rtnetlink.c 104.9 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
	unsigned int		flags;
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 __rcu *rtnl_msg_handlers[RTNL_FAMILY_MAX + 1];
131
static refcount_t rtnl_msg_handlers_ref[RTNL_FAMILY_MAX + 1];
132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152

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;
}

/**
 * __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
153
 * @flags: rtnl_link_flags to modifiy behaviour of doit/dumpit functions
154 155 156 157 158 159 160 161 162 163 164 165
 *
 * 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,
166
		    rtnl_doit_func doit, rtnl_dumpit_func dumpit,
167
		    unsigned int flags)
168 169 170 171
{
	struct rtnl_link *tab;
	int msgindex;

172
	BUG_ON(protocol < 0 || protocol > RTNL_FAMILY_MAX);
173 174
	msgindex = rtm_msgindex(msgtype);

175
	tab = rcu_dereference_raw(rtnl_msg_handlers[protocol]);
176 177 178 179 180
	if (tab == NULL) {
		tab = kcalloc(RTM_NR_MSGTYPES, sizeof(*tab), GFP_KERNEL);
		if (tab == NULL)
			return -ENOBUFS;

181
		rcu_assign_pointer(rtnl_msg_handlers[protocol], tab);
182 183 184 185 186 187
	}

	if (doit)
		tab[msgindex].doit = doit;
	if (dumpit)
		tab[msgindex].dumpit = dumpit;
188
	tab[msgindex].flags |= flags;
189 190 191 192 193 194 195 196 197 198 199 200

	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
201
 * of memory implies no sense in continuing.
202 203
 */
void rtnl_register(int protocol, int msgtype,
204
		   rtnl_doit_func doit, rtnl_dumpit_func dumpit,
205
		   unsigned int flags)
206
{
207
	if (__rtnl_register(protocol, msgtype, doit, dumpit, flags) < 0)
208 209 210 211 212 213 214 215 216 217 218 219 220 221 222
		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)
{
223
	struct rtnl_link *handlers;
224 225
	int msgindex;

226
	BUG_ON(protocol < 0 || protocol > RTNL_FAMILY_MAX);
227 228
	msgindex = rtm_msgindex(msgtype);

229 230 231 232
	rtnl_lock();
	handlers = rtnl_dereference(rtnl_msg_handlers[protocol]);
	if (!handlers) {
		rtnl_unlock();
233
		return -ENOENT;
234
	}
235

236 237
	handlers[msgindex].doit = NULL;
	handlers[msgindex].dumpit = NULL;
238
	handlers[msgindex].flags = 0;
239
	rtnl_unlock();
240 241 242 243 244 245 246 247 248 249 250 251 252 253

	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)
{
254 255
	struct rtnl_link *handlers;

256
	BUG_ON(protocol < 0 || protocol > RTNL_FAMILY_MAX);
257

258
	rtnl_lock();
259 260
	handlers = rtnl_dereference(rtnl_msg_handlers[protocol]);
	RCU_INIT_POINTER(rtnl_msg_handlers[protocol], NULL);
261 262
	rtnl_unlock();

263 264
	synchronize_net();

265
	while (refcount_read(&rtnl_msg_handlers_ref[protocol]) > 1)
266 267
		schedule();
	kfree(handlers);
268 269
}
EXPORT_SYMBOL_GPL(rtnl_unregister_all);
L
Linus Torvalds 已提交
270

P
Patrick McHardy 已提交
271 272
static LIST_HEAD(link_ops);

273 274 275 276 277 278 279 280 281 282 283
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
284 285 286 287 288 289 290 291 292 293 294 295
/**
 * __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)
{
296 297 298
	if (rtnl_link_ops_get(ops->kind))
		return -EEXIST;

299 300 301 302 303 304
	/* 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)
305
		ops->dellink = unregister_netdevice_queue;
306

P
Patrick McHardy 已提交
307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328
	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);

329 330 331
static void __rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops)
{
	struct net_device *dev;
332 333
	LIST_HEAD(list_kill);

334
	for_each_netdev(net, dev) {
335 336
		if (dev->rtnl_link_ops == ops)
			ops->dellink(dev, &list_kill);
337
	}
338
	unregister_netdevice_many(&list_kill);
339 340
}

P
Patrick McHardy 已提交
341 342 343 344
/**
 * __rtnl_link_unregister - Unregister rtnl_link_ops from rtnetlink.
 * @ops: struct rtnl_link_ops * to unregister
 *
345
 * The caller must hold the rtnl_mutex.
P
Patrick McHardy 已提交
346 347 348
 */
void __rtnl_link_unregister(struct rtnl_link_ops *ops)
{
349
	struct net *net;
350

351
	for_each_net(net) {
352
		__rtnl_kill_links(net, ops);
353
	}
P
Patrick McHardy 已提交
354 355 356 357
	list_del(&ops->list);
}
EXPORT_SYMBOL_GPL(__rtnl_link_unregister);

358 359 360 361 362 363 364
/* 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;
365
	DEFINE_WAIT_FUNC(wait, woken_wake_function);
366

367
	add_wait_queue(&netdev_unregistering_wq, &wait);
368 369 370 371 372 373 374 375 376 377 378 379
	for (;;) {
		unregistering = false;
		rtnl_lock();
		for_each_net(net) {
			if (net->dev_unreg_count > 0) {
				unregistering = true;
				break;
			}
		}
		if (!unregistering)
			break;
		__rtnl_unlock();
380 381

		wait_woken(&wait, TASK_UNINTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT);
382
	}
383
	remove_wait_queue(&netdev_unregistering_wq, &wait);
384 385
}

P
Patrick McHardy 已提交
386 387 388 389 390 391
/**
 * 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)
{
392 393 394
	/* Close the race with cleanup_net() */
	mutex_lock(&net_mutex);
	rtnl_lock_unregistering_all();
P
Patrick McHardy 已提交
395 396
	__rtnl_link_unregister(ops);
	rtnl_unlock();
397
	mutex_unlock(&net_mutex);
P
Patrick McHardy 已提交
398 399 400
}
EXPORT_SYMBOL_GPL(rtnl_link_unregister);

401 402 403 404
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;
405
	size_t size = 0;
406

407 408 409
	rcu_read_lock();

	master_dev = netdev_master_upper_dev_get_rcu((struct net_device *)dev);
410
	if (!master_dev)
411 412
		goto out;

413
	ops = master_dev->rtnl_link_ops;
414
	if (!ops || !ops->get_slave_size)
415
		goto out;
416
	/* IFLA_INFO_SLAVE_DATA + nested data */
417
	size = nla_total_size(sizeof(struct nlattr)) +
418
	       ops->get_slave_size(master_dev, dev);
419 420 421 422

out:
	rcu_read_unlock();
	return size;
423 424
}

P
Patrick McHardy 已提交
425 426 427 428 429 430 431 432
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;

433 434
	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
435 436 437

	if (ops->get_size)
		/* IFLA_INFO_DATA + nested data */
438
		size += nla_total_size(sizeof(struct nlattr)) +
P
Patrick McHardy 已提交
439 440 441
			ops->get_size(dev);

	if (ops->get_xstats_size)
442 443
		/* IFLA_INFO_XSTATS */
		size += nla_total_size(ops->get_xstats_size(dev));
P
Patrick McHardy 已提交
444

445 446
	size += rtnl_link_get_slave_info_data_size(dev);

P
Patrick McHardy 已提交
447 448 449
	return size;
}

T
Thomas Graf 已提交
450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469
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.
 */
470
void rtnl_af_register(struct rtnl_af_ops *ops)
T
Thomas Graf 已提交
471 472
{
	rtnl_lock();
473
	list_add_tail(&ops->list, &rtnl_af_ops);
T
Thomas Graf 已提交
474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501
	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);

502 503
static size_t rtnl_link_get_af_size(const struct net_device *dev,
				    u32 ext_filter_mask)
T
Thomas Graf 已提交
504 505 506 507 508 509 510 511 512 513 514
{
	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)) +
515
				af_ops->get_link_af_size(dev, ext_filter_mask);
T
Thomas Graf 已提交
516 517 518 519 520 521
		}
	}

	return size;
}

522
static bool rtnl_have_link_slave_info(const struct net_device *dev)
P
Patrick McHardy 已提交
523
{
524
	struct net_device *master_dev;
P
Patrick McHardy 已提交
525

526
	master_dev = netdev_master_upper_dev_get((struct net_device *) dev);
527
	if (master_dev && master_dev->rtnl_link_ops)
528 529 530 531 532 533 534 535 536 537 538
		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
539

540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572
	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
573
	if (nla_put_string(skb, IFLA_INFO_KIND, ops->kind) < 0)
574
		return -EMSGSIZE;
P
Patrick McHardy 已提交
575 576 577
	if (ops->fill_xstats) {
		err = ops->fill_xstats(skb, dev);
		if (err < 0)
578
			return err;
P
Patrick McHardy 已提交
579 580 581
	}
	if (ops->fill_info) {
		data = nla_nest_start(skb, IFLA_INFO_DATA);
582 583
		if (data == NULL)
			return -EMSGSIZE;
P
Patrick McHardy 已提交
584 585 586 587 588 589 590 591 592
		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);
593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615
	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
616 617 618 619 620 621
err_cancel_link:
	nla_nest_cancel(skb, linkinfo);
out:
	return err;
}

622
int rtnetlink_send(struct sk_buff *skb, struct net *net, u32 pid, unsigned int group, int echo)
L
Linus Torvalds 已提交
623
{
624
	struct sock *rtnl = net->rtnl;
L
Linus Torvalds 已提交
625 626
	int err = 0;

627
	NETLINK_CB(skb).dst_group = group;
L
Linus Torvalds 已提交
628
	if (echo)
629
		refcount_inc(&skb->users);
L
Linus Torvalds 已提交
630 631 632 633 634 635
	netlink_broadcast(rtnl, skb, pid, group, GFP_KERNEL);
	if (echo)
		err = netlink_unicast(rtnl, skb, pid, MSG_DONTWAIT);
	return err;
}

636
int rtnl_unicast(struct sk_buff *skb, struct net *net, u32 pid)
637
{
638 639
	struct sock *rtnl = net->rtnl;

640 641
	return nlmsg_unicast(rtnl, skb, pid);
}
E
Eric Dumazet 已提交
642
EXPORT_SYMBOL(rtnl_unicast);
643

644 645
void rtnl_notify(struct sk_buff *skb, struct net *net, u32 pid, u32 group,
		 struct nlmsghdr *nlh, gfp_t flags)
646
{
647
	struct sock *rtnl = net->rtnl;
648 649 650 651 652
	int report = 0;

	if (nlh)
		report = nlmsg_report(nlh);

653
	nlmsg_notify(rtnl, skb, pid, group, report, flags);
654
}
E
Eric Dumazet 已提交
655
EXPORT_SYMBOL(rtnl_notify);
656

657
void rtnl_set_sk_err(struct net *net, u32 group, int error)
658
{
659 660
	struct sock *rtnl = net->rtnl;

661 662
	netlink_set_err(rtnl, 0, group, error);
}
E
Eric Dumazet 已提交
663
EXPORT_SYMBOL(rtnl_set_sk_err);
664

L
Linus Torvalds 已提交
665 666
int rtnetlink_put_metrics(struct sk_buff *skb, u32 *metrics)
{
667 668 669 670 671 672 673 674 675
	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]) {
676 677 678 679 680 681 682 683
			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;
684 685 686
			} else if (i == RTAX_FEATURES - 1) {
				u32 user_features = metrics[i] & RTAX_FEATURE_MASK;

687 688
				if (!user_features)
					continue;
689 690 691
				BUILD_BUG_ON(RTAX_FEATURE_MASK & DST_FEATURE_MASK);
				if (nla_put_u32(skb, i + 1, user_features))
					goto nla_put_failure;
692 693 694 695
			} else {
				if (nla_put_u32(skb, i + 1, metrics[i]))
					goto nla_put_failure;
			}
696 697
			valid++;
		}
L
Linus Torvalds 已提交
698 699
	}

700 701 702 703
	if (!valid) {
		nla_nest_cancel(skb, mx);
		return 0;
	}
704 705 706 707

	return nla_nest_end(skb, mx);

nla_put_failure:
708 709
	nla_nest_cancel(skb, mx);
	return -EMSGSIZE;
L
Linus Torvalds 已提交
710
}
E
Eric Dumazet 已提交
711
EXPORT_SYMBOL(rtnetlink_put_metrics);
L
Linus Torvalds 已提交
712

713
int rtnl_put_cacheinfo(struct sk_buff *skb, struct dst_entry *dst, u32 id,
714
		       long expires, u32 error)
715 716
{
	struct rta_cacheinfo ci = {
717
		.rta_lastuse = jiffies_delta_to_clock_t(jiffies - dst->lastuse),
718 719 720 721 722 723
		.rta_used = dst->__use,
		.rta_clntref = atomic_read(&(dst->__refcnt)),
		.rta_error = error,
		.rta_id =  id,
	};

724 725
	if (expires) {
		unsigned long clock;
726

727 728 729 730
		clock = jiffies_to_clock_t(abs(expires));
		clock = min_t(unsigned long, clock, INT_MAX);
		ci.rta_expires = (expires > 0) ? clock : -clock;
	}
731 732 733
	return nla_put(skb, RTA_CACHEINFO, sizeof(ci), &ci);
}
EXPORT_SYMBOL_GPL(rtnl_put_cacheinfo);
L
Linus Torvalds 已提交
734

735
static void set_operstate(struct net_device *dev, unsigned char transition)
S
Stefan Rompf 已提交
736 737 738
{
	unsigned char operstate = dev->operstate;

E
Eric Dumazet 已提交
739
	switch (transition) {
S
Stefan Rompf 已提交
740 741 742 743 744 745 746 747 748 749 750 751
	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;
752
	}
S
Stefan Rompf 已提交
753 754 755 756 757

	if (dev->operstate != operstate) {
		write_lock_bh(&dev_base_lock);
		dev->operstate = operstate;
		write_unlock_bh(&dev_base_lock);
758 759
		netdev_state_change(dev);
	}
S
Stefan Rompf 已提交
760 761
}

762 763 764 765 766 767
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));
}

768 769 770 771 772 773 774 775
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) |
776
			(rtnl_dev_get_flags(dev) & ~ifm->ifi_change);
777 778 779 780

	return flags;
}

781
static void copy_rtnl_link_stats(struct rtnl_link_stats *a,
782
				 const struct rtnl_link_stats64 *b)
L
Linus Torvalds 已提交
783
{
784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810
	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;
811 812

	a->rx_nohandler = b->rx_nohandler;
813 814
}

815
/* All VF info */
816 817
static inline int rtnl_vfinfo_size(const struct net_device *dev,
				   u32 ext_filter_mask)
818
{
819
	if (dev->dev.parent && (ext_filter_mask & RTEXT_FILTER_VF)) {
820
		int num_vfs = dev_num_vf(dev->dev.parent);
S
Sabrina Dubroca 已提交
821
		size_t size = nla_total_size(0);
822
		size += num_vfs *
S
Sabrina Dubroca 已提交
823 824 825 826
			(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 */
827 828
			 nla_total_size(MAX_VLAN_LIST_LEN *
					sizeof(struct ifla_vf_vlan_info)) +
829
			 nla_total_size(sizeof(struct ifla_vf_spoofchk)) +
S
Sabrina Dubroca 已提交
830
			 nla_total_size(sizeof(struct ifla_vf_tx_rate)) +
J
Jiri Benc 已提交
831
			 nla_total_size(sizeof(struct ifla_vf_rate)) +
832
			 nla_total_size(sizeof(struct ifla_vf_link_state)) +
833
			 nla_total_size(sizeof(struct ifla_vf_rss_query_en)) +
S
Sabrina Dubroca 已提交
834
			 nla_total_size(0) + /* nest IFLA_VF_STATS */
835
			 /* IFLA_VF_STATS_RX_PACKETS */
N
Nicolas Dichtel 已提交
836
			 nla_total_size_64bit(sizeof(__u64)) +
837
			 /* IFLA_VF_STATS_TX_PACKETS */
N
Nicolas Dichtel 已提交
838
			 nla_total_size_64bit(sizeof(__u64)) +
839
			 /* IFLA_VF_STATS_RX_BYTES */
N
Nicolas Dichtel 已提交
840
			 nla_total_size_64bit(sizeof(__u64)) +
841
			 /* IFLA_VF_STATS_TX_BYTES */
N
Nicolas Dichtel 已提交
842
			 nla_total_size_64bit(sizeof(__u64)) +
843
			 /* IFLA_VF_STATS_BROADCAST */
N
Nicolas Dichtel 已提交
844
			 nla_total_size_64bit(sizeof(__u64)) +
845
			 /* IFLA_VF_STATS_MULTICAST */
N
Nicolas Dichtel 已提交
846
			 nla_total_size_64bit(sizeof(__u64)) +
H
Hiroshi Shimamoto 已提交
847
			 nla_total_size(sizeof(struct ifla_vf_trust)));
848 849
		return size;
	} else
850 851 852
		return 0;
}

853 854
static size_t rtnl_port_size(const struct net_device *dev,
			     u32 ext_filter_mask)
855 856 857 858 859 860 861 862 863 864 865 866 867
{
	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;

868 869
	if (!dev->netdev_ops->ndo_get_vf_port || !dev->dev.parent ||
	    !(ext_filter_mask & RTEXT_FILTER_VF))
870 871 872 873 874 875 876 877
		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
878
static size_t rtnl_xdp_size(void)
879
{
880
	size_t xdp_size = nla_total_size(0) +	/* nest IFLA_XDP */
M
Martin KaFai Lau 已提交
881 882
			  nla_total_size(1) +	/* XDP_ATTACHED */
			  nla_total_size(4);	/* XDP_PROG_ID */
883

D
David S. Miller 已提交
884
	return xdp_size;
885 886
}

887 888
static noinline size_t if_nlmsg_size(const struct net_device *dev,
				     u32 ext_filter_mask)
889 890 891
{
	return NLMSG_ALIGN(sizeof(struct ifinfomsg))
	       + nla_total_size(IFNAMSIZ) /* IFLA_IFNAME */
892
	       + nla_total_size(IFALIASZ) /* IFLA_IFALIAS */
893
	       + nla_total_size(IFNAMSIZ) /* IFLA_QDISC */
894
	       + nla_total_size_64bit(sizeof(struct rtnl_link_ifmap))
895
	       + nla_total_size(sizeof(struct rtnl_link_stats))
896
	       + nla_total_size_64bit(sizeof(struct rtnl_link_stats64))
897 898 899 900 901 902 903
	       + 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 */
904
	       + nla_total_size(1) /* IFLA_CARRIER */
905
	       + nla_total_size(4) /* IFLA_PROMISCUITY */
906 907
	       + nla_total_size(4) /* IFLA_NUM_TX_QUEUES */
	       + nla_total_size(4) /* IFLA_NUM_RX_QUEUES */
908 909
	       + nla_total_size(4) /* IFLA_GSO_MAX_SEGS */
	       + nla_total_size(4) /* IFLA_GSO_MAX_SIZE */
910
	       + nla_total_size(1) /* IFLA_OPERSTATE */
P
Patrick McHardy 已提交
911
	       + nla_total_size(1) /* IFLA_LINKMODE */
912
	       + nla_total_size(4) /* IFLA_CARRIER_CHANGES */
913
	       + nla_total_size(4) /* IFLA_LINK_NETNSID */
914
	       + nla_total_size(4) /* IFLA_GROUP */
915 916 917
	       + nla_total_size(ext_filter_mask
			        & RTEXT_FILTER_VF ? 4 : 0) /* IFLA_NUM_VF */
	       + rtnl_vfinfo_size(dev, ext_filter_mask) /* IFLA_VFINFO_LIST */
918
	       + rtnl_port_size(dev, ext_filter_mask) /* IFLA_VF_PORTS + IFLA_PORT_SELF */
T
Thomas Graf 已提交
919
	       + rtnl_link_get_size(dev) /* IFLA_LINKINFO */
920
	       + rtnl_link_get_af_size(dev, ext_filter_mask) /* IFLA_AF_SPEC */
921
	       + nla_total_size(MAX_PHYS_ITEM_ID_LEN) /* IFLA_PHYS_PORT_ID */
922
	       + nla_total_size(MAX_PHYS_ITEM_ID_LEN) /* IFLA_PHYS_SWITCH_ID */
923
	       + nla_total_size(IFNAMSIZ) /* IFLA_PHYS_PORT_NAME */
D
David S. Miller 已提交
924
	       + rtnl_xdp_size() /* IFLA_XDP */
925
	       + nla_total_size(4)  /* IFLA_EVENT */
926 927
	       + nla_total_size(1); /* IFLA_PROTO_DOWN */

928 929
}

930 931 932 933 934 935 936 937 938 939 940 941 942
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);
943 944
		if (!vf_port)
			goto nla_put_failure;
945 946
		if (nla_put_u32(skb, IFLA_PORT_VF, vf))
			goto nla_put_failure;
947
		err = dev->netdev_ops->ndo_get_vf_port(dev, vf, skb);
948 949
		if (err == -EMSGSIZE)
			goto nla_put_failure;
950 951 952 953 954 955 956 957 958 959
		if (err) {
			nla_nest_cancel(skb, vf_port);
			continue;
		}
		nla_nest_end(skb, vf_port);
	}

	nla_nest_end(skb, vf_ports);

	return 0;
960 961 962 963

nla_put_failure:
	nla_nest_cancel(skb, vf_ports);
	return -EMSGSIZE;
964 965 966 967 968 969 970 971 972 973 974 975 976 977
}

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);
978
		return (err == -EMSGSIZE) ? err : 0;
979 980 981 982 983 984 985
	}

	nla_nest_end(skb, port_self);

	return 0;
}

986 987
static int rtnl_port_fill(struct sk_buff *skb, struct net_device *dev,
			  u32 ext_filter_mask)
988 989 990
{
	int err;

991 992
	if (!dev->netdev_ops->ndo_get_vf_port || !dev->dev.parent ||
	    !(ext_filter_mask & RTEXT_FILTER_VF))
993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007
		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;
}

1008 1009 1010
static int rtnl_phys_port_id_fill(struct sk_buff *skb, struct net_device *dev)
{
	int err;
1011
	struct netdev_phys_item_id ppid;
1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025

	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;
}

1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037
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;
	}

1038
	if (nla_put_string(skb, IFLA_PHYS_PORT_NAME, name))
1039 1040 1041 1042 1043
		return -EMSGSIZE;

	return 0;
}

1044 1045 1046
static int rtnl_phys_switch_id_fill(struct sk_buff *skb, struct net_device *dev)
{
	int err;
1047
	struct switchdev_attr attr = {
1048
		.orig_dev = dev,
1049
		.id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID,
1050 1051
		.flags = SWITCHDEV_F_NO_RECURSE,
	};
1052

1053
	err = switchdev_port_attr_get(dev, &attr);
1054 1055 1056 1057 1058 1059
	if (err) {
		if (err == -EOPNOTSUPP)
			return 0;
		return err;
	}

1060 1061
	if (nla_put(skb, IFLA_PHYS_SWITCH_ID, attr.u.ppid.id_len,
		    attr.u.ppid.id))
1062 1063 1064 1065 1066
		return -EMSGSIZE;

	return 0;
}

1067 1068 1069
static noinline_for_stack int rtnl_fill_stats(struct sk_buff *skb,
					      struct net_device *dev)
{
1070
	struct rtnl_link_stats64 *sp;
1071
	struct nlattr *attr;
1072

1073 1074
	attr = nla_reserve_64bit(skb, IFLA_STATS64,
				 sizeof(struct rtnl_link_stats64), IFLA_PAD);
1075 1076 1077
	if (!attr)
		return -EMSGSIZE;

1078 1079
	sp = nla_data(attr);
	dev_get_stats(dev, sp);
1080

1081 1082
	attr = nla_reserve(skb, IFLA_STATS,
			   sizeof(struct rtnl_link_stats));
1083 1084 1085
	if (!attr)
		return -EMSGSIZE;

1086
	copy_rtnl_link_stats(nla_data(attr), sp);
1087 1088 1089 1090 1091 1092 1093 1094 1095 1096

	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;
1097
	struct nlattr *vf, *vfstats, *vfvlanlist;
1098
	struct ifla_vf_link_state vf_linkstate;
1099
	struct ifla_vf_vlan_info vf_vlan_info;
1100 1101 1102 1103 1104 1105 1106 1107 1108
	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;

1109 1110
	memset(&ivi, 0, sizeof(ivi));

1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122
	/* 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;
1123 1124
	/* VLAN Protocol by default is 802.1Q */
	ivi.vlan_proto = htons(ETH_P_8021Q);
1125 1126 1127
	if (dev->netdev_ops->ndo_get_vf_config(dev, vfs_num, &ivi))
		return 0;

1128 1129
	memset(&vf_vlan_info, 0, sizeof(vf_vlan_info));

1130 1131
	vf_mac.vf =
		vf_vlan.vf =
1132
		vf_vlan_info.vf =
1133 1134 1135 1136 1137 1138 1139 1140 1141 1142
		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;
1143 1144 1145
	vf_vlan_info.vlan = ivi.vlan;
	vf_vlan_info.qos = ivi.qos;
	vf_vlan_info.vlan_proto = ivi.vlan_proto;
1146 1147 1148 1149 1150 1151 1152 1153
	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);
1154 1155
	if (!vf)
		goto nla_put_vfinfo_failure;
1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170
	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))
1171 1172 1173 1174 1175 1176 1177 1178 1179 1180
		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);
1181 1182 1183 1184 1185
	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);
1186 1187
	if (!vfstats)
		goto nla_put_vf_failure;
N
Nicolas Dichtel 已提交
1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198
	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,
1199 1200 1201 1202
			      vf_stats.multicast, IFLA_VF_STATS_PAD)) {
		nla_nest_cancel(skb, vfstats);
		goto nla_put_vf_failure;
	}
1203 1204 1205
	nla_nest_end(skb, vfstats);
	nla_nest_end(skb, vf);
	return 0;
1206 1207 1208 1209 1210 1211

nla_put_vf_failure:
	nla_nest_cancel(skb, vf);
nla_put_vfinfo_failure:
	nla_nest_cancel(skb, vfinfo);
	return -EMSGSIZE;
1212 1213 1214 1215
}

static int rtnl_fill_link_ifmap(struct sk_buff *skb, struct net_device *dev)
{
K
Kangjie Lu 已提交
1216 1217 1218 1219 1220 1221 1222 1223 1224 1225
	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;

1226
	if (nla_put_64bit(skb, IFLA_MAP, sizeof(map), &map, IFLA_PAD))
1227 1228 1229 1230 1231
		return -EMSGSIZE;

	return 0;
}

M
Martin KaFai Lau 已提交
1232
static u8 rtnl_xdp_attached_mode(struct net_device *dev, u32 *prog_id)
1233 1234
{
	const struct net_device_ops *ops = dev->netdev_ops;
M
Martin KaFai Lau 已提交
1235
	const struct bpf_prog *generic_xdp_prog;
1236 1237 1238

	ASSERT_RTNL();

M
Martin KaFai Lau 已提交
1239 1240 1241 1242
	*prog_id = 0;
	generic_xdp_prog = rtnl_dereference(dev->xdp_prog);
	if (generic_xdp_prog) {
		*prog_id = generic_xdp_prog->aux->id;
1243
		return XDP_ATTACHED_SKB;
M
Martin KaFai Lau 已提交
1244
	}
1245 1246
	if (!ops->ndo_xdp)
		return XDP_ATTACHED_NONE;
1247

1248
	return __dev_xdp_attached(dev, ops->ndo_xdp, prog_id);
1249 1250
}

1251 1252 1253
static int rtnl_xdp_fill(struct sk_buff *skb, struct net_device *dev)
{
	struct nlattr *xdp;
M
Martin KaFai Lau 已提交
1254
	u32 prog_id;
1255 1256 1257 1258 1259
	int err;

	xdp = nla_nest_start(skb, IFLA_XDP);
	if (!xdp)
		return -EMSGSIZE;
1260 1261

	err = nla_put_u8(skb, IFLA_XDP_ATTACHED,
M
Martin KaFai Lau 已提交
1262
			 rtnl_xdp_attached_mode(dev, &prog_id));
1263 1264 1265
	if (err)
		goto err_cancel;

M
Martin KaFai Lau 已提交
1266 1267 1268 1269 1270 1271
	if (prog_id) {
		err = nla_put_u32(skb, IFLA_XDP_PROG_ID, prog_id);
		if (err)
			goto err_cancel;
	}

1272 1273 1274 1275 1276 1277 1278 1279
	nla_nest_end(skb, xdp);
	return 0;

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

1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309
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;
}

1310
static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
1311
			    int type, u32 pid, u32 seq, u32 change,
1312 1313
			    unsigned int flags, u32 ext_filter_mask,
			    u32 event)
1314 1315 1316
{
	struct ifinfomsg *ifm;
	struct nlmsghdr *nlh;
1317
	struct nlattr *af_spec;
T
Thomas Graf 已提交
1318
	struct rtnl_af_ops *af_ops;
1319
	struct net_device *upper_dev = netdev_master_upper_dev_get(dev);
L
Linus Torvalds 已提交
1320

1321
	ASSERT_RTNL();
1322 1323
	nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ifm), flags);
	if (nlh == NULL)
1324
		return -EMSGSIZE;
L
Linus Torvalds 已提交
1325

1326 1327 1328 1329 1330 1331 1332 1333
	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;

1334 1335 1336 1337 1338 1339 1340
	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) ||
1341
	    nla_put_u32(skb, IFLA_PROMISCUITY, dev->promiscuity) ||
1342
	    nla_put_u32(skb, IFLA_NUM_TX_QUEUES, dev->num_tx_queues) ||
1343 1344
	    nla_put_u32(skb, IFLA_GSO_MAX_SEGS, dev->gso_max_segs) ||
	    nla_put_u32(skb, IFLA_GSO_MAX_SIZE, dev->gso_max_size) ||
1345
#ifdef CONFIG_RPS
1346
	    nla_put_u32(skb, IFLA_NUM_RX_QUEUES, dev->num_rx_queues) ||
1347
#endif
1348 1349
	    (dev->ifindex != dev_get_iflink(dev) &&
	     nla_put_u32(skb, IFLA_LINK, dev_get_iflink(dev))) ||
1350 1351
	    (upper_dev &&
	     nla_put_u32(skb, IFLA_MASTER, upper_dev->ifindex)) ||
1352
	    nla_put_u8(skb, IFLA_CARRIER, netif_carrier_ok(dev)) ||
1353 1354 1355
	    (dev->qdisc &&
	     nla_put_string(skb, IFLA_QDISC, dev->qdisc->ops->id)) ||
	    (dev->ifalias &&
1356 1357
	     nla_put_string(skb, IFLA_IFALIAS, dev->ifalias)) ||
	    nla_put_u32(skb, IFLA_CARRIER_CHANGES,
1358 1359
			atomic_read(&dev->carrier_changes)) ||
	    nla_put_u8(skb, IFLA_PROTO_DOWN, dev->proto_down))
1360
		goto nla_put_failure;
1361

1362 1363 1364 1365 1366
	if (event != IFLA_EVENT_NONE) {
		if (nla_put_u32(skb, IFLA_EVENT, event))
			goto nla_put_failure;
	}

1367 1368
	if (rtnl_fill_link_ifmap(skb, dev))
		goto nla_put_failure;
L
Linus Torvalds 已提交
1369 1370

	if (dev->addr_len) {
1371 1372 1373
		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
1374 1375
	}

1376 1377 1378
	if (rtnl_phys_port_id_fill(skb, dev))
		goto nla_put_failure;

1379 1380 1381
	if (rtnl_phys_port_name_fill(skb, dev))
		goto nla_put_failure;

1382 1383 1384
	if (rtnl_phys_switch_id_fill(skb, dev))
		goto nla_put_failure;

1385
	if (rtnl_fill_stats(skb, dev))
1386 1387
		goto nla_put_failure;

1388 1389 1390
	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;
1391

1392 1393
	if (dev->netdev_ops->ndo_get_vf_config && dev->dev.parent &&
	    ext_filter_mask & RTEXT_FILTER_VF) {
1394
		int i;
1395
		struct nlattr *vfinfo;
1396 1397 1398 1399 1400 1401
		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++) {
1402
			if (rtnl_fill_vfinfo(skb, dev, i, vfinfo))
1403
				goto nla_put_failure;
1404
		}
1405

1406
		nla_nest_end(skb, vfinfo);
1407
	}
1408

1409
	if (rtnl_port_fill(skb, dev, ext_filter_mask))
1410 1411
		goto nla_put_failure;

1412 1413 1414
	if (rtnl_xdp_fill(skb, dev))
		goto nla_put_failure;

1415
	if (dev->rtnl_link_ops || rtnl_have_link_slave_info(dev)) {
P
Patrick McHardy 已提交
1416 1417 1418 1419
		if (rtnl_link_fill(skb, dev) < 0)
			goto nla_put_failure;
	}

1420 1421 1422 1423 1424
	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)) {
1425
			int id = peernet2id_alloc(dev_net(dev), link_net);
1426 1427 1428 1429 1430 1431

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

T
Thomas Graf 已提交
1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442
	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;

1443
			err = af_ops->fill_link_af(skb, dev, ext_filter_mask);
T
Thomas Graf 已提交
1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461

			/*
			 * 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);

1462 1463
	nlmsg_end(skb, nlh);
	return 0;
1464 1465

nla_put_failure:
1466 1467
	nlmsg_cancel(skb, nlh);
	return -EMSGSIZE;
L
Linus Torvalds 已提交
1468 1469
}

J
Jiri Pirko 已提交
1470
static const struct nla_policy ifla_policy[IFLA_MAX+1] = {
1471
	[IFLA_IFNAME]		= { .type = NLA_STRING, .len = IFNAMSIZ-1 },
P
Patrick McHardy 已提交
1472 1473
	[IFLA_ADDRESS]		= { .type = NLA_BINARY, .len = MAX_ADDR_LEN },
	[IFLA_BROADCAST]	= { .type = NLA_BINARY, .len = MAX_ADDR_LEN },
1474
	[IFLA_MAP]		= { .len = sizeof(struct rtnl_link_ifmap) },
1475
	[IFLA_MTU]		= { .type = NLA_U32 },
1476
	[IFLA_LINK]		= { .type = NLA_U32 },
1477
	[IFLA_MASTER]		= { .type = NLA_U32 },
1478
	[IFLA_CARRIER]		= { .type = NLA_U8 },
1479 1480 1481 1482
	[IFLA_TXQLEN]		= { .type = NLA_U32 },
	[IFLA_WEIGHT]		= { .type = NLA_U32 },
	[IFLA_OPERSTATE]	= { .type = NLA_U8 },
	[IFLA_LINKMODE]		= { .type = NLA_U8 },
1483
	[IFLA_LINKINFO]		= { .type = NLA_NESTED },
1484
	[IFLA_NET_NS_PID]	= { .type = NLA_U32 },
1485
	[IFLA_NET_NS_FD]	= { .type = NLA_U32 },
1486
	[IFLA_IFALIAS]	        = { .type = NLA_STRING, .len = IFALIASZ-1 },
1487
	[IFLA_VFINFO_LIST]	= {. type = NLA_NESTED },
1488 1489
	[IFLA_VF_PORTS]		= { .type = NLA_NESTED },
	[IFLA_PORT_SELF]	= { .type = NLA_NESTED },
T
Thomas Graf 已提交
1490
	[IFLA_AF_SPEC]		= { .type = NLA_NESTED },
1491
	[IFLA_EXT_MASK]		= { .type = NLA_U32 },
1492
	[IFLA_PROMISCUITY]	= { .type = NLA_U32 },
1493 1494
	[IFLA_NUM_TX_QUEUES]	= { .type = NLA_U32 },
	[IFLA_NUM_RX_QUEUES]	= { .type = NLA_U32 },
1495
	[IFLA_PHYS_PORT_ID]	= { .type = NLA_BINARY, .len = MAX_PHYS_ITEM_ID_LEN },
1496
	[IFLA_CARRIER_CHANGES]	= { .type = NLA_U32 },  /* ignored */
1497
	[IFLA_PHYS_SWITCH_ID]	= { .type = NLA_BINARY, .len = MAX_PHYS_ITEM_ID_LEN },
1498
	[IFLA_LINK_NETNSID]	= { .type = NLA_S32 },
1499
	[IFLA_PROTO_DOWN]	= { .type = NLA_U8 },
1500
	[IFLA_XDP]		= { .type = NLA_NESTED },
1501
	[IFLA_EVENT]		= { .type = NLA_U32 },
1502
	[IFLA_GROUP]		= { .type = NLA_U32 },
1503 1504
};

P
Patrick McHardy 已提交
1505 1506 1507
static const struct nla_policy ifla_info_policy[IFLA_INFO_MAX+1] = {
	[IFLA_INFO_KIND]	= { .type = NLA_STRING },
	[IFLA_INFO_DATA]	= { .type = NLA_NESTED },
1508 1509
	[IFLA_INFO_SLAVE_KIND]	= { .type = NLA_STRING },
	[IFLA_INFO_SLAVE_DATA]	= { .type = NLA_NESTED },
P
Patrick McHardy 已提交
1510 1511
};

1512
static const struct nla_policy ifla_vf_policy[IFLA_VF_MAX+1] = {
1513 1514
	[IFLA_VF_MAC]		= { .len = sizeof(struct ifla_vf_mac) },
	[IFLA_VF_VLAN]		= { .len = sizeof(struct ifla_vf_vlan) },
1515
	[IFLA_VF_VLAN_LIST]     = { .type = NLA_NESTED },
1516 1517 1518 1519
	[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) },
1520
	[IFLA_VF_RSS_QUERY_EN]	= { .len = sizeof(struct ifla_vf_rss_query_en) },
1521
	[IFLA_VF_STATS]		= { .type = NLA_NESTED },
H
Hiroshi Shimamoto 已提交
1522
	[IFLA_VF_TRUST]		= { .len = sizeof(struct ifla_vf_trust) },
1523 1524
	[IFLA_VF_IB_NODE_GUID]	= { .len = sizeof(struct ifla_vf_guid) },
	[IFLA_VF_IB_PORT_GUID]	= { .len = sizeof(struct ifla_vf_guid) },
1525 1526
};

1527 1528 1529 1530 1531 1532 1533 1534 1535 1536
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, },
1537 1538 1539 1540 1541 1542 1543

	/* 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) },
1544 1545
};

1546 1547 1548
static const struct nla_policy ifla_xdp_policy[IFLA_XDP_MAX + 1] = {
	[IFLA_XDP_FD]		= { .type = NLA_S32 },
	[IFLA_XDP_ATTACHED]	= { .type = NLA_U8 },
1549
	[IFLA_XDP_FLAGS]	= { .type = NLA_U32 },
M
Martin KaFai Lau 已提交
1550
	[IFLA_XDP_PROG_ID]	= { .type = NLA_U32 },
1551 1552
};

1553 1554 1555 1556 1557
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];

1558 1559
	if (nla_parse_nested(linfo, IFLA_INFO_MAX, nla,
			     ifla_info_policy, NULL) < 0)
1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605
		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
1606 1607 1608 1609 1610 1611 1612 1613 1614
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;
1615 1616 1617
	const struct rtnl_link_ops *kind_ops = NULL;
	unsigned int flags = NLM_F_MULTI;
	int master_idx = 0;
1618
	int err;
1619
	int hdrlen;
J
Jiri Pirko 已提交
1620 1621 1622 1623

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

1624 1625 1626 1627 1628 1629 1630 1631 1632 1633
	/* 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);

1634 1635
	if (nlmsg_parse(cb->nlh, hdrlen, tb, IFLA_MAX,
			ifla_policy, NULL) >= 0) {
J
Jiri Pirko 已提交
1636 1637
		if (tb[IFLA_EXT_MASK])
			ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);
1638 1639 1640 1641 1642 1643 1644 1645 1646

		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
1647 1648 1649 1650 1651
	}

	for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) {
		idx = 0;
		head = &net->dev_index_head[h];
1652
		hlist_for_each_entry(dev, head, index_hlist) {
1653
			if (link_dump_filtered(dev, master_idx, kind_ops))
1654
				goto cont;
J
Jiri Pirko 已提交
1655 1656
			if (idx < s_idx)
				goto cont;
1657 1658 1659
			err = rtnl_fill_ifinfo(skb, dev, RTM_NEWLINK,
					       NETLINK_CB(cb->skb).portid,
					       cb->nlh->nlmsg_seq, 0,
1660
					       flags,
1661
					       ext_filter_mask, 0);
1662

1663 1664 1665 1666 1667 1668
			if (err < 0) {
				if (likely(skb->len))
					goto out;

				goto out_err;
			}
J
Jiri Pirko 已提交
1669 1670 1671 1672 1673
cont:
			idx++;
		}
	}
out:
1674 1675
	err = skb->len;
out_err:
J
Jiri Pirko 已提交
1676 1677
	cb->args[1] = idx;
	cb->args[0] = h;
1678 1679
	cb->seq = net->dev_base_seq;
	nl_dump_check_consistent(cb, nlmsg_hdr(skb));
J
Jiri Pirko 已提交
1680

1681
	return err;
J
Jiri Pirko 已提交
1682 1683
}

1684 1685
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
1686
{
1687
	return nla_parse(tb, IFLA_MAX, head, len, ifla_policy, exterr);
J
Jiri Pirko 已提交
1688 1689 1690
}
EXPORT_SYMBOL(rtnl_nla_parse_ifla);

1691 1692 1693 1694 1695 1696 1697 1698
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]));
1699 1700
	else if (tb[IFLA_NET_NS_FD])
		net = get_net_ns_by_fd(nla_get_u32(tb[IFLA_NET_NS_FD]));
1701 1702 1703 1704 1705 1706
	else
		net = get_net(src_net);
	return net;
}
EXPORT_SYMBOL(rtnl_link_get_net);

1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718
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;
	}

1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732
	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) {
1733
				err = af_ops->validate_link_af(dev, af);
1734 1735 1736 1737 1738 1739
				if (err < 0)
					return err;
			}
		}
	}

1740 1741 1742
	return 0;
}

1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758
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);
}

1759
static int do_setvfinfo(struct net_device *dev, struct nlattr **tb)
1760 1761
{
	const struct net_device_ops *ops = dev->netdev_ops;
1762
	int err = -EINVAL;
1763

1764 1765
	if (tb[IFLA_VF_MAC]) {
		struct ifla_vf_mac *ivm = nla_data(tb[IFLA_VF_MAC]);
1766

1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780
		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,
1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806
						   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++;
		}
1807 1808 1809
		if (len == 0)
			return -EINVAL;

1810 1811
		err = ops->ndo_set_vf_vlan(dev, ivvl[0]->vf, ivvl[0]->vlan,
					   ivvl[0]->qos, ivvl[0]->vlan_proto);
1812 1813
		if (err < 0)
			return err;
1814
	}
1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 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

	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
1881 1882 1883 1884 1885 1886 1887 1888 1889 1890
	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;
	}

1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908
	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);
	}

1909 1910 1911
	return err;
}

1912 1913
static int do_set_master(struct net_device *dev, int ifindex)
{
1914
	struct net_device *upper_dev = netdev_master_upper_dev_get(dev);
1915 1916 1917
	const struct net_device_ops *ops;
	int err;

1918 1919
	if (upper_dev) {
		if (upper_dev->ifindex == ifindex)
1920
			return 0;
1921
		ops = upper_dev->netdev_ops;
1922
		if (ops->ndo_del_slave) {
1923
			err = ops->ndo_del_slave(upper_dev, dev);
1924 1925 1926 1927 1928 1929 1930 1931
			if (err)
				return err;
		} else {
			return -EOPNOTSUPP;
		}
	}

	if (ifindex) {
1932 1933
		upper_dev = __dev_get_by_index(dev_net(dev), ifindex);
		if (!upper_dev)
1934
			return -EINVAL;
1935
		ops = upper_dev->netdev_ops;
1936
		if (ops->ndo_add_slave) {
1937
			err = ops->ndo_add_slave(upper_dev, dev);
1938 1939 1940 1941 1942 1943 1944 1945 1946
			if (err)
				return err;
		} else {
			return -EOPNOTSUPP;
		}
	}
	return 0;
}

1947
#define DO_SETLINK_MODIFIED	0x01
1948 1949
/* notify flag means notify + modified. */
#define DO_SETLINK_NOTIFY	0x03
1950 1951
static int do_setlink(const struct sk_buff *skb,
		      struct net_device *dev, struct ifinfomsg *ifm,
1952
		      struct netlink_ext_ack *extack,
1953
		      struct nlattr **tb, char *ifname, int status)
L
Linus Torvalds 已提交
1954
{
1955
	const struct net_device_ops *ops = dev->netdev_ops;
1956
	int err;
L
Linus Torvalds 已提交
1957

1958
	if (tb[IFLA_NET_NS_PID] || tb[IFLA_NET_NS_FD]) {
1959
		struct net *net = rtnl_link_get_net(dev_net(dev), tb);
1960 1961 1962 1963
		if (IS_ERR(net)) {
			err = PTR_ERR(net);
			goto errout;
		}
1964
		if (!netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) {
1965
			put_net(net);
1966 1967 1968
			err = -EPERM;
			goto errout;
		}
1969 1970 1971 1972
		err = dev_change_net_namespace(dev, net, ifname);
		put_net(net);
		if (err)
			goto errout;
1973
		status |= DO_SETLINK_MODIFIED;
1974 1975
	}

1976
	if (tb[IFLA_MAP]) {
L
Linus Torvalds 已提交
1977 1978 1979
		struct rtnl_link_ifmap *u_map;
		struct ifmap k_map;

1980
		if (!ops->ndo_set_config) {
L
Linus Torvalds 已提交
1981
			err = -EOPNOTSUPP;
1982
			goto errout;
L
Linus Torvalds 已提交
1983 1984 1985 1986
		}

		if (!netif_device_present(dev)) {
			err = -ENODEV;
1987
			goto errout;
L
Linus Torvalds 已提交
1988 1989
		}

1990
		u_map = nla_data(tb[IFLA_MAP]);
L
Linus Torvalds 已提交
1991 1992 1993 1994 1995 1996 1997
		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;

1998
		err = ops->ndo_set_config(dev, &k_map);
1999
		if (err < 0)
2000
			goto errout;
L
Linus Torvalds 已提交
2001

2002
		status |= DO_SETLINK_NOTIFY;
L
Linus Torvalds 已提交
2003 2004
	}

2005
	if (tb[IFLA_ADDRESS]) {
2006 2007 2008
		struct sockaddr *sa;
		int len;

2009 2010
		len = sizeof(sa_family_t) + max_t(size_t, dev->addr_len,
						  sizeof(*sa));
2011 2012 2013
		sa = kmalloc(len, GFP_KERNEL);
		if (!sa) {
			err = -ENOMEM;
2014
			goto errout;
2015 2016
		}
		sa->sa_family = dev->type;
2017
		memcpy(sa->sa_data, nla_data(tb[IFLA_ADDRESS]),
2018
		       dev->addr_len);
2019
		err = dev_set_mac_address(dev, sa);
2020
		kfree(sa);
L
Linus Torvalds 已提交
2021
		if (err)
2022
			goto errout;
2023
		status |= DO_SETLINK_MODIFIED;
L
Linus Torvalds 已提交
2024 2025
	}

2026 2027 2028
	if (tb[IFLA_MTU]) {
		err = dev_set_mtu(dev, nla_get_u32(tb[IFLA_MTU]));
		if (err < 0)
2029
			goto errout;
2030
		status |= DO_SETLINK_MODIFIED;
L
Linus Torvalds 已提交
2031 2032
	}

2033 2034
	if (tb[IFLA_GROUP]) {
		dev_set_group(dev, nla_get_u32(tb[IFLA_GROUP]));
2035
		status |= DO_SETLINK_NOTIFY;
2036 2037
	}

2038 2039 2040 2041 2042
	/*
	 * Interface selected by interface index but interface
	 * name provided implies that a name change has been
	 * requested.
	 */
2043
	if (ifm->ifi_index > 0 && ifname[0]) {
2044 2045
		err = dev_change_name(dev, ifname);
		if (err < 0)
2046
			goto errout;
2047
		status |= DO_SETLINK_MODIFIED;
L
Linus Torvalds 已提交
2048 2049
	}

2050 2051 2052 2053 2054
	if (tb[IFLA_IFALIAS]) {
		err = dev_set_alias(dev, nla_data(tb[IFLA_IFALIAS]),
				    nla_len(tb[IFLA_IFALIAS]));
		if (err < 0)
			goto errout;
2055
		status |= DO_SETLINK_NOTIFY;
2056 2057
	}

2058 2059
	if (tb[IFLA_BROADCAST]) {
		nla_memcpy(dev->broadcast, tb[IFLA_BROADCAST], dev->addr_len);
2060
		call_netdevice_notifiers(NETDEV_CHANGEADDR, dev);
L
Linus Torvalds 已提交
2061 2062
	}

2063
	if (ifm->ifi_flags || ifm->ifi_change) {
2064
		err = dev_change_flags(dev, rtnl_dev_combine_flags(dev, ifm));
2065 2066
		if (err < 0)
			goto errout;
2067
	}
L
Linus Torvalds 已提交
2068

2069 2070 2071 2072
	if (tb[IFLA_MASTER]) {
		err = do_set_master(dev, nla_get_u32(tb[IFLA_MASTER]));
		if (err)
			goto errout;
2073
		status |= DO_SETLINK_MODIFIED;
2074 2075
	}

2076 2077 2078 2079
	if (tb[IFLA_CARRIER]) {
		err = dev_change_carrier(dev, nla_get_u8(tb[IFLA_CARRIER]));
		if (err)
			goto errout;
2080
		status |= DO_SETLINK_MODIFIED;
2081 2082
	}

2083
	if (tb[IFLA_TXQLEN]) {
2084 2085
		unsigned int value = nla_get_u32(tb[IFLA_TXQLEN]);
		unsigned int orig_len = dev->tx_queue_len;
2086 2087 2088 2089 2090 2091 2092 2093 2094 2095

		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;
			}
2096
			status |= DO_SETLINK_NOTIFY;
2097
		}
2098
	}
S
Stefan Rompf 已提交
2099

2100
	if (tb[IFLA_OPERSTATE])
2101
		set_operstate(dev, nla_get_u8(tb[IFLA_OPERSTATE]));
S
Stefan Rompf 已提交
2102

2103
	if (tb[IFLA_LINKMODE]) {
2104 2105
		unsigned char value = nla_get_u8(tb[IFLA_LINKMODE]);

2106
		write_lock_bh(&dev_base_lock);
2107
		if (dev->link_mode ^ value)
2108
			status |= DO_SETLINK_NOTIFY;
2109
		dev->link_mode = value;
2110
		write_unlock_bh(&dev_base_lock);
S
Stefan Rompf 已提交
2111 2112
	}

2113
	if (tb[IFLA_VFINFO_LIST]) {
2114
		struct nlattr *vfinfo[IFLA_VF_MAX + 1];
2115 2116
		struct nlattr *attr;
		int rem;
2117

2118
		nla_for_each_nested(attr, tb[IFLA_VFINFO_LIST], rem) {
2119 2120
			if (nla_type(attr) != IFLA_VF_INFO ||
			    nla_len(attr) < NLA_HDRLEN) {
2121
				err = -EINVAL;
2122
				goto errout;
2123
			}
2124
			err = nla_parse_nested(vfinfo, IFLA_VF_MAX, attr,
2125
					       ifla_vf_policy, NULL);
2126 2127 2128
			if (err < 0)
				goto errout;
			err = do_setvfinfo(dev, vfinfo);
2129 2130
			if (err < 0)
				goto errout;
2131
			status |= DO_SETLINK_NOTIFY;
2132
		}
2133
	}
L
Linus Torvalds 已提交
2134 2135
	err = 0;

2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146
	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) {
2147 2148 2149 2150 2151 2152
			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,
2153
					       ifla_port_policy, NULL);
2154 2155 2156 2157 2158 2159 2160 2161 2162 2163
			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;
2164
			status |= DO_SETLINK_NOTIFY;
2165 2166 2167 2168 2169 2170 2171 2172
		}
	}
	err = 0;

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

		err = nla_parse_nested(port, IFLA_PORT_MAX,
2173 2174
				       tb[IFLA_PORT_SELF], ifla_port_policy,
				       NULL);
2175 2176 2177 2178 2179 2180 2181 2182
		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;
2183
		status |= DO_SETLINK_NOTIFY;
2184
	}
T
Thomas Graf 已提交
2185 2186 2187 2188 2189 2190 2191 2192 2193

	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))))
2194
				BUG();
T
Thomas Graf 已提交
2195

2196
			err = af_ops->set_link_af(dev, af);
T
Thomas Graf 已提交
2197 2198 2199
			if (err < 0)
				goto errout;

2200
			status |= DO_SETLINK_NOTIFY;
T
Thomas Graf 已提交
2201 2202
		}
	}
2203 2204
	err = 0;

2205 2206 2207 2208 2209 2210 2211 2212
	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;
	}

2213 2214
	if (tb[IFLA_XDP]) {
		struct nlattr *xdp[IFLA_XDP_MAX + 1];
2215
		u32 xdp_flags = 0;
2216 2217

		err = nla_parse_nested(xdp, IFLA_XDP_MAX, tb[IFLA_XDP],
2218
				       ifla_xdp_policy, NULL);
2219 2220 2221
		if (err < 0)
			goto errout;

M
Martin KaFai Lau 已提交
2222
		if (xdp[IFLA_XDP_ATTACHED] || xdp[IFLA_XDP_PROG_ID]) {
2223 2224 2225
			err = -EINVAL;
			goto errout;
		}
2226 2227 2228 2229 2230 2231 2232

		if (xdp[IFLA_XDP_FLAGS]) {
			xdp_flags = nla_get_u32(xdp[IFLA_XDP_FLAGS]);
			if (xdp_flags & ~XDP_FLAGS_MASK) {
				err = -EINVAL;
				goto errout;
			}
2233
			if (hweight32(xdp_flags & XDP_FLAGS_MODES) > 1) {
2234 2235 2236
				err = -EINVAL;
				goto errout;
			}
2237 2238
		}

2239
		if (xdp[IFLA_XDP_FD]) {
2240
			err = dev_change_xdp_fd(dev, extack,
2241 2242
						nla_get_s32(xdp[IFLA_XDP_FD]),
						xdp_flags);
2243 2244 2245 2246 2247 2248
			if (err)
				goto errout;
			status |= DO_SETLINK_NOTIFY;
		}
	}

2249
errout:
2250 2251 2252 2253 2254 2255 2256 2257
	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);
	}
2258

2259 2260
	return err;
}
L
Linus Torvalds 已提交
2261

2262 2263
static int rtnl_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
2264
{
2265
	struct net *net = sock_net(skb->sk);
2266 2267 2268 2269 2270 2271
	struct ifinfomsg *ifm;
	struct net_device *dev;
	int err;
	struct nlattr *tb[IFLA_MAX+1];
	char ifname[IFNAMSIZ];

2272 2273
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy,
			  extack);
2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284
	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)
2285
		dev = __dev_get_by_index(net, ifm->ifi_index);
2286
	else if (tb[IFLA_IFNAME])
2287
		dev = __dev_get_by_name(net, ifname);
2288 2289 2290 2291 2292 2293 2294 2295
	else
		goto errout;

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

E
Eric Dumazet 已提交
2296 2297
	err = validate_linkmsg(dev, tb);
	if (err < 0)
2298
		goto errout;
2299

2300
	err = do_setlink(skb, dev, ifm, extack, tb, ifname, 0);
2301
errout:
L
Linus Torvalds 已提交
2302 2303 2304
	return err;
}

2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340
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;
}

2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356
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);

2357 2358
static int rtnl_dellink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
P
Patrick McHardy 已提交
2359
{
2360
	struct net *net = sock_net(skb->sk);
P
Patrick McHardy 已提交
2361 2362 2363 2364 2365 2366
	struct net_device *dev;
	struct ifinfomsg *ifm;
	char ifname[IFNAMSIZ];
	struct nlattr *tb[IFLA_MAX+1];
	int err;

2367
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy, extack);
P
Patrick McHardy 已提交
2368 2369 2370 2371 2372 2373 2374 2375
	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)
2376
		dev = __dev_get_by_index(net, ifm->ifi_index);
P
Patrick McHardy 已提交
2377
	else if (tb[IFLA_IFNAME])
2378
		dev = __dev_get_by_name(net, ifname);
2379 2380
	else if (tb[IFLA_GROUP])
		return rtnl_group_dellink(net, nla_get_u32(tb[IFLA_GROUP]));
P
Patrick McHardy 已提交
2381 2382 2383 2384 2385 2386
	else
		return -EINVAL;

	if (!dev)
		return -ENODEV;

2387
	return rtnl_delete_link(dev);
P
Patrick McHardy 已提交
2388 2389
}

2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403
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;

2404
	__dev_notify_flags(dev, old_flags, ~0U);
2405 2406 2407 2408
	return 0;
}
EXPORT_SYMBOL(rtnl_configure_link);

2409
struct net_device *rtnl_create_link(struct net *net,
2410
	const char *ifname, unsigned char name_assign_type,
2411
	const struct rtnl_link_ops *ops, struct nlattr *tb[])
2412 2413
{
	struct net_device *dev;
2414 2415
	unsigned int num_tx_queues = 1;
	unsigned int num_rx_queues = 1;
2416

2417 2418 2419
	if (tb[IFLA_NUM_TX_QUEUES])
		num_tx_queues = nla_get_u32(tb[IFLA_NUM_TX_QUEUES]);
	else if (ops->get_num_tx_queues)
2420
		num_tx_queues = ops->get_num_tx_queues();
2421 2422 2423 2424

	if (tb[IFLA_NUM_RX_QUEUES])
		num_rx_queues = nla_get_u32(tb[IFLA_NUM_RX_QUEUES]);
	else if (ops->get_num_rx_queues)
2425
		num_rx_queues = ops->get_num_rx_queues();
2426

2427
	dev = alloc_netdev_mqs(ops->priv_size, ifname, name_assign_type,
2428
			       ops->setup, num_tx_queues, num_rx_queues);
2429
	if (!dev)
2430
		return ERR_PTR(-ENOMEM);
2431

2432 2433
	dev_net_set(dev, net);
	dev->rtnl_link_ops = ops;
2434
	dev->rtnl_link_state = RTNL_LINK_INITIALIZING;
2435

2436 2437
	if (tb[IFLA_MTU])
		dev->mtu = nla_get_u32(tb[IFLA_MTU]);
2438
	if (tb[IFLA_ADDRESS]) {
2439 2440
		memcpy(dev->dev_addr, nla_data(tb[IFLA_ADDRESS]),
				nla_len(tb[IFLA_ADDRESS]));
2441 2442
		dev->addr_assign_type = NET_ADDR_SET;
	}
2443 2444 2445 2446 2447 2448
	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])
2449
		set_operstate(dev, nla_get_u8(tb[IFLA_OPERSTATE]));
2450 2451
	if (tb[IFLA_LINKMODE])
		dev->link_mode = nla_get_u8(tb[IFLA_LINKMODE]);
2452 2453
	if (tb[IFLA_GROUP])
		dev_set_group(dev, nla_get_u32(tb[IFLA_GROUP]));
2454 2455 2456

	return dev;
}
E
Eric Dumazet 已提交
2457
EXPORT_SYMBOL(rtnl_create_link);
2458

2459 2460
static int rtnl_group_changelink(const struct sk_buff *skb,
		struct net *net, int group,
2461
		struct ifinfomsg *ifm,
2462
		struct netlink_ext_ack *extack,
2463 2464
		struct nlattr **tb)
{
2465
	struct net_device *dev, *aux;
2466 2467
	int err;

2468
	for_each_netdev_safe(net, dev, aux) {
2469
		if (dev->group == group) {
2470
			err = do_setlink(skb, dev, ifm, extack, tb, NULL, 0);
2471 2472 2473 2474 2475 2476 2477 2478
			if (err < 0)
				return err;
		}
	}

	return 0;
}

2479 2480
static int rtnl_newlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
P
Patrick McHardy 已提交
2481
{
2482
	struct net *net = sock_net(skb->sk);
P
Patrick McHardy 已提交
2483
	const struct rtnl_link_ops *ops;
2484
	const struct rtnl_link_ops *m_ops = NULL;
P
Patrick McHardy 已提交
2485
	struct net_device *dev;
2486
	struct net_device *master_dev = NULL;
P
Patrick McHardy 已提交
2487 2488 2489 2490 2491
	struct ifinfomsg *ifm;
	char kind[MODULE_NAME_LEN];
	char ifname[IFNAMSIZ];
	struct nlattr *tb[IFLA_MAX+1];
	struct nlattr *linkinfo[IFLA_INFO_MAX+1];
2492
	unsigned char name_assign_type = NET_NAME_USER;
P
Patrick McHardy 已提交
2493 2494
	int err;

2495
#ifdef CONFIG_MODULES
P
Patrick McHardy 已提交
2496
replay:
2497
#endif
2498
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy, extack);
P
Patrick McHardy 已提交
2499 2500 2501 2502 2503 2504 2505 2506 2507 2508
	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)
2509
		dev = __dev_get_by_index(net, ifm->ifi_index);
2510 2511 2512 2513 2514 2515
	else {
		if (ifname[0])
			dev = __dev_get_by_name(net, ifname);
		else
			dev = NULL;
	}
P
Patrick McHardy 已提交
2516

2517 2518 2519 2520 2521 2522
	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
2523 2524
	err = validate_linkmsg(dev, tb);
	if (err < 0)
2525 2526
		return err;

P
Patrick McHardy 已提交
2527 2528
	if (tb[IFLA_LINKINFO]) {
		err = nla_parse_nested(linkinfo, IFLA_INFO_MAX,
2529 2530
				       tb[IFLA_LINKINFO], ifla_info_policy,
				       NULL);
P
Patrick McHardy 已提交
2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544
		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
2545 2546
		struct nlattr *attr[ops ? ops->maxtype + 1 : 1];
		struct nlattr *slave_attr[m_ops ? m_ops->slave_maxtype + 1 : 1];
2547 2548
		struct nlattr **data = NULL;
		struct nlattr **slave_data = NULL;
2549
		struct net *dest_net, *link_net = NULL;
P
Patrick McHardy 已提交
2550 2551 2552 2553 2554

		if (ops) {
			if (ops->maxtype && linkinfo[IFLA_INFO_DATA]) {
				err = nla_parse_nested(attr, ops->maxtype,
						       linkinfo[IFLA_INFO_DATA],
2555
						       ops->policy, NULL);
P
Patrick McHardy 已提交
2556 2557 2558 2559 2560
				if (err < 0)
					return err;
				data = attr;
			}
			if (ops->validate) {
2561
				err = ops->validate(tb, data, extack);
P
Patrick McHardy 已提交
2562 2563 2564 2565 2566
				if (err < 0)
					return err;
			}
		}

2567 2568 2569 2570 2571 2572
		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],
2573 2574
						       m_ops->slave_policy,
						       NULL);
2575 2576 2577 2578 2579
				if (err < 0)
					return err;
				slave_data = slave_attr;
			}
			if (m_ops->slave_validate) {
2580 2581
				err = m_ops->slave_validate(tb, slave_data,
							    extack);
2582 2583 2584 2585 2586
				if (err < 0)
					return err;
			}
		}

P
Patrick McHardy 已提交
2587
		if (dev) {
2588
			int status = 0;
P
Patrick McHardy 已提交
2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599

			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;

2600
				err = ops->changelink(dev, tb, data, extack);
P
Patrick McHardy 已提交
2601 2602
				if (err < 0)
					return err;
2603
				status |= DO_SETLINK_NOTIFY;
P
Patrick McHardy 已提交
2604 2605
			}

2606 2607 2608 2609 2610
			if (linkinfo[IFLA_INFO_SLAVE_DATA]) {
				if (!m_ops || !m_ops->slave_changelink)
					return -EOPNOTSUPP;

				err = m_ops->slave_changelink(master_dev, dev,
2611 2612
							      tb, slave_data,
							      extack);
2613 2614
				if (err < 0)
					return err;
2615
				status |= DO_SETLINK_NOTIFY;
2616 2617
			}

2618 2619
			return do_setlink(skb, dev, ifm, extack, tb, ifname,
					  status);
P
Patrick McHardy 已提交
2620 2621
		}

2622 2623
		if (!(nlh->nlmsg_flags & NLM_F_CREATE)) {
			if (ifm->ifi_index == 0 && tb[IFLA_GROUP])
2624
				return rtnl_group_changelink(skb, net,
2625
						nla_get_u32(tb[IFLA_GROUP]),
2626
						ifm, extack, tb);
P
Patrick McHardy 已提交
2627
			return -ENODEV;
2628
		}
P
Patrick McHardy 已提交
2629

2630
		if (tb[IFLA_MAP] || tb[IFLA_PROTINFO])
P
Patrick McHardy 已提交
2631 2632 2633
			return -EOPNOTSUPP;

		if (!ops) {
2634
#ifdef CONFIG_MODULES
P
Patrick McHardy 已提交
2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646
			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;
		}

2647 2648 2649
		if (!ops->setup)
			return -EOPNOTSUPP;

2650
		if (!ifname[0]) {
P
Patrick McHardy 已提交
2651
			snprintf(ifname, IFNAMSIZ, "%s%%d", ops->kind);
2652 2653
			name_assign_type = NET_NAME_ENUM;
		}
2654

2655
		dest_net = rtnl_link_get_net(net, tb);
2656 2657 2658
		if (IS_ERR(dest_net))
			return PTR_ERR(dest_net);

2659 2660 2661 2662
		err = -EPERM;
		if (!netlink_ns_capable(skb, dest_net->user_ns, CAP_NET_ADMIN))
			goto out;

2663 2664 2665 2666 2667 2668 2669 2670
		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;
			}
2671 2672 2673
			err = -EPERM;
			if (!netlink_ns_capable(skb, link_net->user_ns, CAP_NET_ADMIN))
				goto out;
2674 2675 2676 2677
		}

		dev = rtnl_create_link(link_net ? : dest_net, ifname,
				       name_assign_type, ops, tb);
2678
		if (IS_ERR(dev)) {
2679
			err = PTR_ERR(dev);
2680 2681 2682 2683 2684
			goto out;
		}

		dev->ifindex = ifm->ifi_index;

2685
		if (ops->newlink) {
2686 2687
			err = ops->newlink(link_net ? : net, dev, tb, data,
					   extack);
2688
			/* Drivers should call free_netdev() in ->destructor
2689 2690
			 * and unregister it on failure after registration
			 * so that device could be finally freed in rtnl_unlock.
2691
			 */
2692 2693 2694 2695
			if (err < 0) {
				/* If device is not registered at all, free it now */
				if (dev->reg_state == NETREG_UNINITIALIZED)
					free_netdev(dev);
2696
				goto out;
2697
			}
2698
		} else {
2699
			err = register_netdevice(dev);
2700 2701 2702 2703
			if (err < 0) {
				free_netdev(dev);
				goto out;
			}
2704
		}
2705
		err = rtnl_configure_link(dev, ifm);
2706 2707
		if (err < 0)
			goto out_unregister;
2708
		if (link_net) {
2709
			err = dev_change_net_namespace(dev, dest_net, ifname);
2710
			if (err < 0)
2711
				goto out_unregister;
2712
		}
2713 2714 2715 2716 2717
		if (tb[IFLA_MASTER]) {
			err = do_set_master(dev, nla_get_u32(tb[IFLA_MASTER]));
			if (err)
				goto out_unregister;
		}
2718
out:
2719 2720
		if (link_net)
			put_net(link_net);
2721
		put_net(dest_net);
P
Patrick McHardy 已提交
2722
		return err;
2723 2724 2725 2726 2727 2728 2729 2730 2731 2732
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
2733 2734 2735
	}
}

2736 2737
static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
2738
{
2739
	struct net *net = sock_net(skb->sk);
2740
	struct ifinfomsg *ifm;
2741
	char ifname[IFNAMSIZ];
2742 2743 2744
	struct nlattr *tb[IFLA_MAX+1];
	struct net_device *dev = NULL;
	struct sk_buff *nskb;
2745
	int err;
2746
	u32 ext_filter_mask = 0;
2747

2748
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy, extack);
2749
	if (err < 0)
2750
		return err;
2751

2752 2753 2754
	if (tb[IFLA_IFNAME])
		nla_strlcpy(ifname, tb[IFLA_IFNAME], IFNAMSIZ);

2755 2756 2757
	if (tb[IFLA_EXT_MASK])
		ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);

2758
	ifm = nlmsg_data(nlh);
2759 2760 2761 2762 2763
	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
2764 2765
		return -EINVAL;

2766 2767 2768
	if (dev == NULL)
		return -ENODEV;

2769
	nskb = nlmsg_new(if_nlmsg_size(dev, ext_filter_mask), GFP_KERNEL);
2770 2771
	if (nskb == NULL)
		return -ENOBUFS;
2772

2773
	err = rtnl_fill_ifinfo(nskb, dev, RTM_NEWLINK, NETLINK_CB(skb).portid,
2774
			       nlh->nlmsg_seq, 0, 0, ext_filter_mask, 0);
2775 2776 2777 2778
	if (err < 0) {
		/* -EMSGSIZE implies BUG in if_nlmsg_size */
		WARN_ON(err == -EMSGSIZE);
		kfree_skb(nskb);
2779
	} else
2780
		err = rtnl_unicast(nskb, net, NETLINK_CB(skb).portid);
2781

2782
	return err;
2783 2784
}

2785
static u16 rtnl_calcit(struct sk_buff *skb, struct nlmsghdr *nlh)
2786
{
2787 2788 2789 2790 2791
	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;
2792 2793 2794 2795 2796
	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);
2797

2798
	if (nlmsg_parse(nlh, hdrlen, tb, IFLA_MAX, ifla_policy, NULL) >= 0) {
2799 2800 2801
		if (tb[IFLA_EXT_MASK])
			ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);
	}
2802 2803 2804 2805 2806 2807 2808

	if (!ext_filter_mask)
		return NLMSG_GOODSIZE;
	/*
	 * traverse the list of net devices and compute the minimum
	 * buffer size based upon the filter mask.
	 */
2809 2810
	rcu_read_lock();
	for_each_netdev_rcu(net, dev) {
2811 2812 2813 2814
		min_ifinfo_dump_size = max_t(u16, min_ifinfo_dump_size,
					     if_nlmsg_size(dev,
						           ext_filter_mask));
	}
2815
	rcu_read_unlock();
2816

2817
	return nlmsg_total_size(min_ifinfo_dump_size);
2818 2819
}

A
Adrian Bunk 已提交
2820
static int rtnl_dump_all(struct sk_buff *skb, struct netlink_callback *cb)
L
Linus Torvalds 已提交
2821 2822 2823 2824 2825 2826
{
	int idx;
	int s_idx = cb->family;

	if (s_idx == 0)
		s_idx = 1;
2827

2828
	for (idx = 1; idx <= RTNL_FAMILY_MAX; idx++) {
L
Linus Torvalds 已提交
2829
		int type = cb->nlh->nlmsg_type-RTM_BASE;
2830 2831 2832
		struct rtnl_link *handlers;
		rtnl_dumpit_func dumpit;

L
Linus Torvalds 已提交
2833 2834
		if (idx < s_idx || idx == PF_PACKET)
			continue;
2835 2836 2837

		handlers = rtnl_dereference(rtnl_msg_handlers[idx]);
		if (!handlers)
L
Linus Torvalds 已提交
2838
			continue;
2839 2840 2841 2842 2843

		dumpit = READ_ONCE(handlers[type].dumpit);
		if (!dumpit)
			continue;

2844
		if (idx > s_idx) {
L
Linus Torvalds 已提交
2845
			memset(&cb->args[0], 0, sizeof(cb->args));
2846 2847 2848
			cb->prev_seq = 0;
			cb->seq = 0;
		}
2849
		if (dumpit(skb, cb))
L
Linus Torvalds 已提交
2850 2851 2852 2853 2854 2855 2856
			break;
	}
	cb->family = idx;

	return skb->len;
}

2857
struct sk_buff *rtmsg_ifinfo_build_skb(int type, struct net_device *dev,
2858 2859
				       unsigned int change,
				       u32 event, gfp_t flags)
L
Linus Torvalds 已提交
2860
{
2861
	struct net *net = dev_net(dev);
L
Linus Torvalds 已提交
2862
	struct sk_buff *skb;
2863
	int err = -ENOBUFS;
2864
	size_t if_info_size;
L
Linus Torvalds 已提交
2865

2866
	skb = nlmsg_new((if_info_size = if_nlmsg_size(dev, 0)), flags);
2867 2868
	if (skb == NULL)
		goto errout;
L
Linus Torvalds 已提交
2869

2870
	err = rtnl_fill_ifinfo(skb, dev, type, 0, 0, change, 0, 0, event);
2871 2872 2873 2874 2875 2876
	if (err < 0) {
		/* -EMSGSIZE implies BUG in if_nlmsg_size() */
		WARN_ON(err == -EMSGSIZE);
		kfree_skb(skb);
		goto errout;
	}
2877
	return skb;
2878 2879
errout:
	if (err < 0)
2880
		rtnl_set_sk_err(net, RTNLGRP_LINK, err);
2881 2882 2883 2884 2885 2886 2887 2888 2889 2890
	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);
}

2891 2892 2893
static void rtmsg_ifinfo_event(int type, struct net_device *dev,
			       unsigned int change, u32 event,
			       gfp_t flags)
2894 2895 2896
{
	struct sk_buff *skb;

2897 2898 2899
	if (dev->reg_state != NETREG_REGISTERED)
		return;

2900
	skb = rtmsg_ifinfo_build_skb(type, dev, change, event, flags);
2901 2902
	if (skb)
		rtmsg_ifinfo_send(skb, dev, flags);
L
Linus Torvalds 已提交
2903
}
2904 2905 2906 2907

void rtmsg_ifinfo(int type, struct net_device *dev, unsigned int change,
		  gfp_t flags)
{
2908
	rtmsg_ifinfo_event(type, dev, change, rtnl_get_event(0), flags);
2909
}
2910
EXPORT_SYMBOL(rtmsg_ifinfo);
L
Linus Torvalds 已提交
2911

2912 2913
static int nlmsg_populate_fdb_fill(struct sk_buff *skb,
				   struct net_device *dev,
2914
				   u8 *addr, u16 vid, u32 pid, u32 seq,
2915
				   int type, unsigned int flags,
2916
				   int nlflags, u16 ndm_state)
2917 2918 2919 2920
{
	struct nlmsghdr *nlh;
	struct ndmsg *ndm;

2921
	nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ndm), nlflags);
2922 2923 2924 2925 2926 2927 2928 2929 2930 2931
	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;
2932
	ndm->ndm_state   = ndm_state;
2933 2934 2935

	if (nla_put(skb, NDA_LLADDR, ETH_ALEN, addr))
		goto nla_put_failure;
2936 2937 2938
	if (vid)
		if (nla_put(skb, NDA_VLAN, sizeof(u16), &vid))
			goto nla_put_failure;
2939

2940 2941
	nlmsg_end(skb, nlh);
	return 0;
2942 2943 2944 2945 2946 2947

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

2948 2949
static inline size_t rtnl_fdb_nlmsg_size(void)
{
2950 2951 2952 2953
	return NLMSG_ALIGN(sizeof(struct ndmsg)) +
	       nla_total_size(ETH_ALEN) +	/* NDA_LLADDR */
	       nla_total_size(sizeof(u16)) +	/* NDA_VLAN */
	       0;
2954 2955
}

2956 2957
static void rtnl_fdb_notify(struct net_device *dev, u8 *addr, u16 vid, int type,
			    u16 ndm_state)
2958 2959 2960 2961 2962 2963 2964 2965 2966
{
	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;

2967
	err = nlmsg_populate_fdb_fill(skb, dev, addr, vid,
2968
				      0, 0, type, NTF_SELF, 0, ndm_state);
2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979
	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);
}

2980 2981 2982 2983 2984 2985
/**
 * 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,
2986
		     const unsigned char *addr, u16 vid,
2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998
		     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;
	}

2999 3000 3001 3002 3003
	if (vid) {
		pr_info("%s: vlans aren't supported yet for dev_uc|mc_add()\n", dev->name);
		return err;
	}

3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016
	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);

3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038
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;
}

3039 3040
static int rtnl_fdb_add(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
3041 3042 3043 3044 3045 3046
{
	struct net *net = sock_net(skb->sk);
	struct ndmsg *ndm;
	struct nlattr *tb[NDA_MAX+1];
	struct net_device *dev;
	u8 *addr;
3047
	u16 vid;
3048 3049
	int err;

3050
	err = nlmsg_parse(nlh, sizeof(*ndm), tb, NDA_MAX, NULL, extack);
3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072
	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]);

3073 3074 3075 3076
	err = fdb_vid_parse(tb[NDA_VLAN], &vid);
	if (err)
		return err;

3077 3078 3079 3080 3081
	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)) {
3082 3083 3084
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);
		const struct net_device_ops *ops = br_dev->netdev_ops;

3085 3086
		err = ops->ndo_fdb_add(ndm, tb, dev, addr, vid,
				       nlh->nlmsg_flags);
3087 3088 3089 3090 3091 3092 3093
		if (err)
			goto out;
		else
			ndm->ndm_flags &= ~NTF_MASTER;
	}

	/* Embedded bridge, macvlan, and any other device support */
3094 3095 3096
	if ((ndm->ndm_flags & NTF_SELF)) {
		if (dev->netdev_ops->ndo_fdb_add)
			err = dev->netdev_ops->ndo_fdb_add(ndm, tb, dev, addr,
3097
							   vid,
3098 3099
							   nlh->nlmsg_flags);
		else
3100
			err = ndo_dflt_fdb_add(ndm, tb, dev, addr, vid,
3101
					       nlh->nlmsg_flags);
3102

3103
		if (!err) {
3104 3105
			rtnl_fdb_notify(dev, addr, vid, RTM_NEWNEIGH,
					ndm->ndm_state);
3106
			ndm->ndm_flags &= ~NTF_SELF;
3107
		}
3108 3109 3110 3111 3112
	}
out:
	return err;
}

3113 3114 3115 3116 3117 3118
/**
 * 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,
3119
		     const unsigned char *addr, u16 vid)
3120
{
3121
	int err = -EINVAL;
3122 3123 3124 3125

	/* If aging addresses are supported device will need to
	 * implement its own handler for this.
	 */
3126
	if (!(ndm->ndm_state & NUD_PERMANENT)) {
3127
		pr_info("%s: FDB only supports static addresses\n", dev->name);
3128
		return err;
3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139
	}

	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);

3140 3141
static int rtnl_fdb_del(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
3142 3143 3144
{
	struct net *net = sock_net(skb->sk);
	struct ndmsg *ndm;
3145
	struct nlattr *tb[NDA_MAX+1];
3146 3147 3148
	struct net_device *dev;
	int err = -EINVAL;
	__u8 *addr;
3149
	u16 vid;
3150

3151
	if (!netlink_capable(skb, CAP_NET_ADMIN))
3152 3153
		return -EPERM;

3154
	err = nlmsg_parse(nlh, sizeof(*ndm), tb, NDA_MAX, NULL, extack);
3155 3156
	if (err < 0)
		return err;
3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169

	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;
	}

3170 3171 3172 3173 3174 3175
	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]);
3176

3177 3178 3179 3180
	err = fdb_vid_parse(tb[NDA_VLAN], &vid);
	if (err)
		return err;

3181 3182 3183 3184 3185
	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)) {
3186 3187
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);
		const struct net_device_ops *ops = br_dev->netdev_ops;
3188

3189
		if (ops->ndo_fdb_del)
3190
			err = ops->ndo_fdb_del(ndm, tb, dev, addr, vid);
3191 3192 3193 3194 3195 3196 3197 3198

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

	/* Embedded bridge, macvlan, and any other device support */
3199 3200
	if (ndm->ndm_flags & NTF_SELF) {
		if (dev->netdev_ops->ndo_fdb_del)
3201 3202
			err = dev->netdev_ops->ndo_fdb_del(ndm, tb, dev, addr,
							   vid);
3203
		else
3204
			err = ndo_dflt_fdb_del(ndm, tb, dev, addr, vid);
3205

3206
		if (!err) {
3207 3208
			rtnl_fdb_notify(dev, addr, vid, RTM_DELNEIGH,
					ndm->ndm_state);
3209
			ndm->ndm_flags &= ~NTF_SELF;
3210
		}
3211 3212 3213 3214 3215
	}
out:
	return err;
}

3216 3217 3218 3219 3220 3221 3222 3223
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;
3224
	u32 portid, seq;
3225

3226
	portid = NETLINK_CB(cb->skb).portid;
3227 3228 3229
	seq = cb->nlh->nlmsg_seq;

	list_for_each_entry(ha, &list->list, list) {
3230
		if (*idx < cb->args[2])
3231 3232
			goto skip;

3233
		err = nlmsg_populate_fdb_fill(skb, dev, ha->addr, 0,
3234
					      portid, seq,
3235
					      RTM_NEWNEIGH, NTF_SELF,
3236
					      NLM_F_MULTI, NUD_PERMANENT);
3237 3238 3239 3240 3241 3242 3243 3244 3245
		if (err < 0)
			return err;
skip:
		*idx += 1;
	}
	return 0;
}

/**
3246
 * ndo_dflt_fdb_dump - default netdevice operation to dump an FDB table.
3247 3248 3249 3250
 * @nlh: netlink message header
 * @dev: netdevice
 *
 * Default netdevice operation to dump the existing unicast address list.
3251
 * Returns number of addresses from list put in skb.
3252 3253 3254 3255
 */
int ndo_dflt_fdb_dump(struct sk_buff *skb,
		      struct netlink_callback *cb,
		      struct net_device *dev,
3256
		      struct net_device *filter_dev,
3257
		      int *idx)
3258 3259 3260 3261
{
	int err;

	netif_addr_lock_bh(dev);
3262
	err = nlmsg_populate_fdb(skb, cb, dev, idx, &dev->uc);
3263 3264
	if (err)
		goto out;
3265
	err = nlmsg_populate_fdb(skb, cb, dev, idx, &dev->mc);
3266 3267
out:
	netif_addr_unlock_bh(dev);
3268
	return err;
3269 3270 3271
}
EXPORT_SYMBOL(ndo_dflt_fdb_dump);

3272 3273 3274
static int rtnl_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb)
{
	struct net_device *dev;
3275 3276 3277 3278 3279 3280
	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);
3281
	struct hlist_head *head;
3282 3283
	int brport_idx = 0;
	int br_idx = 0;
3284 3285 3286 3287
	int h, s_h;
	int idx = 0, s_idx;
	int err = 0;
	int fidx = 0;
3288

3289 3290 3291 3292 3293
	err = nlmsg_parse(cb->nlh, sizeof(struct ifinfomsg), tb,
			  IFLA_MAX, ifla_policy, NULL);
	if (err < 0) {
		return -EINVAL;
	} else if (err == 0) {
3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307
		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;
	}

3308 3309
	s_h = cb->args[0];
	s_idx = cb->args[1];
3310

3311 3312 3313 3314
	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) {
3315

3316
			if (brport_idx && (dev->ifindex != brport_idx))
3317 3318
				continue;

3319 3320 3321 3322 3323 3324 3325 3326 3327
			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;
3328

3329 3330 3331 3332 3333
				if (br_dev != netdev_master_upper_dev_get(dev) &&
				    !(dev->priv_flags & IFF_EBRIDGE))
					continue;
				cops = ops;
			}
3334

3335 3336
			if (idx < s_idx)
				goto cont;
3337

3338 3339 3340 3341 3342 3343 3344 3345 3346
			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;
				}
			}
3347

3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365
			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++;
		}
3366 3367
	}

3368 3369 3370 3371 3372
out:
	cb->args[0] = h;
	cb->args[1] = idx;
	cb->args[2] = fidx;

3373 3374 3375
	return skb->len;
}

3376 3377 3378 3379 3380 3381 3382 3383
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;
}

3384
int ndo_dflt_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
3385
			    struct net_device *dev, u16 mode,
3386 3387 3388 3389 3390
			    u32 flags, u32 mask, int nlflags,
			    u32 filter_mask,
			    int (*vlan_fill)(struct sk_buff *skb,
					     struct net_device *dev,
					     u32 filter_mask))
3391 3392 3393 3394
{
	struct nlmsghdr *nlh;
	struct ifinfomsg *ifm;
	struct nlattr *br_afspec;
3395
	struct nlattr *protinfo;
3396
	u8 operstate = netif_running(dev) ? dev->operstate : IF_OPER_DOWN;
3397
	struct net_device *br_dev = netdev_master_upper_dev_get(dev);
3398
	int err = 0;
3399

3400
	nlh = nlmsg_put(skb, pid, seq, RTM_NEWLINK, sizeof(*ifm), nlflags);
3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415
	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) ||
3416 3417
	    (br_dev &&
	     nla_put_u32(skb, IFLA_MASTER, br_dev->ifindex)) ||
3418 3419
	    (dev->addr_len &&
	     nla_put(skb, IFLA_ADDRESS, dev->addr_len, dev->dev_addr)) ||
3420 3421
	    (dev->ifindex != dev_get_iflink(dev) &&
	     nla_put_u32(skb, IFLA_LINK, dev_get_iflink(dev))))
3422 3423 3424 3425 3426 3427
		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
3428
	if (nla_put_u16(skb, IFLA_BRIDGE_FLAGS, BRIDGE_FLAGS_SELF)) {
3429 3430 3431
		nla_nest_cancel(skb, br_afspec);
		goto nla_put_failure;
	}
R
Roopa Prabhu 已提交
3432 3433 3434 3435 3436 3437 3438

	if (mode != BRIDGE_MODE_UNDEF) {
		if (nla_put_u16(skb, IFLA_BRIDGE_MODE, mode)) {
			nla_nest_cancel(skb, br_afspec);
			goto nla_put_failure;
		}
	}
3439 3440 3441 3442 3443 3444 3445
	if (vlan_fill) {
		err = vlan_fill(skb, dev, filter_mask);
		if (err) {
			nla_nest_cancel(skb, br_afspec);
			goto nla_put_failure;
		}
	}
3446 3447
	nla_nest_end(skb, br_afspec);

3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474
	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);

3475 3476
	nlmsg_end(skb, nlh);
	return 0;
3477 3478
nla_put_failure:
	nlmsg_cancel(skb, nlh);
3479
	return err ? err : -EMSGSIZE;
3480
}
3481
EXPORT_SYMBOL_GPL(ndo_dflt_bridge_getlink);
3482

J
John Fastabend 已提交
3483 3484 3485 3486 3487 3488 3489
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;
3490
	u32 filter_mask = 0;
3491
	int err;
3492

3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504
	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
3505 3506 3507 3508

	rcu_read_lock();
	for_each_netdev_rcu(net, dev) {
		const struct net_device_ops *ops = dev->netdev_ops;
3509
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);
J
John Fastabend 已提交
3510

3511
		if (br_dev && br_dev->netdev_ops->ndo_bridge_getlink) {
3512 3513 3514 3515
			if (idx >= cb->args[0]) {
				err = br_dev->netdev_ops->ndo_bridge_getlink(
						skb, portid, seq, dev,
						filter_mask, NLM_F_MULTI);
3516 3517 3518 3519 3520 3521
				if (err < 0 && err != -EOPNOTSUPP) {
					if (likely(skb->len))
						break;

					goto out_err;
				}
3522
			}
3523
			idx++;
J
John Fastabend 已提交
3524 3525 3526
		}

		if (ops->ndo_bridge_getlink) {
3527 3528 3529 3530 3531
			if (idx >= cb->args[0]) {
				err = ops->ndo_bridge_getlink(skb, portid,
							      seq, dev,
							      filter_mask,
							      NLM_F_MULTI);
3532 3533 3534 3535 3536 3537
				if (err < 0 && err != -EOPNOTSUPP) {
					if (likely(skb->len))
						break;

					goto out_err;
				}
3538
			}
3539
			idx++;
J
John Fastabend 已提交
3540 3541
		}
	}
3542 3543
	err = skb->len;
out_err:
J
John Fastabend 已提交
3544 3545 3546
	rcu_read_unlock();
	cb->args[0] = idx;

3547
	return err;
J
John Fastabend 已提交
3548 3549
}

3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564
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 */
}

3565
static int rtnl_bridge_notify(struct net_device *dev)
3566 3567 3568 3569 3570
{
	struct net *net = dev_net(dev);
	struct sk_buff *skb;
	int err = -EOPNOTSUPP;

3571 3572 3573
	if (!dev->netdev_ops->ndo_bridge_getlink)
		return 0;

3574 3575 3576 3577 3578 3579
	skb = nlmsg_new(bridge_nlmsg_size(), GFP_ATOMIC);
	if (!skb) {
		err = -ENOMEM;
		goto errout;
	}

3580
	err = dev->netdev_ops->ndo_bridge_getlink(skb, 0, 0, dev, 0, 0);
3581 3582
	if (err < 0)
		goto errout;
3583

3584 3585 3586
	if (!skb->len)
		goto errout;

3587 3588 3589 3590 3591
	rtnl_notify(skb, net, 0, RTNLGRP_LINK, NULL, GFP_ATOMIC);
	return 0;
errout:
	WARN_ON(err == -EMSGSIZE);
	kfree_skb(skb);
3592 3593
	if (err)
		rtnl_set_sk_err(net, RTNLGRP_LINK, err);
3594 3595 3596
	return err;
}

3597 3598
static int rtnl_bridge_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			       struct netlink_ext_ack *extack)
J
John Fastabend 已提交
3599 3600 3601 3602
{
	struct net *net = sock_net(skb->sk);
	struct ifinfomsg *ifm;
	struct net_device *dev;
3603 3604
	struct nlattr *br_spec, *attr = NULL;
	int rem, err = -EOPNOTSUPP;
3605
	u16 flags = 0;
3606
	bool have_flags = false;
J
John Fastabend 已提交
3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620

	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;
	}

3621 3622 3623 3624
	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) {
3625 3626 3627
				if (nla_len(attr) < sizeof(flags))
					return -EINVAL;

3628
				have_flags = true;
3629 3630 3631 3632 3633 3634 3635
				flags = nla_get_u16(attr);
				break;
			}
		}
	}

	if (!flags || (flags & BRIDGE_FLAGS_MASTER)) {
3636 3637 3638
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);

		if (!br_dev || !br_dev->netdev_ops->ndo_bridge_setlink) {
3639 3640 3641 3642
			err = -EOPNOTSUPP;
			goto out;
		}

3643
		err = br_dev->netdev_ops->ndo_bridge_setlink(dev, nlh, flags);
J
John Fastabend 已提交
3644 3645
		if (err)
			goto out;
3646 3647

		flags &= ~BRIDGE_FLAGS_MASTER;
J
John Fastabend 已提交
3648 3649
	}

3650 3651 3652 3653
	if ((flags & BRIDGE_FLAGS_SELF)) {
		if (!dev->netdev_ops->ndo_bridge_setlink)
			err = -EOPNOTSUPP;
		else
3654 3655
			err = dev->netdev_ops->ndo_bridge_setlink(dev, nlh,
								  flags);
3656
		if (!err) {
3657
			flags &= ~BRIDGE_FLAGS_SELF;
3658 3659 3660 3661 3662 3663

			/* Generate event to notify upper layer of bridge
			 * change
			 */
			err = rtnl_bridge_notify(dev);
		}
3664
	}
J
John Fastabend 已提交
3665

3666
	if (have_flags)
3667
		memcpy(nla_data(attr), &flags, sizeof(flags));
J
John Fastabend 已提交
3668 3669 3670 3671
out:
	return err;
}

3672 3673
static int rtnl_bridge_dellink(struct sk_buff *skb, struct nlmsghdr *nlh,
			       struct netlink_ext_ack *extack)
3674 3675 3676 3677 3678 3679
{
	struct net *net = sock_net(skb->sk);
	struct ifinfomsg *ifm;
	struct net_device *dev;
	struct nlattr *br_spec, *attr = NULL;
	int rem, err = -EOPNOTSUPP;
3680
	u16 flags = 0;
3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 3697 3698 3699
	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) {
3700 3701 3702
				if (nla_len(attr) < sizeof(flags))
					return -EINVAL;

3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717
				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;
		}

3718
		err = br_dev->netdev_ops->ndo_bridge_dellink(dev, nlh, flags);
3719 3720 3721 3722 3723 3724 3725 3726 3727 3728
		if (err)
			goto out;

		flags &= ~BRIDGE_FLAGS_MASTER;
	}

	if ((flags & BRIDGE_FLAGS_SELF)) {
		if (!dev->netdev_ops->ndo_bridge_dellink)
			err = -EOPNOTSUPP;
		else
3729 3730
			err = dev->netdev_ops->ndo_bridge_dellink(dev, nlh,
								  flags);
3731

3732
		if (!err) {
3733
			flags &= ~BRIDGE_FLAGS_SELF;
3734 3735 3736 3737 3738 3739

			/* Generate event to notify upper layer of bridge
			 * change
			 */
			err = rtnl_bridge_notify(dev);
		}
3740 3741 3742 3743 3744 3745 3746 3747
	}

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

3748 3749 3750 3751 3752 3753
static bool stats_attr_valid(unsigned int mask, int attrid, int idxattr)
{
	return (mask & IFLA_STATS_FILTER_BIT(attrid)) &&
	       (!idxattr || idxattr == attrid);
}

3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785
#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;

3786
		if (!dev->netdev_ops->ndo_has_offload_stats(dev, attr_id))
3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 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
			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++) {
3827
		if (!dev->netdev_ops->ndo_has_offload_stats(dev, attr_id))
3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838
			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;
}

3839 3840
static int rtnl_fill_statsinfo(struct sk_buff *skb, struct net_device *dev,
			       int type, u32 pid, u32 seq, u32 change,
3841 3842
			       unsigned int flags, unsigned int filter_mask,
			       int *idxattr, int *prividx)
3843 3844 3845 3846
{
	struct if_stats_msg *ifsm;
	struct nlmsghdr *nlh;
	struct nlattr *attr;
3847
	int s_prividx = *prividx;
3848
	int err;
3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859

	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;

3860
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_64, *idxattr)) {
3861 3862
		struct rtnl_link_stats64 *sp;

3863 3864 3865
		attr = nla_reserve_64bit(skb, IFLA_STATS_LINK_64,
					 sizeof(struct rtnl_link_stats64),
					 IFLA_STATS_UNSPEC);
3866 3867 3868 3869 3870 3871 3872
		if (!attr)
			goto nla_put_failure;

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

3873 3874 3875 3876 3877 3878 3879 3880 3881 3882
	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;

3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906
			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);
3907 3908 3909 3910 3911 3912 3913
			nla_nest_end(skb, attr);
			if (err)
				goto nla_put_failure;
			*idxattr = 0;
		}
	}

3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928 3929 3930 3931
	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;
	}

3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964
	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;
	}

3965 3966 3967 3968 3969
	nlmsg_end(skb, nlh);

	return 0;

nla_put_failure:
3970 3971 3972 3973 3974
	/* 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);
3975 3976 3977 3978 3979 3980 3981 3982 3983

	return -EMSGSIZE;
}

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

3984
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_64, 0))
3985 3986
		size += nla_total_size_64bit(sizeof(struct rtnl_link_stats64));

3987 3988
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_XSTATS, 0)) {
		const struct rtnl_link_ops *ops = dev->rtnl_link_ops;
3989
		int attr = IFLA_STATS_LINK_XSTATS;
3990 3991

		if (ops && ops->get_linkxstats_size) {
3992 3993
			size += nla_total_size(ops->get_linkxstats_size(dev,
									attr));
3994 3995 3996 3997 3998
			/* for IFLA_STATS_LINK_XSTATS */
			size += nla_total_size(0);
		}
	}

3999 4000 4001 4002 4003 4004 4005 4006 4007 4008 4009 4010 4011 4012 4013 4014 4015 4016 4017
	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);
		}
	}

4018 4019 4020
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_OFFLOAD_XSTATS, 0))
		size += rtnl_get_offload_stats_size(dev);

4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037
	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);
			}
		}
	}

4038 4039 4040
	return size;
}

4041 4042
static int rtnl_stats_get(struct sk_buff *skb, struct nlmsghdr *nlh,
			  struct netlink_ext_ack *extack)
4043 4044 4045
{
	struct net *net = sock_net(skb->sk);
	struct net_device *dev = NULL;
4046 4047
	int idxattr = 0, prividx = 0;
	struct if_stats_msg *ifsm;
4048 4049 4050 4051
	struct sk_buff *nskb;
	u32 filter_mask;
	int err;

4052 4053 4054
	if (nlmsg_len(nlh) < sizeof(*ifsm))
		return -EINVAL;

4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073
	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,
4074
				  0, filter_mask, &idxattr, &prividx);
4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086 4087
	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)
{
4088
	int h, s_h, err, s_idx, s_idxattr, s_prividx;
4089
	struct net *net = sock_net(skb->sk);
4090
	unsigned int flags = NLM_F_MULTI;
4091 4092
	struct if_stats_msg *ifsm;
	struct hlist_head *head;
4093
	struct net_device *dev;
4094
	u32 filter_mask = 0;
4095
	int idx = 0;
4096 4097 4098

	s_h = cb->args[0];
	s_idx = cb->args[1];
4099 4100
	s_idxattr = cb->args[2];
	s_prividx = cb->args[3];
4101 4102 4103

	cb->seq = net->dev_base_seq;

4104 4105 4106
	if (nlmsg_len(cb->nlh) < sizeof(*ifsm))
		return -EINVAL;

4107 4108 4109 4110 4111 4112 4113 4114 4115 4116 4117 4118 4119 4120
	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,
4121 4122
						  flags, filter_mask,
						  &s_idxattr, &s_prividx);
4123 4124 4125 4126 4127 4128 4129
			/* 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;
4130 4131
			s_prividx = 0;
			s_idxattr = 0;
4132 4133 4134 4135 4136 4137
			nl_dump_check_consistent(cb, nlmsg_hdr(skb));
cont:
			idx++;
		}
	}
out:
4138 4139
	cb->args[3] = s_prividx;
	cb->args[2] = s_idxattr;
4140 4141 4142 4143 4144 4145
	cb->args[1] = idx;
	cb->args[0] = h;

	return skb->len;
}

L
Linus Torvalds 已提交
4146 4147
/* Process one rtnetlink message. */

J
Johannes Berg 已提交
4148 4149
static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
			     struct netlink_ext_ack *extack)
L
Linus Torvalds 已提交
4150
{
4151
	struct net *net = sock_net(skb->sk);
4152 4153
	struct rtnl_link *handlers;
	int err = -EOPNOTSUPP;
4154
	rtnl_doit_func doit;
4155
	unsigned int flags;
4156
	int kind;
L
Linus Torvalds 已提交
4157 4158 4159 4160 4161
	int family;
	int type;

	type = nlh->nlmsg_type;
	if (type > RTM_MAX)
4162
		return -EOPNOTSUPP;
L
Linus Torvalds 已提交
4163 4164 4165 4166

	type -= RTM_BASE;

	/* All the messages must have at least 1 byte length */
4167
	if (nlmsg_len(nlh) < sizeof(struct rtgenmsg))
L
Linus Torvalds 已提交
4168 4169
		return 0;

4170
	family = ((struct rtgenmsg *)nlmsg_data(nlh))->rtgen_family;
L
Linus Torvalds 已提交
4171 4172
	kind = type&3;

4173
	if (kind != 2 && !netlink_net_capable(skb, CAP_NET_ADMIN))
4174
		return -EPERM;
L
Linus Torvalds 已提交
4175

4176
	if (family >= ARRAY_SIZE(rtnl_msg_handlers))
4177 4178 4179 4180 4181 4182 4183 4184 4185
		family = PF_UNSPEC;

	rcu_read_lock();
	handlers = rcu_dereference(rtnl_msg_handlers[family]);
	if (!handlers) {
		family = PF_UNSPEC;
		handlers = rcu_dereference(rtnl_msg_handlers[family]);
	}

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

4191 4192 4193 4194 4195 4196
		dumpit = READ_ONCE(handlers[type].dumpit);
		if (!dumpit) {
			family = PF_UNSPEC;
			handlers = rcu_dereference(rtnl_msg_handlers[PF_UNSPEC]);
			if (!handlers)
				goto err_unlock;
4197

4198 4199 4200 4201
			dumpit = READ_ONCE(handlers[type].dumpit);
			if (!dumpit)
				goto err_unlock;
		}
4202

4203 4204
		refcount_inc(&rtnl_msg_handlers_ref[family]);

4205
		if (type == RTM_GETLINK - RTM_BASE)
4206
			min_dump_alloc = rtnl_calcit(skb, nlh);
4207

4208 4209
		rcu_read_unlock();

4210
		rtnl = net->rtnl;
4211 4212 4213 4214 4215 4216 4217
		{
			struct netlink_dump_control c = {
				.dump		= dumpit,
				.min_dump_alloc	= min_dump_alloc,
			};
			err = netlink_dump_start(rtnl, skb, nlh, &c);
		}
4218
		refcount_dec(&rtnl_msg_handlers_ref[family]);
4219
		return err;
L
Linus Torvalds 已提交
4220 4221
	}

4222 4223 4224 4225 4226 4227
	doit = READ_ONCE(handlers[type].doit);
	if (!doit) {
		family = PF_UNSPEC;
		handlers = rcu_dereference(rtnl_msg_handlers[family]);
	}

4228 4229 4230 4231 4232 4233 4234 4235 4236 4237 4238
	flags = READ_ONCE(handlers[type].flags);
	if (flags & RTNL_FLAG_DOIT_UNLOCKED) {
		refcount_inc(&rtnl_msg_handlers_ref[family]);
		doit = READ_ONCE(handlers[type].doit);
		rcu_read_unlock();
		if (doit)
			err = doit(skb, nlh, extack);
		refcount_dec(&rtnl_msg_handlers_ref[family]);
		return err;
	}

4239
	rcu_read_unlock();
L
Linus Torvalds 已提交
4240

4241 4242 4243 4244 4245 4246 4247
	rtnl_lock();
	handlers = rtnl_dereference(rtnl_msg_handlers[family]);
	if (handlers) {
		doit = READ_ONCE(handlers[type].doit);
		if (doit)
			err = doit(skb, nlh, extack);
	}
4248 4249 4250 4251
	rtnl_unlock();
	return err;

err_unlock:
4252
	rcu_read_unlock();
4253
	return -EOPNOTSUPP;
L
Linus Torvalds 已提交
4254 4255
}

4256
static void rtnetlink_rcv(struct sk_buff *skb)
L
Linus Torvalds 已提交
4257
{
4258
	netlink_rcv_skb(skb, &rtnetlink_rcv_msg);
L
Linus Torvalds 已提交
4259 4260
}

4261 4262 4263 4264 4265 4266 4267 4268 4269 4270 4271 4272
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
4273 4274
static int rtnetlink_event(struct notifier_block *this, unsigned long event, void *ptr)
{
4275
	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
4276

L
Linus Torvalds 已提交
4277
	switch (event) {
4278
	case NETDEV_REBOOT:
4279
	case NETDEV_CHANGEADDR:
4280 4281 4282 4283 4284 4285
	case NETDEV_CHANGENAME:
	case NETDEV_FEAT_CHANGE:
	case NETDEV_BONDING_FAILOVER:
	case NETDEV_NOTIFY_PEERS:
	case NETDEV_RESEND_IGMP:
	case NETDEV_CHANGEINFODATA:
4286 4287
		rtmsg_ifinfo_event(RTM_NEWLINK, dev, 0, rtnl_get_event(event),
				   GFP_KERNEL);
L
Linus Torvalds 已提交
4288 4289 4290 4291 4292 4293 4294 4295 4296 4297 4298
		break;
	default:
		break;
	}
	return NOTIFY_DONE;
}

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

4299

4300
static int __net_init rtnetlink_net_init(struct net *net)
4301 4302
{
	struct sock *sk;
4303 4304 4305 4306
	struct netlink_kernel_cfg cfg = {
		.groups		= RTNLGRP_MAX,
		.input		= rtnetlink_rcv,
		.cb_mutex	= &rtnl_mutex,
4307
		.flags		= NL_CFG_F_NONROOT_RECV,
4308
		.bind		= rtnetlink_bind,
4309 4310
	};

4311
	sk = netlink_kernel_create(net, NETLINK_ROUTE, &cfg);
4312 4313 4314 4315 4316 4317
	if (!sk)
		return -ENOMEM;
	net->rtnl = sk;
	return 0;
}

4318
static void __net_exit rtnetlink_net_exit(struct net *net)
4319
{
4320 4321
	netlink_kernel_release(net->rtnl);
	net->rtnl = NULL;
4322 4323 4324 4325 4326 4327 4328
}

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

L
Linus Torvalds 已提交
4329 4330
void __init rtnetlink_init(void)
{
4331 4332 4333 4334 4335
	int i;

	for (i = 0; i < ARRAY_SIZE(rtnl_msg_handlers_ref); i++)
		refcount_set(&rtnl_msg_handlers_ref[i], 1);

4336
	if (register_pernet_subsys(&rtnetlink_net_ops))
L
Linus Torvalds 已提交
4337
		panic("rtnetlink_init: cannot initialize rtnetlink\n");
4338

L
Linus Torvalds 已提交
4339
	register_netdevice_notifier(&rtnetlink_dev_notifier);
4340

4341
	rtnl_register(PF_UNSPEC, RTM_GETLINK, rtnl_getlink,
4342 4343 4344 4345
		      rtnl_dump_ifinfo, 0);
	rtnl_register(PF_UNSPEC, RTM_SETLINK, rtnl_setlink, NULL, 0);
	rtnl_register(PF_UNSPEC, RTM_NEWLINK, rtnl_newlink, NULL, 0);
	rtnl_register(PF_UNSPEC, RTM_DELLINK, rtnl_dellink, NULL, 0);
4346

4347 4348 4349
	rtnl_register(PF_UNSPEC, RTM_GETADDR, NULL, rtnl_dump_all, 0);
	rtnl_register(PF_UNSPEC, RTM_GETROUTE, NULL, rtnl_dump_all, 0);
	rtnl_register(PF_UNSPEC, RTM_GETNETCONF, NULL, rtnl_dump_all, 0);
4350

4351 4352 4353
	rtnl_register(PF_BRIDGE, RTM_NEWNEIGH, rtnl_fdb_add, NULL, 0);
	rtnl_register(PF_BRIDGE, RTM_DELNEIGH, rtnl_fdb_del, NULL, 0);
	rtnl_register(PF_BRIDGE, RTM_GETNEIGH, NULL, rtnl_fdb_dump, 0);
J
John Fastabend 已提交
4354

4355 4356 4357
	rtnl_register(PF_BRIDGE, RTM_GETLINK, NULL, rtnl_bridge_getlink, 0);
	rtnl_register(PF_BRIDGE, RTM_DELLINK, rtnl_bridge_dellink, NULL, 0);
	rtnl_register(PF_BRIDGE, RTM_SETLINK, rtnl_bridge_setlink, NULL, 0);
4358 4359

	rtnl_register(PF_UNSPEC, RTM_GETSTATS, rtnl_stats_get, rtnl_stats_dump,
4360
		      0);
L
Linus Torvalds 已提交
4361
}