679
		dest_net = rtnl_link_get_net(net, tb);
2680 2681 2682
		if (IS_ERR(dest_net))
			return PTR_ERR(dest_net);

2683 2684 2685 2686
		err = -EPERM;
		if (!netlink_ns_capable(skb, dest_net->user_ns, CAP_NET_ADMIN))
			goto out;

2687 2688 2689 2690 2691 2692 2693 2694
		if (tb[IFLA_LINK_NETNSID]) {
			int id = nla_get_s32(tb[IFLA_LINK_NETNSID]);

			link_net = get_net_ns_by_id(dest_net, id);
			if (!link_net) {
				err =  -EINVAL;
				goto out;
			}
2695 2696 2697
			err = -EPERM;
			if (!netlink_ns_capable(skb, link_net->user_ns, CAP_NET_ADMIN))
				goto out;
2698 2699 2700 2701
		}

		dev = rtnl_create_link(link_net ? : dest_net, ifname,
				       name_assign_type, ops, tb);
2702
		if (IS_ERR(dev)) {
2703
			err = PTR_ERR(dev);
2704 2705 2706 2707 2708
			goto out;
		}

		dev->ifindex = ifm->ifi_index;

2709
		if (ops->newlink) {
2710 2711
			err = ops->newlink(link_net ? : net, dev, tb, data,
					   extack);
2712
			/* Drivers should call free_netdev() in ->destructor
2713 2714
			 * and unregister it on failure after registration
			 * so that device could be finally freed in rtnl_unlock.
2715
			 */
2716 2717 2718 2719
			if (err < 0) {
				/* If device is not registered at all, free it now */
				if (dev->reg_state == NETREG_UNINITIALIZED)
					free_netdev(dev);
2720
				goto out;
2721
			}
2722
		} else {
2723
			err = register_netdevice(dev);
2724 2725 2726 2727
			if (err < 0) {
				free_netdev(dev);
				goto out;
			}
2728
		}
2729
		err = rtnl_configure_link(dev, ifm);
2730 2731
		if (err < 0)
			goto out_unregister;
2732
		if (link_net) {
2733
			err = dev_change_net_namespace(dev, dest_net, ifname);
2734
			if (err < 0)
2735
				goto out_unregister;
2736
		}
2737 2738 2739 2740 2741
		if (tb[IFLA_MASTER]) {
			err = do_set_master(dev, nla_get_u32(tb[IFLA_MASTER]));
			if (err)
				goto out_unregister;
		}
2742
out:
2743 2744
		if (link_net)
			put_net(link_net);
2745
		put_net(dest_net);
P
Patrick McHardy 已提交
2746
		return err;
2747 2748 2749 2750 2751 2752 2753 2754 2755 2756
out_unregister:
		if (ops->newlink) {
			LIST_HEAD(list_kill);

			ops->dellink(dev, &list_kill);
			unregister_netdevice_many(&list_kill);
		} else {
			unregister_netdevice(dev);
		}
		goto out;
P
Patrick McHardy 已提交
2757 2758 2759
	}
}

2760 2761
static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
2762
{
2763
	struct net *net = sock_net(skb->sk);
2764
	struct ifinfomsg *ifm;
2765
	char ifname[IFNAMSIZ];
2766 2767 2768
	struct nlattr *tb[IFLA_MAX+1];
	struct net_device *dev = NULL;
	struct sk_buff *nskb;
2769
	int err;
2770
	u32 ext_filter_mask = 0;
2771

2772
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy, extack);
2773
	if (err < 0)
2774
		return err;
2775

2776 2777 2778
	if (tb[IFLA_IFNAME])
		nla_strlcpy(ifname, tb[IFLA_IFNAME], IFNAMSIZ);

2779 2780 2781
	if (tb[IFLA_EXT_MASK])
		ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);

2782
	ifm = nlmsg_data(nlh);
2783 2784 2785 2786 2787
	if (ifm->ifi_index > 0)
		dev = __dev_get_by_index(net, ifm->ifi_index);
	else if (tb[IFLA_IFNAME])
		dev = __dev_get_by_name(net, ifname);
	else
2788 2789
		return -EINVAL;

2790 2791 2792
	if (dev == NULL)
		return -ENODEV;

2793
	nskb = nlmsg_new(if_nlmsg_size(dev, ext_filter_mask), GFP_KERNEL);
2794 2795
	if (nskb == NULL)
		return -ENOBUFS;
2796

2797
	err = rtnl_fill_ifinfo(nskb, dev, RTM_NEWLINK, NETLINK_CB(skb).portid,
2798
			       nlh->nlmsg_seq, 0, 0, ext_filter_mask, 0);
2799 2800 2801 2802
	if (err < 0) {
		/* -EMSGSIZE implies BUG in if_nlmsg_size */
		WARN_ON(err == -EMSGSIZE);
		kfree_skb(nskb);
2803
	} else
2804
		err = rtnl_unicast(nskb, net, NETLINK_CB(skb).portid);
2805

2806
	return err;
2807 2808
}

2809
static u16 rtnl_calcit(struct sk_buff *skb, struct nlmsghdr *nlh)
2810
{
2811 2812 2813 2814 2815
	struct net *net = sock_net(skb->sk);
	struct net_device *dev;
	struct nlattr *tb[IFLA_MAX+1];
	u32 ext_filter_mask = 0;
	u16 min_ifinfo_dump_size = 0;
2816 2817 2818 2819 2820
	int hdrlen;

	/* Same kernel<->userspace interface hack as in rtnl_dump_ifinfo. */
	hdrlen = nlmsg_len(nlh) < sizeof(struct ifinfomsg) ?
		 sizeof(struct rtgenmsg) : sizeof(struct ifinfomsg);
2821

2822
	if (nlmsg_parse(nlh, hdrlen, tb, IFLA_MAX, ifla_policy, NULL) >= 0) {
2823 2824 2825
		if (tb[IFLA_EXT_MASK])
			ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);
	}
2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838

	if (!ext_filter_mask)
		return NLMSG_GOODSIZE;
	/*
	 * traverse the list of net devices and compute the minimum
	 * buffer size based upon the filter mask.
	 */
	list_for_each_entry(dev, &net->dev_base_head, dev_list) {
		min_ifinfo_dump_size = max_t(u16, min_ifinfo_dump_size,
					     if_nlmsg_size(dev,
						           ext_filter_mask));
	}

2839
	return nlmsg_total_size(min_ifinfo_dump_size);
2840 2841
}

A
Adrian Bunk 已提交
2842
static int rtnl_dump_all(struct sk_buff *skb, struct netlink_callback *cb)
L
Linus Torvalds 已提交
2843 2844 2845 2846 2847 2848
{
	int idx;
	int s_idx = cb->family;

	if (s_idx == 0)
		s_idx = 1;
2849
	for (idx = 1; idx <= RTNL_FAMILY_MAX; idx++) {
L
Linus Torvalds 已提交
2850 2851 2852
		int type = cb->nlh->nlmsg_type-RTM_BASE;
		if (idx < s_idx || idx == PF_PACKET)
			continue;
2853 2854
		if (rtnl_msg_handlers[idx] == NULL ||
		    rtnl_msg_handlers[idx][type].dumpit == NULL)
L
Linus Torvalds 已提交
2855
			continue;
2856
		if (idx > s_idx) {
L
Linus Torvalds 已提交
2857
			memset(&cb->args[0], 0, sizeof(cb->args));
2858 2859 2860
			cb->prev_seq = 0;
			cb->seq = 0;
		}
2861
		if (rtnl_msg_handlers[idx][type].dumpit(skb, cb))
L
Linus Torvalds 已提交
2862 2863 2864 2865 2866 2867 2868
			break;
	}
	cb->family = idx;

	return skb->len;
}

2869
struct sk_buff *rtmsg_ifinfo_build_skb(int type, struct net_device *dev,
2870 2871
				       unsigned int change,
				       u32 event, gfp_t flags)
L
Linus Torvalds 已提交
2872
{
2873
	struct net *net = dev_net(dev);
L
Linus Torvalds 已提交
2874
	struct sk_buff *skb;
2875
	int err = -ENOBUFS;
2876
	size_t if_info_size;
L
Linus Torvalds 已提交
2877

2878
	skb = nlmsg_new((if_info_size = if_nlmsg_size(dev, 0)), flags);
2879 2880
	if (skb == NULL)
		goto errout;
L
Linus Torvalds 已提交
2881

2882
	err = rtnl_fill_ifinfo(skb, dev, type, 0, 0, change, 0, 0, event);
2883 2884 2885 2886 2887 2888
	if (err < 0) {
		/* -EMSGSIZE implies BUG in if_nlmsg_size() */
		WARN_ON(err == -EMSGSIZE);
		kfree_skb(skb);
		goto errout;
	}
2889
	return skb;
2890 2891
errout:
	if (err < 0)
2892
		rtnl_set_sk_err(net, RTNLGRP_LINK, err);
2893 2894 2895 2896 2897 2898 2899 2900 2901 2902
	return NULL;
}

void rtmsg_ifinfo_send(struct sk_buff *skb, struct net_device *dev, gfp_t flags)
{
	struct net *net = dev_net(dev);

	rtnl_notify(skb, net, 0, RTNLGRP_LINK, NULL, flags);
}

2903 2904 2905
static void rtmsg_ifinfo_event(int type, struct net_device *dev,
			       unsigned int change, u32 event,
			       gfp_t flags)
2906 2907 2908
{
	struct sk_buff *skb;

2909 2910 2911
	if (dev->reg_state != NETREG_REGISTERED)
		return;

2912
	skb = rtmsg_ifinfo_build_skb(type, dev, change, event, flags);
2913 2914
	if (skb)
		rtmsg_ifinfo_send(skb, dev, flags);
L
Linus Torvalds 已提交
2915
}
2916 2917 2918 2919

void rtmsg_ifinfo(int type, struct net_device *dev, unsigned int change,
		  gfp_t flags)
{
2920
	rtmsg_ifinfo_event(type, dev, change, rtnl_get_event(0), flags);
2921
}
2922
EXPORT_SYMBOL(rtmsg_ifinfo);
L
Linus Torvalds 已提交
2923

2924 2925
static int nlmsg_populate_fdb_fill(struct sk_buff *skb,
				   struct net_device *dev,
2926
				   u8 *addr, u16 vid, u32 pid, u32 seq,
2927
				   int type, unsigned int flags,
2928
				   int nlflags, u16 ndm_state)
2929 2930 2931 2932
{
	struct nlmsghdr *nlh;
	struct ndmsg *ndm;

2933
	nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ndm), nlflags);
2934 2935 2936 2937 2938 2939 2940 2941 2942 2943
	if (!nlh)
		return -EMSGSIZE;

	ndm = nlmsg_data(nlh);
	ndm->ndm_family  = AF_BRIDGE;
	ndm->ndm_pad1	 = 0;
	ndm->ndm_pad2    = 0;
	ndm->ndm_flags	 = flags;
	ndm->ndm_type	 = 0;
	ndm->ndm_ifindex = dev->ifindex;
2944
	ndm->ndm_state   = ndm_state;
2945 2946 2947

	if (nla_put(skb, NDA_LLADDR, ETH_ALEN, addr))
		goto nla_put_failure;
2948 2949 2950
	if (vid)
		if (nla_put(skb, NDA_VLAN, sizeof(u16), &vid))
			goto nla_put_failure;
2951

2952 2953
	nlmsg_end(skb, nlh);
	return 0;
2954 2955 2956 2957 2958 2959

nla_put_failure:
	nlmsg_cancel(skb, nlh);
	return -EMSGSIZE;
}

2960 2961
static inline size_t rtnl_fdb_nlmsg_size(void)
{
2962 2963 2964 2965
	return NLMSG_ALIGN(sizeof(struct ndmsg)) +
	       nla_total_size(ETH_ALEN) +	/* NDA_LLADDR */
	       nla_total_size(sizeof(u16)) +	/* NDA_VLAN */
	       0;
2966 2967
}

2968 2969
static void rtnl_fdb_notify(struct net_device *dev, u8 *addr, u16 vid, int type,
			    u16 ndm_state)
2970 2971 2972 2973 2974 2975 2976 2977 2978
{
	struct net *net = dev_net(dev);
	struct sk_buff *skb;
	int err = -ENOBUFS;

	skb = nlmsg_new(rtnl_fdb_nlmsg_size(), GFP_ATOMIC);
	if (!skb)
		goto errout;

2979
	err = nlmsg_populate_fdb_fill(skb, dev, addr, vid,
2980
				      0, 0, type, NTF_SELF, 0, ndm_state);
2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991
	if (err < 0) {
		kfree_skb(skb);
		goto errout;
	}

	rtnl_notify(skb, net, 0, RTNLGRP_NEIGH, NULL, GFP_ATOMIC);
	return;
errout:
	rtnl_set_sk_err(net, RTNLGRP_NEIGH, err);
}

2992 2993 2994 2995 2996 2997
/**
 * ndo_dflt_fdb_add - default netdevice operation to add an FDB entry
 */
int ndo_dflt_fdb_add(struct ndmsg *ndm,
		     struct nlattr *tb[],
		     struct net_device *dev,
2998
		     const unsigned char *addr, u16 vid,
2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010
		     u16 flags)
{
	int err = -EINVAL;

	/* If aging addresses are supported device will need to
	 * implement its own handler for this.
	 */
	if (ndm->ndm_state && !(ndm->ndm_state & NUD_PERMANENT)) {
		pr_info("%s: FDB only supports static addresses\n", dev->name);
		return err;
	}

3011 3012 3013 3014 3015
	if (vid) {
		pr_info("%s: vlans aren't supported yet for dev_uc|mc_add()\n", dev->name);
		return err;
	}

3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028
	if (is_unicast_ether_addr(addr) || is_link_local_ether_addr(addr))
		err = dev_uc_add_excl(dev, addr);
	else if (is_multicast_ether_addr(addr))
		err = dev_mc_add_excl(dev, addr);

	/* Only return duplicate errors if NLM_F_EXCL is set */
	if (err == -EEXIST && !(flags & NLM_F_EXCL))
		err = 0;

	return err;
}
EXPORT_SYMBOL(ndo_dflt_fdb_add);

3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050
static int fdb_vid_parse(struct nlattr *vlan_attr, u16 *p_vid)
{
	u16 vid = 0;

	if (vlan_attr) {
		if (nla_len(vlan_attr) != sizeof(u16)) {
			pr_info("PF_BRIDGE: RTM_NEWNEIGH with invalid vlan\n");
			return -EINVAL;
		}

		vid = nla_get_u16(vlan_attr);

		if (!vid || vid >= VLAN_VID_MASK) {
			pr_info("PF_BRIDGE: RTM_NEWNEIGH with invalid vlan id %d\n",
				vid);
			return -EINVAL;
		}
	}
	*p_vid = vid;
	return 0;
}

3051 3052
static int rtnl_fdb_add(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
3053 3054 3055 3056 3057 3058
{
	struct net *net = sock_net(skb->sk);
	struct ndmsg *ndm;
	struct nlattr *tb[NDA_MAX+1];
	struct net_device *dev;
	u8 *addr;
3059
	u16 vid;
3060 3061
	int err;

3062
	err = nlmsg_parse(nlh, sizeof(*ndm), tb, NDA_MAX, NULL, extack);
3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084
	if (err < 0)
		return err;

	ndm = nlmsg_data(nlh);
	if (ndm->ndm_ifindex == 0) {
		pr_info("PF_BRIDGE: RTM_NEWNEIGH with invalid ifindex\n");
		return -EINVAL;
	}

	dev = __dev_get_by_index(net, ndm->ndm_ifindex);
	if (dev == NULL) {
		pr_info("PF_BRIDGE: RTM_NEWNEIGH with unknown ifindex\n");
		return -ENODEV;
	}

	if (!tb[NDA_LLADDR] || nla_len(tb[NDA_LLADDR]) != ETH_ALEN) {
		pr_info("PF_BRIDGE: RTM_NEWNEIGH with invalid address\n");
		return -EINVAL;
	}

	addr = nla_data(tb[NDA_LLADDR]);

3085 3086 3087 3088
	err = fdb_vid_parse(tb[NDA_VLAN], &vid);
	if (err)
		return err;

3089 3090 3091 3092 3093
	err = -EOPNOTSUPP;

	/* Support fdb on master device the net/bridge default case */
	if ((!ndm->ndm_flags || ndm->ndm_flags & NTF_MASTER) &&
	    (dev->priv_flags & IFF_BRIDGE_PORT)) {
3094 3095 3096
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);
		const struct net_device_ops *ops = br_dev->netdev_ops;

3097 3098
		err = ops->ndo_fdb_add(ndm, tb, dev, addr, vid,
				       nlh->nlmsg_flags);
3099 3100 3101 3102 3103 3104 3105
		if (err)
			goto out;
		else
			ndm->ndm_flags &= ~NTF_MASTER;
	}

	/* Embedded bridge, macvlan, and any other device support */
3106 3107 3108
	if ((ndm->ndm_flags & NTF_SELF)) {
		if (dev->netdev_ops->ndo_fdb_add)
			err = dev->netdev_ops->ndo_fdb_add(ndm, tb, dev, addr,
3109
							   vid,
3110 3111
							   nlh->nlmsg_flags);
		else
3112
			err = ndo_dflt_fdb_add(ndm, tb, dev, addr, vid,
3113
					       nlh->nlmsg_flags);
3114

3115
		if (!err) {
3116 3117
			rtnl_fdb_notify(dev, addr, vid, RTM_NEWNEIGH,
					ndm->ndm_state);
3118
			ndm->ndm_flags &= ~NTF_SELF;
3119
		}
3120 3121 3122 3123 3124
	}
out:
	return err;
}

3125 3126 3127 3128 3129 3130
/**
 * ndo_dflt_fdb_del - default netdevice operation to delete an FDB entry
 */
int ndo_dflt_fdb_del(struct ndmsg *ndm,
		     struct nlattr *tb[],
		     struct net_device *dev,
3131
		     const unsigned char *addr, u16 vid)
3132
{
3133
	int err = -EINVAL;
3134 3135 3136 3137

	/* If aging addresses are supported device will need to
	 * implement its own handler for this.
	 */
3138
	if (!(ndm->ndm_state & NUD_PERMANENT)) {
3139
		pr_info("%s: FDB only supports static addresses\n", dev->name);
3140
		return err;
3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151
	}

	if (is_unicast_ether_addr(addr) || is_link_local_ether_addr(addr))
		err = dev_uc_del(dev, addr);
	else if (is_multicast_ether_addr(addr))
		err = dev_mc_del(dev, addr);

	return err;
}
EXPORT_SYMBOL(ndo_dflt_fdb_del);

3152 3153
static int rtnl_fdb_del(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
3154 3155 3156
{
	struct net *net = sock_net(skb->sk);
	struct ndmsg *ndm;
3157
	struct nlattr *tb[NDA_MAX+1];
3158 3159 3160
	struct net_device *dev;
	int err = -EINVAL;
	__u8 *addr;
3161
	u16 vid;
3162

3163
	if (!netlink_capable(skb, CAP_NET_ADMIN))
3164 3165
		return -EPERM;

3166
	err = nlmsg_parse(nlh, sizeof(*ndm), tb, NDA_MAX, NULL, extack);
3167 3168
	if (err < 0)
		return err;
3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181

	ndm = nlmsg_data(nlh);
	if (ndm->ndm_ifindex == 0) {
		pr_info("PF_BRIDGE: RTM_DELNEIGH with invalid ifindex\n");
		return -EINVAL;
	}

	dev = __dev_get_by_index(net, ndm->ndm_ifindex);
	if (dev == NULL) {
		pr_info("PF_BRIDGE: RTM_DELNEIGH with unknown ifindex\n");
		return -ENODEV;
	}

3182 3183 3184 3185 3186 3187
	if (!tb[NDA_LLADDR] || nla_len(tb[NDA_LLADDR]) != ETH_ALEN) {
		pr_info("PF_BRIDGE: RTM_DELNEIGH with invalid address\n");
		return -EINVAL;
	}

	addr = nla_data(tb[NDA_LLADDR]);
3188

3189 3190 3191 3192
	err = fdb_vid_parse(tb[NDA_VLAN], &vid);
	if (err)
		return err;

3193 3194 3195 3196 3197
	err = -EOPNOTSUPP;

	/* Support fdb on master device the net/bridge default case */
	if ((!ndm->ndm_flags || ndm->ndm_flags & NTF_MASTER) &&
	    (dev->priv_flags & IFF_BRIDGE_PORT)) {
3198 3199
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);
		const struct net_device_ops *ops = br_dev->netdev_ops;
3200

3201
		if (ops->ndo_fdb_del)
3202
			err = ops->ndo_fdb_del(ndm, tb, dev, addr, vid);
3203 3204 3205 3206 3207 3208 3209 3210

		if (err)
			goto out;
		else
			ndm->ndm_flags &= ~NTF_MASTER;
	}

	/* Embedded bridge, macvlan, and any other device support */
3211 3212
	if (ndm->ndm_flags & NTF_SELF) {
		if (dev->netdev_ops->ndo_fdb_del)
3213 3214
			err = dev->netdev_ops->ndo_fdb_del(ndm, tb, dev, addr,
							   vid);
3215
		else
3216
			err = ndo_dflt_fdb_del(ndm, tb, dev, addr, vid);
3217

3218
		if (!err) {
3219 3220
			rtnl_fdb_notify(dev, addr, vid, RTM_DELNEIGH,
					ndm->ndm_state);
3221
			ndm->ndm_flags &= ~NTF_SELF;
3222
		}
3223 3224 3225 3226 3227
	}
out:
	return err;
}

3228 3229 3230 3231 3232 3233 3234 3235
static int nlmsg_populate_fdb(struct sk_buff *skb,
			      struct netlink_callback *cb,
			      struct net_device *dev,
			      int *idx,
			      struct netdev_hw_addr_list *list)
{
	struct netdev_hw_addr *ha;
	int err;
3236
	u32 portid, seq;
3237

3238
	portid = NETLINK_CB(cb->skb).portid;
3239 3240 3241
	seq = cb->nlh->nlmsg_seq;

	list_for_each_entry(ha, &list->list, list) {
3242
		if (*idx < cb->args[2])
3243 3244
			goto skip;

3245
		err = nlmsg_populate_fdb_fill(skb, dev, ha->addr, 0,
3246
					      portid, seq,
3247
					      RTM_NEWNEIGH, NTF_SELF,
3248
					      NLM_F_MULTI, NUD_PERMANENT);
3249 3250 3251 3252 3253 3254 3255 3256 3257
		if (err < 0)
			return err;
skip:
		*idx += 1;
	}
	return 0;
}

/**
3258
 * ndo_dflt_fdb_dump - default netdevice operation to dump an FDB table.
3259 3260 3261 3262
 * @nlh: netlink message header
 * @dev: netdevice
 *
 * Default netdevice operation to dump the existing unicast address list.
3263
 * Returns number of addresses from list put in skb.
3264 3265 3266 3267
 */
int ndo_dflt_fdb_dump(struct sk_buff *skb,
		      struct netlink_callback *cb,
		      struct net_device *dev,
3268
		      struct net_device *filter_dev,
3269
		      int *idx)
3270 3271 3272 3273
{
	int err;

	netif_addr_lock_bh(dev);
3274
	err = nlmsg_populate_fdb(skb, cb, dev, idx, &dev->uc);
3275 3276
	if (err)
		goto out;
3277
	err = nlmsg_populate_fdb(skb, cb, dev, idx, &dev->mc);
3278 3279
out:
	netif_addr_unlock_bh(dev);
3280
	return err;
3281 3282 3283
}
EXPORT_SYMBOL(ndo_dflt_fdb_dump);

3284 3285 3286
static int rtnl_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb)
{
	struct net_device *dev;
3287 3288 3289 3290 3291 3292
	struct nlattr *tb[IFLA_MAX+1];
	struct net_device *br_dev = NULL;
	const struct net_device_ops *ops = NULL;
	const struct net_device_ops *cops = NULL;
	struct ifinfomsg *ifm = nlmsg_data(cb->nlh);
	struct net *net = sock_net(skb->sk);
3293
	struct hlist_head *head;
3294 3295
	int brport_idx = 0;
	int br_idx = 0;
3296 3297 3298 3299
	int h, s_h;
	int idx = 0, s_idx;
	int err = 0;
	int fidx = 0;
3300

3301 3302 3303 3304 3305
	err = nlmsg_parse(cb->nlh, sizeof(struct ifinfomsg), tb,
			  IFLA_MAX, ifla_policy, NULL);
	if (err < 0) {
		return -EINVAL;
	} else if (err == 0) {
3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319
		if (tb[IFLA_MASTER])
			br_idx = nla_get_u32(tb[IFLA_MASTER]);
	}

	brport_idx = ifm->ifi_index;

	if (br_idx) {
		br_dev = __dev_get_by_index(net, br_idx);
		if (!br_dev)
			return -ENODEV;

		ops = br_dev->netdev_ops;
	}

3320 3321
	s_h = cb->args[0];
	s_idx = cb->args[1];
3322

3323 3324 3325 3326
	for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) {
		idx = 0;
		head = &net->dev_index_head[h];
		hlist_for_each_entry(dev, head, index_hlist) {
3327

3328
			if (brport_idx && (dev->ifindex != brport_idx))
3329 3330
				continue;

3331 3332 3333 3334 3335 3336 3337 3338 3339
			if (!br_idx) { /* user did not specify a specific bridge */
				if (dev->priv_flags & IFF_BRIDGE_PORT) {
					br_dev = netdev_master_upper_dev_get(dev);
					cops = br_dev->netdev_ops;
				}
			} else {
				if (dev != br_dev &&
				    !(dev->priv_flags & IFF_BRIDGE_PORT))
					continue;
3340

3341 3342 3343 3344 3345
				if (br_dev != netdev_master_upper_dev_get(dev) &&
				    !(dev->priv_flags & IFF_EBRIDGE))
					continue;
				cops = ops;
			}
3346

3347 3348
			if (idx < s_idx)
				goto cont;
3349

3350 3351 3352 3353 3354 3355 3356 3357 3358
			if (dev->priv_flags & IFF_BRIDGE_PORT) {
				if (cops && cops->ndo_fdb_dump) {
					err = cops->ndo_fdb_dump(skb, cb,
								br_dev, dev,
								&fidx);
					if (err == -EMSGSIZE)
						goto out;
				}
			}
3359

3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377
			if (dev->netdev_ops->ndo_fdb_dump)
				err = dev->netdev_ops->ndo_fdb_dump(skb, cb,
								    dev, NULL,
								    &fidx);
			else
				err = ndo_dflt_fdb_dump(skb, cb, dev, NULL,
							&fidx);
			if (err == -EMSGSIZE)
				goto out;

			cops = NULL;

			/* reset fdb offset to 0 for rest of the interfaces */
			cb->args[2] = 0;
			fidx = 0;
cont:
			idx++;
		}
3378 3379
	}

3380 3381 3382 3383 3384
out:
	cb->args[0] = h;
	cb->args[1] = idx;
	cb->args[2] = fidx;

3385 3386 3387
	return skb->len;
}

3388 3389 3390 3391 3392 3393 3394 3395
static int brport_nla_put_flag(struct sk_buff *skb, u32 flags, u32 mask,
			       unsigned int attrnum, unsigned int flag)
{
	if (mask & flag)
		return nla_put_u8(skb, attrnum, !!(flags & flag));
	return 0;
}

3396
int ndo_dflt_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
3397
			    struct net_device *dev, u16 mode,
3398 3399 3400 3401 3402
			    u32 flags, u32 mask, int nlflags,
			    u32 filter_mask,
			    int (*vlan_fill)(struct sk_buff *skb,
					     struct net_device *dev,
					     u32 filter_mask))
3403 3404 3405 3406
{
	struct nlmsghdr *nlh;
	struct ifinfomsg *ifm;
	struct nlattr *br_afspec;
3407
	struct nlattr *protinfo;
3408
	u8 operstate = netif_running(dev) ? dev->operstate : IF_OPER_DOWN;
3409
	struct net_device *br_dev = netdev_master_upper_dev_get(dev);
3410
	int err = 0;
3411

3412
	nlh = nlmsg_put(skb, pid, seq, RTM_NEWLINK, sizeof(*ifm), nlflags);
3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427
	if (nlh == NULL)
		return -EMSGSIZE;

	ifm = nlmsg_data(nlh);
	ifm->ifi_family = AF_BRIDGE;
	ifm->__ifi_pad = 0;
	ifm->ifi_type = dev->type;
	ifm->ifi_index = dev->ifindex;
	ifm->ifi_flags = dev_get_flags(dev);
	ifm->ifi_change = 0;


	if (nla_put_string(skb, IFLA_IFNAME, dev->name) ||
	    nla_put_u32(skb, IFLA_MTU, dev->mtu) ||
	    nla_put_u8(skb, IFLA_OPERSTATE, operstate) ||
3428 3429
	    (br_dev &&
	     nla_put_u32(skb, IFLA_MASTER, br_dev->ifindex)) ||
3430 3431
	    (dev->addr_len &&
	     nla_put(skb, IFLA_ADDRESS, dev->addr_len, dev->dev_addr)) ||
3432 3433
	    (dev->ifindex != dev_get_iflink(dev) &&
	     nla_put_u32(skb, IFLA_LINK, dev_get_iflink(dev))))
3434 3435 3436 3437 3438 3439
		goto nla_put_failure;

	br_afspec = nla_nest_start(skb, IFLA_AF_SPEC);
	if (!br_afspec)
		goto nla_put_failure;

R
Roopa Prabhu 已提交
3440
	if (nla_put_u16(skb, IFLA_BRIDGE_FLAGS, BRIDGE_FLAGS_SELF)) {
3441 3442 3443
		nla_nest_cancel(skb, br_afspec);
		goto nla_put_failure;
	}
R
Roopa Prabhu 已提交
3444 3445 3446 3447 3448 3449 3450

	if (mode != BRIDGE_MODE_UNDEF) {
		if (nla_put_u16(skb, IFLA_BRIDGE_MODE, mode)) {
			nla_nest_cancel(skb, br_afspec);
			goto nla_put_failure;
		}
	}
3451 3452 3453 3454 3455 3456 3457
	if (vlan_fill) {
		err = vlan_fill(skb, dev, filter_mask);
		if (err) {
			nla_nest_cancel(skb, br_afspec);
			goto nla_put_failure;
		}
	}
3458 3459
	nla_nest_end(skb, br_afspec);

3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486
	protinfo = nla_nest_start(skb, IFLA_PROTINFO | NLA_F_NESTED);
	if (!protinfo)
		goto nla_put_failure;

	if (brport_nla_put_flag(skb, flags, mask,
				IFLA_BRPORT_MODE, BR_HAIRPIN_MODE) ||
	    brport_nla_put_flag(skb, flags, mask,
				IFLA_BRPORT_GUARD, BR_BPDU_GUARD) ||
	    brport_nla_put_flag(skb, flags, mask,
				IFLA_BRPORT_FAST_LEAVE,
				BR_MULTICAST_FAST_LEAVE) ||
	    brport_nla_put_flag(skb, flags, mask,
				IFLA_BRPORT_PROTECT, BR_ROOT_BLOCK) ||
	    brport_nla_put_flag(skb, flags, mask,
				IFLA_BRPORT_LEARNING, BR_LEARNING) ||
	    brport_nla_put_flag(skb, flags, mask,
				IFLA_BRPORT_LEARNING_SYNC, BR_LEARNING_SYNC) ||
	    brport_nla_put_flag(skb, flags, mask,
				IFLA_BRPORT_UNICAST_FLOOD, BR_FLOOD) ||
	    brport_nla_put_flag(skb, flags, mask,
				IFLA_BRPORT_PROXYARP, BR_PROXYARP)) {
		nla_nest_cancel(skb, protinfo);
		goto nla_put_failure;
	}

	nla_nest_end(skb, protinfo);

3487 3488
	nlmsg_end(skb, nlh);
	return 0;
3489 3490
nla_put_failure:
	nlmsg_cancel(skb, nlh);
3491
	return err ? err : -EMSGSIZE;
3492
}
3493
EXPORT_SYMBOL_GPL(ndo_dflt_bridge_getlink);
3494

J
John Fastabend 已提交
3495 3496 3497 3498 3499 3500 3501
static int rtnl_bridge_getlink(struct sk_buff *skb, struct netlink_callback *cb)
{
	struct net *net = sock_net(skb->sk);
	struct net_device *dev;
	int idx = 0;
	u32 portid = NETLINK_CB(cb->skb).portid;
	u32 seq = cb->nlh->nlmsg_seq;
3502
	u32 filter_mask = 0;
3503
	int err;
3504

3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516
	if (nlmsg_len(cb->nlh) > sizeof(struct ifinfomsg)) {
		struct nlattr *extfilt;

		extfilt = nlmsg_find_attr(cb->nlh, sizeof(struct ifinfomsg),
					  IFLA_EXT_MASK);
		if (extfilt) {
			if (nla_len(extfilt) < sizeof(filter_mask))
				return -EINVAL;

			filter_mask = nla_get_u32(extfilt);
		}
	}
J
John Fastabend 已提交
3517 3518 3519 3520

	rcu_read_lock();
	for_each_netdev_rcu(net, dev) {
		const struct net_device_ops *ops = dev->netdev_ops;
3521
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);
J
John Fastabend 已提交
3522

3523
		if (br_dev && br_dev->netdev_ops->ndo_bridge_getlink) {
3524 3525 3526 3527
			if (idx >= cb->args[0]) {
				err = br_dev->netdev_ops->ndo_bridge_getlink(
						skb, portid, seq, dev,
						filter_mask, NLM_F_MULTI);
3528 3529 3530 3531 3532 3533
				if (err < 0 && err != -EOPNOTSUPP) {
					if (likely(skb->len))
						break;

					goto out_err;
				}
3534
			}
3535
			idx++;
J
John Fastabend 已提交
3536 3537 3538
		}

		if (ops->ndo_bridge_getlink) {
3539 3540 3541 3542 3543
			if (idx >= cb->args[0]) {
				err = ops->ndo_bridge_getlink(skb, portid,
							      seq, dev,
							      filter_mask,
							      NLM_F_MULTI);
3544 3545 3546 3547 3548 3549
				if (err < 0 && err != -EOPNOTSUPP) {
					if (likely(skb->len))
						break;

					goto out_err;
				}
3550
			}
3551
			idx++;
J
John Fastabend 已提交
3552 3553
		}
	}
3554 3555
	err = skb->len;
out_err:
J
John Fastabend 已提交
3556 3557 3558
	rcu_read_unlock();
	cb->args[0] = idx;

3559
	return err;
J
John Fastabend 已提交
3560 3561
}

3562 3563 3564 3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576
static inline size_t bridge_nlmsg_size(void)
{
	return NLMSG_ALIGN(sizeof(struct ifinfomsg))
		+ nla_total_size(IFNAMSIZ)	/* IFLA_IFNAME */
		+ nla_total_size(MAX_ADDR_LEN)	/* IFLA_ADDRESS */
		+ nla_total_size(sizeof(u32))	/* IFLA_MASTER */
		+ nla_total_size(sizeof(u32))	/* IFLA_MTU */
		+ nla_total_size(sizeof(u32))	/* IFLA_LINK */
		+ nla_total_size(sizeof(u32))	/* IFLA_OPERSTATE */
		+ nla_total_size(sizeof(u8))	/* IFLA_PROTINFO */
		+ nla_total_size(sizeof(struct nlattr))	/* IFLA_AF_SPEC */
		+ nla_total_size(sizeof(u16))	/* IFLA_BRIDGE_FLAGS */
		+ nla_total_size(sizeof(u16));	/* IFLA_BRIDGE_MODE */
}

3577
static int rtnl_bridge_notify(struct net_device *dev)
3578 3579 3580 3581 3582
{
	struct net *net = dev_net(dev);
	struct sk_buff *skb;
	int err = -EOPNOTSUPP;

3583 3584 3585
	if (!dev->netdev_ops->ndo_bridge_getlink)
		return 0;

3586 3587 3588 3589 3590 3591
	skb = nlmsg_new(bridge_nlmsg_size(), GFP_ATOMIC);
	if (!skb) {
		err = -ENOMEM;
		goto errout;
	}

3592
	err = dev->netdev_ops->ndo_bridge_getlink(skb, 0, 0, dev, 0, 0);
3593 3594
	if (err < 0)
		goto errout;
3595

3596 3597 3598
	if (!skb->len)
		goto errout;

3599 3600 3601 3602 3603
	rtnl_notify(skb, net, 0, RTNLGRP_LINK, NULL, GFP_ATOMIC);
	return 0;
errout:
	WARN_ON(err == -EMSGSIZE);
	kfree_skb(skb);
3604 3605
	if (err)
		rtnl_set_sk_err(net, RTNLGRP_LINK, err);
3606 3607 3608
	return err;
}

3609 3610
static int rtnl_bridge_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			       struct netlink_ext_ack *extack)
J
John Fastabend 已提交
3611 3612 3613 3614
{
	struct net *net = sock_net(skb->sk);
	struct ifinfomsg *ifm;
	struct net_device *dev;
3615 3616
	struct nlattr *br_spec, *attr = NULL;
	int rem, err = -EOPNOTSUPP;
3617
	u16 flags = 0;
3618
	bool have_flags = false;
J
John Fastabend 已提交
3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632

	if (nlmsg_len(nlh) < sizeof(*ifm))
		return -EINVAL;

	ifm = nlmsg_data(nlh);
	if (ifm->ifi_family != AF_BRIDGE)
		return -EPFNOSUPPORT;

	dev = __dev_get_by_index(net, ifm->ifi_index);
	if (!dev) {
		pr_info("PF_BRIDGE: RTM_SETLINK with unknown ifindex\n");
		return -ENODEV;
	}

3633 3634 3635 3636
	br_spec = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), IFLA_AF_SPEC);
	if (br_spec) {
		nla_for_each_nested(attr, br_spec, rem) {
			if (nla_type(attr) == IFLA_BRIDGE_FLAGS) {
3637 3638 3639
				if (nla_len(attr) < sizeof(flags))
					return -EINVAL;

3640
				have_flags = true;
3641 3642 3643 3644 3645 3646 3647
				flags = nla_get_u16(attr);
				break;
			}
		}
	}

	if (!flags || (flags & BRIDGE_FLAGS_MASTER)) {
3648 3649 3650
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);

		if (!br_dev || !br_dev->netdev_ops->ndo_bridge_setlink) {
3651 3652 3653 3654
			err = -EOPNOTSUPP;
			goto out;
		}

3655
		err = br_dev->netdev_ops->ndo_bridge_setlink(dev, nlh, flags);
J
John Fastabend 已提交
3656 3657
		if (err)
			goto out;
3658 3659

		flags &= ~BRIDGE_FLAGS_MASTER;
J
John Fastabend 已提交
3660 3661
	}

3662 3663 3664 3665
	if ((flags & BRIDGE_FLAGS_SELF)) {
		if (!dev->netdev_ops->ndo_bridge_setlink)
			err = -EOPNOTSUPP;
		else
3666 3667
			err = dev->netdev_ops->ndo_bridge_setlink(dev, nlh,
								  flags);
3668
		if (!err) {
3669
			flags &= ~BRIDGE_FLAGS_SELF;
3670 3671 3672 3673 3674 3675

			/* Generate event to notify upper layer of bridge
			 * change
			 */
			err = rtnl_bridge_notify(dev);
		}
3676
	}
J
John Fastabend 已提交
3677

3678
	if (have_flags)
3679
		memcpy(nla_data(attr), &flags, sizeof(flags));
J
John Fastabend 已提交
3680 3681 3682 3683
out:
	return err;
}

3684 3685
static int rtnl_bridge_dellink(struct sk_buff *skb, struct nlmsghdr *nlh,
			       struct netlink_ext_ack *extack)
3686 3687 3688 3689 3690 3691
{
	struct net *net = sock_net(skb->sk);
	struct ifinfomsg *ifm;
	struct net_device *dev;
	struct nlattr *br_spec, *attr = NULL;
	int rem, err = -EOPNOTSUPP;
3692
	u16 flags = 0;
3693 3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711
	bool have_flags = false;

	if (nlmsg_len(nlh) < sizeof(*ifm))
		return -EINVAL;

	ifm = nlmsg_data(nlh);
	if (ifm->ifi_family != AF_BRIDGE)
		return -EPFNOSUPPORT;

	dev = __dev_get_by_index(net, ifm->ifi_index);
	if (!dev) {
		pr_info("PF_BRIDGE: RTM_SETLINK with unknown ifindex\n");
		return -ENODEV;
	}

	br_spec = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), IFLA_AF_SPEC);
	if (br_spec) {
		nla_for_each_nested(attr, br_spec, rem) {
			if (nla_type(attr) == IFLA_BRIDGE_FLAGS) {
3712 3713 3714
				if (nla_len(attr) < sizeof(flags))
					return -EINVAL;

3715 3716 3717 3718 3719 3720 3721 3722 3723 3724 3725 3726 3727 3728 3729
				have_flags = true;
				flags = nla_get_u16(attr);
				break;
			}
		}
	}

	if (!flags || (flags & BRIDGE_FLAGS_MASTER)) {
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);

		if (!br_dev || !br_dev->netdev_ops->ndo_bridge_dellink) {
			err = -EOPNOTSUPP;
			goto out;
		}

3730
		err = br_dev->netdev_ops->ndo_bridge_dellink(dev, nlh, flags);
3731 3732 3733 3734 3735 3736 3737 3738 3739 3740
		if (err)
			goto out;

		flags &= ~BRIDGE_FLAGS_MASTER;
	}

	if ((flags & BRIDGE_FLAGS_SELF)) {
		if (!dev->netdev_ops->ndo_bridge_dellink)
			err = -EOPNOTSUPP;
		else
3741 3742
			err = dev->netdev_ops->ndo_bridge_dellink(dev, nlh,
								  flags);
3743

3744
		if (!err) {
3745
			flags &= ~BRIDGE_FLAGS_SELF;
3746 3747 3748 3749 3750 3751

			/* Generate event to notify upper layer of bridge
			 * change
			 */
			err = rtnl_bridge_notify(dev);
		}
3752 3753 3754 3755 3756 3757 3758 3759
	}

	if (have_flags)
		memcpy(nla_data(attr), &flags, sizeof(flags));
out:
	return err;
}

3760 3761 3762 3763 3764 3765
static bool stats_attr_valid(unsigned int mask, int attrid, int idxattr)
{
	return (mask & IFLA_STATS_FILTER_BIT(attrid)) &&
	       (!idxattr || idxattr == attrid);
}

3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797
#define IFLA_OFFLOAD_XSTATS_FIRST (IFLA_OFFLOAD_XSTATS_UNSPEC + 1)
static int rtnl_get_offload_stats_attr_size(int attr_id)
{
	switch (attr_id) {
	case IFLA_OFFLOAD_XSTATS_CPU_HIT:
		return sizeof(struct rtnl_link_stats64);
	}

	return 0;
}

static int rtnl_get_offload_stats(struct sk_buff *skb, struct net_device *dev,
				  int *prividx)
{
	struct nlattr *attr = NULL;
	int attr_id, size;
	void *attr_data;
	int err;

	if (!(dev->netdev_ops && dev->netdev_ops->ndo_has_offload_stats &&
	      dev->netdev_ops->ndo_get_offload_stats))
		return -ENODATA;

	for (attr_id = IFLA_OFFLOAD_XSTATS_FIRST;
	     attr_id <= IFLA_OFFLOAD_XSTATS_MAX; attr_id++) {
		if (attr_id < *prividx)
			continue;

		size = rtnl_get_offload_stats_attr_size(attr_id);
		if (!size)
			continue;

3798
		if (!dev->netdev_ops->ndo_has_offload_stats(dev, attr_id))
3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838
			continue;

		attr = nla_reserve_64bit(skb, attr_id, size,
					 IFLA_OFFLOAD_XSTATS_UNSPEC);
		if (!attr)
			goto nla_put_failure;

		attr_data = nla_data(attr);
		memset(attr_data, 0, size);
		err = dev->netdev_ops->ndo_get_offload_stats(attr_id, dev,
							     attr_data);
		if (err)
			goto get_offload_stats_failure;
	}

	if (!attr)
		return -ENODATA;

	*prividx = 0;
	return 0;

nla_put_failure:
	err = -EMSGSIZE;
get_offload_stats_failure:
	*prividx = attr_id;
	return err;
}

static int rtnl_get_offload_stats_size(const struct net_device *dev)
{
	int nla_size = 0;
	int attr_id;
	int size;

	if (!(dev->netdev_ops && dev->netdev_ops->ndo_has_offload_stats &&
	      dev->netdev_ops->ndo_get_offload_stats))
		return 0;

	for (attr_id = IFLA_OFFLOAD_XSTATS_FIRST;
	     attr_id <= IFLA_OFFLOAD_XSTATS_MAX; attr_id++) {
3839
		if (!dev->netdev_ops->ndo_has_offload_stats(dev, attr_id))
3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850
			continue;
		size = rtnl_get_offload_stats_attr_size(attr_id);
		nla_size += nla_total_size_64bit(size);
	}

	if (nla_size != 0)
		nla_size += nla_total_size(0);

	return nla_size;
}

3851 3852
static int rtnl_fill_statsinfo(struct sk_buff *skb, struct net_device *dev,
			       int type, u32 pid, u32 seq, u32 change,
3853 3854
			       unsigned int flags, unsigned int filter_mask,
			       int *idxattr, int *prividx)
3855 3856 3857 3858
{
	struct if_stats_msg *ifsm;
	struct nlmsghdr *nlh;
	struct nlattr *attr;
3859
	int s_prividx = *prividx;
3860
	int err;
3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871

	ASSERT_RTNL();

	nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ifsm), flags);
	if (!nlh)
		return -EMSGSIZE;

	ifsm = nlmsg_data(nlh);
	ifsm->ifindex = dev->ifindex;
	ifsm->filter_mask = filter_mask;

3872
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_64, *idxattr)) {
3873 3874
		struct rtnl_link_stats64 *sp;

3875 3876 3877
		attr = nla_reserve_64bit(skb, IFLA_STATS_LINK_64,
					 sizeof(struct rtnl_link_stats64),
					 IFLA_STATS_UNSPEC);
3878 3879 3880 3881 3882 3883 3884
		if (!attr)
			goto nla_put_failure;

		sp = nla_data(attr);
		dev_get_stats(dev, sp);
	}

3885 3886 3887 3888 3889 3890 3891 3892 3893 3894
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_XSTATS, *idxattr)) {
		const struct rtnl_link_ops *ops = dev->rtnl_link_ops;

		if (ops && ops->fill_linkxstats) {
			*idxattr = IFLA_STATS_LINK_XSTATS;
			attr = nla_nest_start(skb,
					      IFLA_STATS_LINK_XSTATS);
			if (!attr)
				goto nla_put_failure;

3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906 3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918
			err = ops->fill_linkxstats(skb, dev, prividx, *idxattr);
			nla_nest_end(skb, attr);
			if (err)
				goto nla_put_failure;
			*idxattr = 0;
		}
	}

	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_XSTATS_SLAVE,
			     *idxattr)) {
		const struct rtnl_link_ops *ops = NULL;
		const struct net_device *master;

		master = netdev_master_upper_dev_get(dev);
		if (master)
			ops = master->rtnl_link_ops;
		if (ops && ops->fill_linkxstats) {
			*idxattr = IFLA_STATS_LINK_XSTATS_SLAVE;
			attr = nla_nest_start(skb,
					      IFLA_STATS_LINK_XSTATS_SLAVE);
			if (!attr)
				goto nla_put_failure;

			err = ops->fill_linkxstats(skb, dev, prividx, *idxattr);
3919 3920 3921 3922 3923 3924 3925
			nla_nest_end(skb, attr);
			if (err)
				goto nla_put_failure;
			*idxattr = 0;
		}
	}

3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_OFFLOAD_XSTATS,
			     *idxattr)) {
		*idxattr = IFLA_STATS_LINK_OFFLOAD_XSTATS;
		attr = nla_nest_start(skb, IFLA_STATS_LINK_OFFLOAD_XSTATS);
		if (!attr)
			goto nla_put_failure;

		err = rtnl_get_offload_stats(skb, dev, prividx);
		if (err == -ENODATA)
			nla_nest_cancel(skb, attr);
		else
			nla_nest_end(skb, attr);

		if (err && err != -ENODATA)
			goto nla_put_failure;
		*idxattr = 0;
	}

3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976
	if (stats_attr_valid(filter_mask, IFLA_STATS_AF_SPEC, *idxattr)) {
		struct rtnl_af_ops *af_ops;

		*idxattr = IFLA_STATS_AF_SPEC;
		attr = nla_nest_start(skb, IFLA_STATS_AF_SPEC);
		if (!attr)
			goto nla_put_failure;

		list_for_each_entry(af_ops, &rtnl_af_ops, list) {
			if (af_ops->fill_stats_af) {
				struct nlattr *af;
				int err;

				af = nla_nest_start(skb, af_ops->family);
				if (!af)
					goto nla_put_failure;

				err = af_ops->fill_stats_af(skb, dev);

				if (err == -ENODATA)
					nla_nest_cancel(skb, af);
				else if (err < 0)
					goto nla_put_failure;

				nla_nest_end(skb, af);
			}
		}

		nla_nest_end(skb, attr);

		*idxattr = 0;
	}

3977 3978 3979 3980 3981
	nlmsg_end(skb, nlh);

	return 0;

nla_put_failure:
3982 3983 3984 3985 3986
	/* not a multi message or no progress mean a real error */
	if (!(flags & NLM_F_MULTI) || s_prividx == *prividx)
		nlmsg_cancel(skb, nlh);
	else
		nlmsg_end(skb, nlh);
3987 3988 3989 3990 3991 3992 3993 3994 3995

	return -EMSGSIZE;
}

static size_t if_nlmsg_stats_size(const struct net_device *dev,
				  u32 filter_mask)
{
	size_t size = 0;

3996
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_64, 0))
3997 3998
		size += nla_total_size_64bit(sizeof(struct rtnl_link_stats64));

3999 4000
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_XSTATS, 0)) {
		const struct rtnl_link_ops *ops = dev->rtnl_link_ops;
4001
		int attr = IFLA_STATS_LINK_XSTATS;
4002 4003

		if (ops && ops->get_linkxstats_size) {
4004 4005
			size += nla_total_size(ops->get_linkxstats_size(dev,
									attr));
4006 4007 4008 4009 4010
			/* for IFLA_STATS_LINK_XSTATS */
			size += nla_total_size(0);
		}
	}

4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_XSTATS_SLAVE, 0)) {
		struct net_device *_dev = (struct net_device *)dev;
		const struct rtnl_link_ops *ops = NULL;
		const struct net_device *master;

		/* netdev_master_upper_dev_get can't take const */
		master = netdev_master_upper_dev_get(_dev);
		if (master)
			ops = master->rtnl_link_ops;
		if (ops && ops->get_linkxstats_size) {
			int attr = IFLA_STATS_LINK_XSTATS_SLAVE;

			size += nla_total_size(ops->get_linkxstats_size(dev,
									attr));
			/* for IFLA_STATS_LINK_XSTATS_SLAVE */
			size += nla_total_size(0);
		}
	}

4030 4031 4032
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_OFFLOAD_XSTATS, 0))
		size += rtnl_get_offload_stats_size(dev);

4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 4045 4046 4047 4048 4049
	if (stats_attr_valid(filter_mask, IFLA_STATS_AF_SPEC, 0)) {
		struct rtnl_af_ops *af_ops;

		/* for IFLA_STATS_AF_SPEC */
		size += nla_total_size(0);

		list_for_each_entry(af_ops, &rtnl_af_ops, list) {
			if (af_ops->get_stats_af_size) {
				size += nla_total_size(
					af_ops->get_stats_af_size(dev));

				/* for AF_* */
				size += nla_total_size(0);
			}
		}
	}

4050 4051 4052
	return size;
}

4053 4054
static int rtnl_stats_get(struct sk_buff *skb, struct nlmsghdr *nlh,
			  struct netlink_ext_ack *extack)
4055 4056 4057
{
	struct net *net = sock_net(skb->sk);
	struct net_device *dev = NULL;
4058 4059
	int idxattr = 0, prividx = 0;
	struct if_stats_msg *ifsm;
4060 4061 4062 4063
	struct sk_buff *nskb;
	u32 filter_mask;
	int err;

4064 4065 4066
	if (nlmsg_len(nlh) < sizeof(*ifsm))
		return -EINVAL;

4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085
	ifsm = nlmsg_data(nlh);
	if (ifsm->ifindex > 0)
		dev = __dev_get_by_index(net, ifsm->ifindex);
	else
		return -EINVAL;

	if (!dev)
		return -ENODEV;

	filter_mask = ifsm->filter_mask;
	if (!filter_mask)
		return -EINVAL;

	nskb = nlmsg_new(if_nlmsg_stats_size(dev, filter_mask), GFP_KERNEL);
	if (!nskb)
		return -ENOBUFS;

	err = rtnl_fill_statsinfo(nskb, dev, RTM_NEWSTATS,
				  NETLINK_CB(skb).portid, nlh->nlmsg_seq, 0,
4086
				  0, filter_mask, &idxattr, &prividx);
4087 4088 4089 4090 4091 4092 4093 4094 4095 4096 4097 4098 4099
	if (err < 0) {
		/* -EMSGSIZE implies BUG in if_nlmsg_stats_size */
		WARN_ON(err == -EMSGSIZE);
		kfree_skb(nskb);
	} else {
		err = rtnl_unicast(nskb, net, NETLINK_CB(skb).portid);
	}

	return err;
}

static int rtnl_stats_dump(struct sk_buff *skb, struct netlink_callback *cb)
{
4100
	int h, s_h, err, s_idx, s_idxattr, s_prividx;
4101
	struct net *net = sock_net(skb->sk);
4102
	unsigned int flags = NLM_F_MULTI;
4103 4104
	struct if_stats_msg *ifsm;
	struct hlist_head *head;
4105
	struct net_device *dev;
4106
	u32 filter_mask = 0;
4107
	int idx = 0;
4108 4109 4110

	s_h = cb->args[0];
	s_idx = cb->args[1];
4111 4112
	s_idxattr = cb->args[2];
	s_prividx = cb->args[3];
4113 4114 4115

	cb->seq = net->dev_base_seq;

4116 4117 4118
	if (nlmsg_len(cb->nlh) < sizeof(*ifsm))
		return -EINVAL;

4119 4120 4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132
	ifsm = nlmsg_data(cb->nlh);
	filter_mask = ifsm->filter_mask;
	if (!filter_mask)
		return -EINVAL;

	for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) {
		idx = 0;
		head = &net->dev_index_head[h];
		hlist_for_each_entry(dev, head, index_hlist) {
			if (idx < s_idx)
				goto cont;
			err = rtnl_fill_statsinfo(skb, dev, RTM_NEWSTATS,
						  NETLINK_CB(cb->skb).portid,
						  cb->nlh->nlmsg_seq, 0,
4133 4134
						  flags, filter_mask,
						  &s_idxattr, &s_prividx);
4135 4136 4137 4138 4139 4140 4141
			/* If we ran out of room on the first message,
			 * we're in trouble
			 */
			WARN_ON((err == -EMSGSIZE) && (skb->len == 0));

			if (err < 0)
				goto out;
4142 4143
			s_prividx = 0;
			s_idxattr = 0;
4144 4145 4146 4147 4148 4149
			nl_dump_check_consistent(cb, nlmsg_hdr(skb));
cont:
			idx++;
		}
	}
out:
4150 4151
	cb->args[3] = s_prividx;
	cb->args[2] = s_idxattr;
4152 4153 4154 4155 4156 4157
	cb->args[1] = idx;
	cb->args[0] = h;

	return skb->len;
}

L
Linus Torvalds 已提交
4158 4159
/* Process one rtnetlink message. */

J
Johannes Berg 已提交
4160 4161
static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
			     struct netlink_ext_ack *extack)
L
Linus Torvalds 已提交
4162
{
4163
	struct net *net = sock_net(skb->sk);
4164
	rtnl_doit_func doit;
4165
	int kind;
L
Linus Torvalds 已提交
4166 4167
	int family;
	int type;
4168
	int err;
L
Linus Torvalds 已提交
4169 4170 4171

	type = nlh->nlmsg_type;
	if (type > RTM_MAX)
4172
		return -EOPNOTSUPP;
L
Linus Torvalds 已提交
4173 4174 4175 4176

	type -= RTM_BASE;

	/* All the messages must have at least 1 byte length */
4177
	if (nlmsg_len(nlh) < sizeof(struct rtgenmsg))
L
Linus Torvalds 已提交
4178 4179
		return 0;

4180
	family = ((struct rtgenmsg *)nlmsg_data(nlh))->rtgen_family;
L
Linus Torvalds 已提交
4181 4182
	kind = type&3;

4183
	if (kind != 2 && !netlink_net_capable(skb, CAP_NET_ADMIN))
4184
		return -EPERM;
L
Linus Torvalds 已提交
4185

4186
	if (kind == 2 && nlh->nlmsg_flags&NLM_F_DUMP) {
4187
		struct sock *rtnl;
4188
		rtnl_dumpit_func dumpit;
4189 4190
		rtnl_calcit_func calcit;
		u16 min_dump_alloc = 0;
L
Linus Torvalds 已提交
4191

4192 4193
		dumpit = rtnl_get_dumpit(family, type);
		if (dumpit == NULL)
4194
			return -EOPNOTSUPP;
4195 4196
		calcit = rtnl_get_calcit(family, type);
		if (calcit)
4197
			min_dump_alloc = calcit(skb, nlh);
4198

4199
		__rtnl_unlock();
4200
		rtnl = net->rtnl;
4201 4202 4203 4204 4205 4206 4207
		{
			struct netlink_dump_control c = {
				.dump		= dumpit,
				.min_dump_alloc	= min_dump_alloc,
			};
			err = netlink_dump_start(rtnl, skb, nlh, &c);
		}
4208 4209
		rtnl_lock();
		return err;
L
Linus Torvalds 已提交
4210 4211
	}

4212 4213
	doit = rtnl_get_doit(family, type);
	if (doit == NULL)
4214
		return -EOPNOTSUPP;
L
Linus Torvalds 已提交
4215

4216
	return doit(skb, nlh, extack);
L
Linus Torvalds 已提交
4217 4218
}

4219
static void rtnetlink_rcv(struct sk_buff *skb)
L
Linus Torvalds 已提交
4220
{
4221 4222 4223
	rtnl_lock();
	netlink_rcv_skb(skb, &rtnetlink_rcv_msg);
	rtnl_unlock();
L
Linus Torvalds 已提交
4224 4225
}

4226 4227 4228 4229 4230 4231 4232 4233 4234 4235 4236 4237
static int rtnetlink_bind(struct net *net, int group)
{
	switch (group) {
	case RTNLGRP_IPV4_MROUTE_R:
	case RTNLGRP_IPV6_MROUTE_R:
		if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
			return -EPERM;
		break;
	}
	return 0;
}

L
Linus Torvalds 已提交
4238 4239
static int rtnetlink_event(struct notifier_block *this, unsigned long event, void *ptr)
{
4240
	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
4241

L
Linus Torvalds 已提交
4242
	switch (event) {
4243 4244 4245 4246 4247 4248 4249
	case NETDEV_REBOOT:
	case NETDEV_CHANGENAME:
	case NETDEV_FEAT_CHANGE:
	case NETDEV_BONDING_FAILOVER:
	case NETDEV_NOTIFY_PEERS:
	case NETDEV_RESEND_IGMP:
	case NETDEV_CHANGEINFODATA:
4250 4251
		rtmsg_ifinfo_event(RTM_NEWLINK, dev, 0, rtnl_get_event(event),
				   GFP_KERNEL);
L
Linus Torvalds 已提交
4252 4253 4254 4255 4256 4257 4258 4259 4260 4261 4262
		break;
	default:
		break;
	}
	return NOTIFY_DONE;
}

static struct notifier_block rtnetlink_dev_notifier = {
	.notifier_call	= rtnetlink_event,
};

4263

4264
static int __net_init rtnetlink_net_init(struct net *net)
4265 4266
{
	struct sock *sk;
4267 4268 4269 4270
	struct netlink_kernel_cfg cfg = {
		.groups		= RTNLGRP_MAX,
		.input		= rtnetlink_rcv,
		.cb_mutex	= &rtnl_mutex,
4271
		.flags		= NL_CFG_F_NONROOT_RECV,
4272
		.bind		= rtnetlink_bind,
4273 4274
	};

4275
	sk = netlink_kernel_create(net, NETLINK_ROUTE, &cfg);
4276 4277 4278 4279 4280 4281
	if (!sk)
		return -ENOMEM;
	net->rtnl = sk;
	return 0;
}

4282
static void __net_exit rtnetlink_net_exit(struct net *net)
4283
{
4284 4285
	netlink_kernel_release(net->rtnl);
	net->rtnl = NULL;
4286 4287 4288 4289 4290 4291 4292
}

static struct pernet_operations rtnetlink_net_ops = {
	.init = rtnetlink_net_init,
	.exit = rtnetlink_net_exit,
};

L
Linus Torvalds 已提交
4293 4294
void __init rtnetlink_init(void)
{
4295
	if (register_pernet_subsys(&rtnetlink_net_ops))
L
Linus Torvalds 已提交
4296
		panic("rtnetlink_init: cannot initialize rtnetlink\n");
4297

L
Linus Torvalds 已提交
4298
	register_netdevice_notifier(&rtnetlink_dev_notifier);
4299

4300 4301 4302 4303 4304
	rtnl_register(PF_UNSPEC, RTM_GETLINK, rtnl_getlink,
		      rtnl_dump_ifinfo, rtnl_calcit);
	rtnl_register(PF_UNSPEC, RTM_SETLINK, rtnl_setlink, NULL, NULL);
	rtnl_register(PF_UNSPEC, RTM_NEWLINK, rtnl_newlink, NULL, NULL);
	rtnl_register(PF_UNSPEC, RTM_DELLINK, rtnl_dellink, NULL, NULL);
4305

4306 4307
	rtnl_register(PF_UNSPEC, RTM_GETADDR, NULL, rtnl_dump_all, NULL);
	rtnl_register(PF_UNSPEC, RTM_GETROUTE, NULL, rtnl_dump_all, NULL);
4308
	rtnl_register(PF_UNSPEC, RTM_GETNETCONF, NULL, rtnl_dump_all, NULL);
4309 4310 4311 4312

	rtnl_register(PF_BRIDGE, RTM_NEWNEIGH, rtnl_fdb_add, NULL, NULL);
	rtnl_register(PF_BRIDGE, RTM_DELNEIGH, rtnl_fdb_del, NULL, NULL);
	rtnl_register(PF_BRIDGE, RTM_GETNEIGH, NULL, rtnl_fdb_dump, NULL);
J
John Fastabend 已提交
4313 4314

	rtnl_register(PF_BRIDGE, RTM_GETLINK, NULL, rtnl_bridge_getlink, NULL);
4315
	rtnl_register(PF_BRIDGE, RTM_DELLINK, rtnl_bridge_dellink, NULL, NULL);
J
John Fastabend 已提交
4316
	rtnl_register(PF_BRIDGE, RTM_SETLINK, rtnl_bridge_setlink, NULL, NULL);
4317 4318 4319

	rtnl_register(PF_UNSPEC, RTM_GETSTATS, rtnl_stats_get, rtnl_stats_dump,
		      NULL);
L
Linus Torvalds 已提交
4320
}