rtnetlink.c 104.5 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();

265 266 267
	while (refcount_read(&rtnl_msg_handlers_ref[protocol]) > 0)
		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);

401 402 403 404 405 406 407 408 409
static size_t rtnl_link_get_slave_info_data_size(const struct net_device *dev)
{
	struct net_device *master_dev;
	const struct rtnl_link_ops *ops;

	master_dev = netdev_master_upper_dev_get((struct net_device *) dev);
	if (!master_dev)
		return 0;
	ops = master_dev->rtnl_link_ops;
410
	if (!ops || !ops->get_slave_size)
411 412 413 414 415 416
		return 0;
	/* IFLA_INFO_SLAVE_DATA + nested data */
	return nla_total_size(sizeof(struct nlattr)) +
	       ops->get_slave_size(master_dev, dev);
}

P
Patrick McHardy 已提交
417 418 419 420 421 422 423 424
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;

425 426
	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
427 428 429

	if (ops->get_size)
		/* IFLA_INFO_DATA + nested data */
430
		size += nla_total_size(sizeof(struct nlattr)) +
P
Patrick McHardy 已提交
431 432 433
			ops->get_size(dev);

	if (ops->get_xstats_size)
434 435
		/* IFLA_INFO_XSTATS */
		size += nla_total_size(ops->get_xstats_size(dev));
P
Patrick McHardy 已提交
436

437 438
	size += rtnl_link_get_slave_info_data_size(dev);

P
Patrick McHardy 已提交
439 440 441
	return size;
}

T
Thomas Graf 已提交
442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461
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.
 */
462
void rtnl_af_register(struct rtnl_af_ops *ops)
T
Thomas Graf 已提交
463 464
{
	rtnl_lock();
465
	list_add_tail(&ops->list, &rtnl_af_ops);
T
Thomas Graf 已提交
466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493
	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);

494 495
static size_t rtnl_link_get_af_size(const struct net_device *dev,
				    u32 ext_filter_mask)
T
Thomas Graf 已提交
496 497 498 499 500 501 502 503 504 505 506
{
	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)) +
507
				af_ops->get_link_af_size(dev, ext_filter_mask);
T
Thomas Graf 已提交
508 509 510 511 512 513
		}
	}

	return size;
}

514
static bool rtnl_have_link_slave_info(const struct net_device *dev)
P
Patrick McHardy 已提交
515
{
516
	struct net_device *master_dev;
P
Patrick McHardy 已提交
517

518
	master_dev = netdev_master_upper_dev_get((struct net_device *) dev);
519
	if (master_dev && master_dev->rtnl_link_ops)
520 521 522 523 524 525 526 527 528 529 530
		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
531

532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564
	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
565
	if (nla_put_string(skb, IFLA_INFO_KIND, ops->kind) < 0)
566
		return -EMSGSIZE;
P
Patrick McHardy 已提交
567 568 569
	if (ops->fill_xstats) {
		err = ops->fill_xstats(skb, dev);
		if (err < 0)
570
			return err;
P
Patrick McHardy 已提交
571 572 573
	}
	if (ops->fill_info) {
		data = nla_nest_start(skb, IFLA_INFO_DATA);
574 575
		if (data == NULL)
			return -EMSGSIZE;
P
Patrick McHardy 已提交
576 577 578 579 580 581 582 583 584
		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);
585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607
	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
608 609 610 611 612 613
err_cancel_link:
	nla_nest_cancel(skb, linkinfo);
out:
	return err;
}

614
int rtnetlink_send(struct sk_buff *skb, struct net *net, u32 pid, unsigned int group, int echo)
L
Linus Torvalds 已提交
615
{
616
	struct sock *rtnl = net->rtnl;
L
Linus Torvalds 已提交
617 618
	int err = 0;

619
	NETLINK_CB(skb).dst_group = group;
L
Linus Torvalds 已提交
620
	if (echo)
621
		refcount_inc(&skb->users);
L
Linus Torvalds 已提交
622 623 624 625 626 627
	netlink_broadcast(rtnl, skb, pid, group, GFP_KERNEL);
	if (echo)
		err = netlink_unicast(rtnl, skb, pid, MSG_DONTWAIT);
	return err;
}

628
int rtnl_unicast(struct sk_buff *skb, struct net *net, u32 pid)
629
{
630 631
	struct sock *rtnl = net->rtnl;

632 633
	return nlmsg_unicast(rtnl, skb, pid);
}
E
Eric Dumazet 已提交
634
EXPORT_SYMBOL(rtnl_unicast);
635

636 637
void rtnl_notify(struct sk_buff *skb, struct net *net, u32 pid, u32 group,
		 struct nlmsghdr *nlh, gfp_t flags)
638
{
639
	struct sock *rtnl = net->rtnl;
640 641 642 643 644
	int report = 0;

	if (nlh)
		report = nlmsg_report(nlh);

645
	nlmsg_notify(rtnl, skb, pid, group, report, flags);
646
}
E
Eric Dumazet 已提交
647
EXPORT_SYMBOL(rtnl_notify);
648

649
void rtnl_set_sk_err(struct net *net, u32 group, int error)
650
{
651 652
	struct sock *rtnl = net->rtnl;

653 654
	netlink_set_err(rtnl, 0, group, error);
}
E
Eric Dumazet 已提交
655
EXPORT_SYMBOL(rtnl_set_sk_err);
656

L
Linus Torvalds 已提交
657 658
int rtnetlink_put_metrics(struct sk_buff *skb, u32 *metrics)
{
659 660 661 662 663 664 665 666 667
	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]) {
668 669 670 671 672 673 674 675
			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;
676 677 678
			} else if (i == RTAX_FEATURES - 1) {
				u32 user_features = metrics[i] & RTAX_FEATURE_MASK;

679 680
				if (!user_features)
					continue;
681 682 683
				BUILD_BUG_ON(RTAX_FEATURE_MASK & DST_FEATURE_MASK);
				if (nla_put_u32(skb, i + 1, user_features))
					goto nla_put_failure;
684 685 686 687
			} else {
				if (nla_put_u32(skb, i + 1, metrics[i]))
					goto nla_put_failure;
			}
688 689
			valid++;
		}
L
Linus Torvalds 已提交
690 691
	}

692 693 694 695
	if (!valid) {
		nla_nest_cancel(skb, mx);
		return 0;
	}
696 697 698 699

	return nla_nest_end(skb, mx);

nla_put_failure:
700 701
	nla_nest_cancel(skb, mx);
	return -EMSGSIZE;
L
Linus Torvalds 已提交
702
}
E
Eric Dumazet 已提交
703
EXPORT_SYMBOL(rtnetlink_put_metrics);
L
Linus Torvalds 已提交
704

705
int rtnl_put_cacheinfo(struct sk_buff *skb, struct dst_entry *dst, u32 id,
706
		       long expires, u32 error)
707 708
{
	struct rta_cacheinfo ci = {
709
		.rta_lastuse = jiffies_delta_to_clock_t(jiffies - dst->lastuse),
710 711 712 713 714 715
		.rta_used = dst->__use,
		.rta_clntref = atomic_read(&(dst->__refcnt)),
		.rta_error = error,
		.rta_id =  id,
	};

716 717
	if (expires) {
		unsigned long clock;
718

719 720 721 722
		clock = jiffies_to_clock_t(abs(expires));
		clock = min_t(unsigned long, clock, INT_MAX);
		ci.rta_expires = (expires > 0) ? clock : -clock;
	}
723 724 725
	return nla_put(skb, RTA_CACHEINFO, sizeof(ci), &ci);
}
EXPORT_SYMBOL_GPL(rtnl_put_cacheinfo);
L
Linus Torvalds 已提交
726

727
static void set_operstate(struct net_device *dev, unsigned char transition)
S
Stefan Rompf 已提交
728 729 730
{
	unsigned char operstate = dev->operstate;

E
Eric Dumazet 已提交
731
	switch (transition) {
S
Stefan Rompf 已提交
732 733 734 735 736 737 738 739 740 741 742 743
	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;
744
	}
S
Stefan Rompf 已提交
745 746 747 748 749

	if (dev->operstate != operstate) {
		write_lock_bh(&dev_base_lock);
		dev->operstate = operstate;
		write_unlock_bh(&dev_base_lock);
750 751
		netdev_state_change(dev);
	}
S
Stefan Rompf 已提交
752 753
}

754 755 756 757 758 759
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));
}

760 761 762 763 764 765 766 767
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) |
768
			(rtnl_dev_get_flags(dev) & ~ifm->ifi_change);
769 770 771 772

	return flags;
}

773
static void copy_rtnl_link_stats(struct rtnl_link_stats *a,
774
				 const struct rtnl_link_stats64 *b)
L
Linus Torvalds 已提交
775
{
776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802
	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;
803 804

	a->rx_nohandler = b->rx_nohandler;
805 806
}

807
/* All VF info */
808 809
static inline int rtnl_vfinfo_size(const struct net_device *dev,
				   u32 ext_filter_mask)
810
{
811
	if (dev->dev.parent && (ext_filter_mask & RTEXT_FILTER_VF)) {
812
		int num_vfs = dev_num_vf(dev->dev.parent);
S
Sabrina Dubroca 已提交
813
		size_t size = nla_total_size(0);
814
		size += num_vfs *
S
Sabrina Dubroca 已提交
815 816 817 818
			(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 */
819 820
			 nla_total_size(MAX_VLAN_LIST_LEN *
					sizeof(struct ifla_vf_vlan_info)) +
821
			 nla_total_size(sizeof(struct ifla_vf_spoofchk)) +
S
Sabrina Dubroca 已提交
822
			 nla_total_size(sizeof(struct ifla_vf_tx_rate)) +
J
Jiri Benc 已提交
823
			 nla_total_size(sizeof(struct ifla_vf_rate)) +
824
			 nla_total_size(sizeof(struct ifla_vf_link_state)) +
825
			 nla_total_size(sizeof(struct ifla_vf_rss_query_en)) +
S
Sabrina Dubroca 已提交
826
			 nla_total_size(0) + /* nest IFLA_VF_STATS */
827
			 /* IFLA_VF_STATS_RX_PACKETS */
N
Nicolas Dichtel 已提交
828
			 nla_total_size_64bit(sizeof(__u64)) +
829
			 /* IFLA_VF_STATS_TX_PACKETS */
N
Nicolas Dichtel 已提交
830
			 nla_total_size_64bit(sizeof(__u64)) +
831
			 /* IFLA_VF_STATS_RX_BYTES */
N
Nicolas Dichtel 已提交
832
			 nla_total_size_64bit(sizeof(__u64)) +
833
			 /* IFLA_VF_STATS_TX_BYTES */
N
Nicolas Dichtel 已提交
834
			 nla_total_size_64bit(sizeof(__u64)) +
835
			 /* IFLA_VF_STATS_BROADCAST */
N
Nicolas Dichtel 已提交
836
			 nla_total_size_64bit(sizeof(__u64)) +
837
			 /* IFLA_VF_STATS_MULTICAST */
N
Nicolas Dichtel 已提交
838
			 nla_total_size_64bit(sizeof(__u64)) +
H
Hiroshi Shimamoto 已提交
839
			 nla_total_size(sizeof(struct ifla_vf_trust)));
840 841
		return size;
	} else
842 843 844
		return 0;
}

845 846
static size_t rtnl_port_size(const struct net_device *dev,
			     u32 ext_filter_mask)
847 848 849 850 851 852 853 854 855 856 857 858 859
{
	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;

860 861
	if (!dev->netdev_ops->ndo_get_vf_port || !dev->dev.parent ||
	    !(ext_filter_mask & RTEXT_FILTER_VF))
862 863 864 865 866 867 868 869
		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
870
static size_t rtnl_xdp_size(void)
871
{
872
	size_t xdp_size = nla_total_size(0) +	/* nest IFLA_XDP */
M
Martin KaFai Lau 已提交
873 874
			  nla_total_size(1) +	/* XDP_ATTACHED */
			  nla_total_size(4);	/* XDP_PROG_ID */
875

D
David S. Miller 已提交
876
	return xdp_size;
877 878
}

879 880
static noinline size_t if_nlmsg_size(const struct net_device *dev,
				     u32 ext_filter_mask)
881 882 883
{
	return NLMSG_ALIGN(sizeof(struct ifinfomsg))
	       + nla_total_size(IFNAMSIZ) /* IFLA_IFNAME */
884
	       + nla_total_size(IFALIASZ) /* IFLA_IFALIAS */
885
	       + nla_total_size(IFNAMSIZ) /* IFLA_QDISC */
886
	       + nla_total_size_64bit(sizeof(struct rtnl_link_ifmap))
887
	       + nla_total_size(sizeof(struct rtnl_link_stats))
888
	       + nla_total_size_64bit(sizeof(struct rtnl_link_stats64))
889 890 891 892 893 894 895
	       + 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 */
896
	       + nla_total_size(1) /* IFLA_CARRIER */
897
	       + nla_total_size(4) /* IFLA_PROMISCUITY */
898 899
	       + nla_total_size(4) /* IFLA_NUM_TX_QUEUES */
	       + nla_total_size(4) /* IFLA_NUM_RX_QUEUES */
900 901
	       + nla_total_size(4) /* IFLA_GSO_MAX_SEGS */
	       + nla_total_size(4) /* IFLA_GSO_MAX_SIZE */
902
	       + nla_total_size(1) /* IFLA_OPERSTATE */
P
Patrick McHardy 已提交
903
	       + nla_total_size(1) /* IFLA_LINKMODE */
904
	       + nla_total_size(4) /* IFLA_CARRIER_CHANGES */
905
	       + nla_total_size(4) /* IFLA_LINK_NETNSID */
906
	       + nla_total_size(4) /* IFLA_GROUP */
907 908 909
	       + nla_total_size(ext_filter_mask
			        & RTEXT_FILTER_VF ? 4 : 0) /* IFLA_NUM_VF */
	       + rtnl_vfinfo_size(dev, ext_filter_mask) /* IFLA_VFINFO_LIST */
910
	       + rtnl_port_size(dev, ext_filter_mask) /* IFLA_VF_PORTS + IFLA_PORT_SELF */
T
Thomas Graf 已提交
911
	       + rtnl_link_get_size(dev) /* IFLA_LINKINFO */
912
	       + rtnl_link_get_af_size(dev, ext_filter_mask) /* IFLA_AF_SPEC */
913
	       + nla_total_size(MAX_PHYS_ITEM_ID_LEN) /* IFLA_PHYS_PORT_ID */
914
	       + nla_total_size(MAX_PHYS_ITEM_ID_LEN) /* IFLA_PHYS_SWITCH_ID */
915
	       + nla_total_size(IFNAMSIZ) /* IFLA_PHYS_PORT_NAME */
D
David S. Miller 已提交
916
	       + rtnl_xdp_size() /* IFLA_XDP */
917
	       + nla_total_size(4)  /* IFLA_EVENT */
918 919
	       + nla_total_size(1); /* IFLA_PROTO_DOWN */

920 921
}

922 923 924 925 926 927 928 929 930 931 932 933 934
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);
935 936
		if (!vf_port)
			goto nla_put_failure;
937 938
		if (nla_put_u32(skb, IFLA_PORT_VF, vf))
			goto nla_put_failure;
939
		err = dev->netdev_ops->ndo_get_vf_port(dev, vf, skb);
940 941
		if (err == -EMSGSIZE)
			goto nla_put_failure;
942 943 944 945 946 947 948 949 950 951
		if (err) {
			nla_nest_cancel(skb, vf_port);
			continue;
		}
		nla_nest_end(skb, vf_port);
	}

	nla_nest_end(skb, vf_ports);

	return 0;
952 953 954 955

nla_put_failure:
	nla_nest_cancel(skb, vf_ports);
	return -EMSGSIZE;
956 957 958 959 960 961 962 963 964 965 966 967 968 969
}

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);
970
		return (err == -EMSGSIZE) ? err : 0;
971 972 973 974 975 976 977
	}

	nla_nest_end(skb, port_self);

	return 0;
}

978 979
static int rtnl_port_fill(struct sk_buff *skb, struct net_device *dev,
			  u32 ext_filter_mask)
980 981 982
{
	int err;

983 984
	if (!dev->netdev_ops->ndo_get_vf_port || !dev->dev.parent ||
	    !(ext_filter_mask & RTEXT_FILTER_VF))
985 986 987 988 989 990 991 992 993 994 995 996 997 998 999
		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;
}

1000 1001 1002
static int rtnl_phys_port_id_fill(struct sk_buff *skb, struct net_device *dev)
{
	int err;
1003
	struct netdev_phys_item_id ppid;
1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017

	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;
}

1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029
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;
	}

1030
	if (nla_put_string(skb, IFLA_PHYS_PORT_NAME, name))
1031 1032 1033 1034 1035
		return -EMSGSIZE;

	return 0;
}

1036 1037 1038
static int rtnl_phys_switch_id_fill(struct sk_buff *skb, struct net_device *dev)
{
	int err;
1039
	struct switchdev_attr attr = {
1040
		.orig_dev = dev,
1041
		.id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID,
1042 1043
		.flags = SWITCHDEV_F_NO_RECURSE,
	};
1044

1045
	err = switchdev_port_attr_get(dev, &attr);
1046 1047 1048 1049 1050 1051
	if (err) {
		if (err == -EOPNOTSUPP)
			return 0;
		return err;
	}

1052 1053
	if (nla_put(skb, IFLA_PHYS_SWITCH_ID, attr.u.ppid.id_len,
		    attr.u.ppid.id))
1054 1055 1056 1057 1058
		return -EMSGSIZE;

	return 0;
}

1059 1060 1061
static noinline_for_stack int rtnl_fill_stats(struct sk_buff *skb,
					      struct net_device *dev)
{
1062
	struct rtnl_link_stats64 *sp;
1063
	struct nlattr *attr;
1064

1065 1066
	attr = nla_reserve_64bit(skb, IFLA_STATS64,
				 sizeof(struct rtnl_link_stats64), IFLA_PAD);
1067 1068 1069
	if (!attr)
		return -EMSGSIZE;

1070 1071
	sp = nla_data(attr);
	dev_get_stats(dev, sp);
1072

1073 1074
	attr = nla_reserve(skb, IFLA_STATS,
			   sizeof(struct rtnl_link_stats));
1075 1076 1077
	if (!attr)
		return -EMSGSIZE;

1078
	copy_rtnl_link_stats(nla_data(attr), sp);
1079 1080 1081 1082 1083 1084 1085 1086 1087 1088

	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;
1089
	struct nlattr *vf, *vfstats, *vfvlanlist;
1090
	struct ifla_vf_link_state vf_linkstate;
1091
	struct ifla_vf_vlan_info vf_vlan_info;
1092 1093 1094 1095 1096 1097 1098 1099 1100
	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;

1101 1102
	memset(&ivi, 0, sizeof(ivi));

1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114
	/* 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;
1115 1116
	/* VLAN Protocol by default is 802.1Q */
	ivi.vlan_proto = htons(ETH_P_8021Q);
1117 1118 1119
	if (dev->netdev_ops->ndo_get_vf_config(dev, vfs_num, &ivi))
		return 0;

1120 1121
	memset(&vf_vlan_info, 0, sizeof(vf_vlan_info));

1122 1123
	vf_mac.vf =
		vf_vlan.vf =
1124
		vf_vlan_info.vf =
1125 1126 1127 1128 1129 1130 1131 1132 1133 1134
		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;
1135 1136 1137
	vf_vlan_info.vlan = ivi.vlan;
	vf_vlan_info.qos = ivi.qos;
	vf_vlan_info.vlan_proto = ivi.vlan_proto;
1138 1139 1140 1141 1142 1143 1144 1145
	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);
1146 1147
	if (!vf)
		goto nla_put_vfinfo_failure;
1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162
	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))
1163 1164 1165 1166 1167 1168 1169 1170 1171 1172
		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);
1173 1174 1175 1176 1177
	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);
1178 1179
	if (!vfstats)
		goto nla_put_vf_failure;
N
Nicolas Dichtel 已提交
1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190
	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,
1191 1192 1193 1194
			      vf_stats.multicast, IFLA_VF_STATS_PAD)) {
		nla_nest_cancel(skb, vfstats);
		goto nla_put_vf_failure;
	}
1195 1196 1197
	nla_nest_end(skb, vfstats);
	nla_nest_end(skb, vf);
	return 0;
1198 1199 1200 1201 1202 1203

nla_put_vf_failure:
	nla_nest_cancel(skb, vf);
nla_put_vfinfo_failure:
	nla_nest_cancel(skb, vfinfo);
	return -EMSGSIZE;
1204 1205 1206 1207
}

static int rtnl_fill_link_ifmap(struct sk_buff *skb, struct net_device *dev)
{
K
Kangjie Lu 已提交
1208 1209 1210 1211 1212 1213 1214 1215 1216 1217
	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;

1218
	if (nla_put_64bit(skb, IFLA_MAP, sizeof(map), &map, IFLA_PAD))
1219 1220 1221 1222 1223
		return -EMSGSIZE;

	return 0;
}

M
Martin KaFai Lau 已提交
1224
static u8 rtnl_xdp_attached_mode(struct net_device *dev, u32 *prog_id)
1225 1226
{
	const struct net_device_ops *ops = dev->netdev_ops;
M
Martin KaFai Lau 已提交
1227
	const struct bpf_prog *generic_xdp_prog;
1228 1229 1230

	ASSERT_RTNL();

M
Martin KaFai Lau 已提交
1231 1232 1233 1234
	*prog_id = 0;
	generic_xdp_prog = rtnl_dereference(dev->xdp_prog);
	if (generic_xdp_prog) {
		*prog_id = generic_xdp_prog->aux->id;
1235
		return XDP_ATTACHED_SKB;
M
Martin KaFai Lau 已提交
1236
	}
1237 1238
	if (!ops->ndo_xdp)
		return XDP_ATTACHED_NONE;
1239

1240
	return __dev_xdp_attached(dev, ops->ndo_xdp, prog_id);
1241 1242
}

1243 1244 1245
static int rtnl_xdp_fill(struct sk_buff *skb, struct net_device *dev)
{
	struct nlattr *xdp;
M
Martin KaFai Lau 已提交
1246
	u32 prog_id;
1247 1248 1249 1250 1251
	int err;

	xdp = nla_nest_start(skb, IFLA_XDP);
	if (!xdp)
		return -EMSGSIZE;
1252 1253

	err = nla_put_u8(skb, IFLA_XDP_ATTACHED,
M
Martin KaFai Lau 已提交
1254
			 rtnl_xdp_attached_mode(dev, &prog_id));
1255 1256 1257
	if (err)
		goto err_cancel;

M
Martin KaFai Lau 已提交
1258 1259 1260 1261 1262 1263
	if (prog_id) {
		err = nla_put_u32(skb, IFLA_XDP_PROG_ID, prog_id);
		if (err)
			goto err_cancel;
	}

1264 1265 1266 1267 1268 1269 1270 1271
	nla_nest_end(skb, xdp);
	return 0;

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

1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301
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;
}

1302
static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
1303
			    int type, u32 pid, u32 seq, u32 change,
1304 1305
			    unsigned int flags, u32 ext_filter_mask,
			    u32 event)
1306 1307 1308
{
	struct ifinfomsg *ifm;
	struct nlmsghdr *nlh;
1309
	struct nlattr *af_spec;
T
Thomas Graf 已提交
1310
	struct rtnl_af_ops *af_ops;
1311
	struct net_device *upper_dev = netdev_master_upper_dev_get(dev);
L
Linus Torvalds 已提交
1312

1313
	ASSERT_RTNL();
1314 1315
	nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ifm), flags);
	if (nlh == NULL)
1316
		return -EMSGSIZE;
L
Linus Torvalds 已提交
1317

1318 1319 1320 1321 1322 1323 1324 1325
	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;

1326 1327 1328 1329 1330 1331 1332
	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) ||
1333
	    nla_put_u32(skb, IFLA_PROMISCUITY, dev->promiscuity) ||
1334
	    nla_put_u32(skb, IFLA_NUM_TX_QUEUES, dev->num_tx_queues) ||
1335 1336
	    nla_put_u32(skb, IFLA_GSO_MAX_SEGS, dev->gso_max_segs) ||
	    nla_put_u32(skb, IFLA_GSO_MAX_SIZE, dev->gso_max_size) ||
1337
#ifdef CONFIG_RPS
1338
	    nla_put_u32(skb, IFLA_NUM_RX_QUEUES, dev->num_rx_queues) ||
1339
#endif
1340 1341
	    (dev->ifindex != dev_get_iflink(dev) &&
	     nla_put_u32(skb, IFLA_LINK, dev_get_iflink(dev))) ||
1342 1343
	    (upper_dev &&
	     nla_put_u32(skb, IFLA_MASTER, upper_dev->ifindex)) ||
1344
	    nla_put_u8(skb, IFLA_CARRIER, netif_carrier_ok(dev)) ||
1345 1346 1347
	    (dev->qdisc &&
	     nla_put_string(skb, IFLA_QDISC, dev->qdisc->ops->id)) ||
	    (dev->ifalias &&
1348 1349
	     nla_put_string(skb, IFLA_IFALIAS, dev->ifalias)) ||
	    nla_put_u32(skb, IFLA_CARRIER_CHANGES,
1350 1351
			atomic_read(&dev->carrier_changes)) ||
	    nla_put_u8(skb, IFLA_PROTO_DOWN, dev->proto_down))
1352
		goto nla_put_failure;
1353

1354 1355 1356 1357 1358
	if (event != IFLA_EVENT_NONE) {
		if (nla_put_u32(skb, IFLA_EVENT, event))
			goto nla_put_failure;
	}

1359 1360
	if (rtnl_fill_link_ifmap(skb, dev))
		goto nla_put_failure;
L
Linus Torvalds 已提交
1361 1362

	if (dev->addr_len) {
1363 1364 1365
		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
1366 1367
	}

1368 1369 1370
	if (rtnl_phys_port_id_fill(skb, dev))
		goto nla_put_failure;

1371 1372 1373
	if (rtnl_phys_port_name_fill(skb, dev))
		goto nla_put_failure;

1374 1375 1376
	if (rtnl_phys_switch_id_fill(skb, dev))
		goto nla_put_failure;

1377
	if (rtnl_fill_stats(skb, dev))
1378 1379
		goto nla_put_failure;

1380 1381 1382
	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;
1383

1384 1385
	if (dev->netdev_ops->ndo_get_vf_config && dev->dev.parent &&
	    ext_filter_mask & RTEXT_FILTER_VF) {
1386
		int i;
1387
		struct nlattr *vfinfo;
1388 1389 1390 1391 1392 1393
		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++) {
1394
			if (rtnl_fill_vfinfo(skb, dev, i, vfinfo))
1395
				goto nla_put_failure;
1396
		}
1397

1398
		nla_nest_end(skb, vfinfo);
1399
	}
1400

1401
	if (rtnl_port_fill(skb, dev, ext_filter_mask))
1402 1403
		goto nla_put_failure;

1404 1405 1406
	if (rtnl_xdp_fill(skb, dev))
		goto nla_put_failure;

1407
	if (dev->rtnl_link_ops || rtnl_have_link_slave_info(dev)) {
P
Patrick McHardy 已提交
1408 1409 1410 1411
		if (rtnl_link_fill(skb, dev) < 0)
			goto nla_put_failure;
	}

1412 1413 1414 1415 1416
	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)) {
1417
			int id = peernet2id_alloc(dev_net(dev), link_net);
1418 1419 1420 1421 1422 1423

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

T
Thomas Graf 已提交
1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434
	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;

1435
			err = af_ops->fill_link_af(skb, dev, ext_filter_mask);
T
Thomas Graf 已提交
1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453

			/*
			 * 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);

1454 1455
	nlmsg_end(skb, nlh);
	return 0;
1456 1457

nla_put_failure:
1458 1459
	nlmsg_cancel(skb, nlh);
	return -EMSGSIZE;
L
Linus Torvalds 已提交
1460 1461
}

J
Jiri Pirko 已提交
1462
static const struct nla_policy ifla_policy[IFLA_MAX+1] = {
1463
	[IFLA_IFNAME]		= { .type = NLA_STRING, .len = IFNAMSIZ-1 },
P
Patrick McHardy 已提交
1464 1465
	[IFLA_ADDRESS]		= { .type = NLA_BINARY, .len = MAX_ADDR_LEN },
	[IFLA_BROADCAST]	= { .type = NLA_BINARY, .len = MAX_ADDR_LEN },
1466
	[IFLA_MAP]		= { .len = sizeof(struct rtnl_link_ifmap) },
1467
	[IFLA_MTU]		= { .type = NLA_U32 },
1468
	[IFLA_LINK]		= { .type = NLA_U32 },
1469
	[IFLA_MASTER]		= { .type = NLA_U32 },
1470
	[IFLA_CARRIER]		= { .type = NLA_U8 },
1471 1472 1473 1474
	[IFLA_TXQLEN]		= { .type = NLA_U32 },
	[IFLA_WEIGHT]		= { .type = NLA_U32 },
	[IFLA_OPERSTATE]	= { .type = NLA_U8 },
	[IFLA_LINKMODE]		= { .type = NLA_U8 },
1475
	[IFLA_LINKINFO]		= { .type = NLA_NESTED },
1476
	[IFLA_NET_NS_PID]	= { .type = NLA_U32 },
1477
	[IFLA_NET_NS_FD]	= { .type = NLA_U32 },
1478
	[IFLA_IFALIAS]	        = { .type = NLA_STRING, .len = IFALIASZ-1 },
1479
	[IFLA_VFINFO_LIST]	= {. type = NLA_NESTED },
1480 1481
	[IFLA_VF_PORTS]		= { .type = NLA_NESTED },
	[IFLA_PORT_SELF]	= { .type = NLA_NESTED },
T
Thomas Graf 已提交
1482
	[IFLA_AF_SPEC]		= { .type = NLA_NESTED },
1483
	[IFLA_EXT_MASK]		= { .type = NLA_U32 },
1484
	[IFLA_PROMISCUITY]	= { .type = NLA_U32 },
1485 1486
	[IFLA_NUM_TX_QUEUES]	= { .type = NLA_U32 },
	[IFLA_NUM_RX_QUEUES]	= { .type = NLA_U32 },
1487
	[IFLA_PHYS_PORT_ID]	= { .type = NLA_BINARY, .len = MAX_PHYS_ITEM_ID_LEN },
1488
	[IFLA_CARRIER_CHANGES]	= { .type = NLA_U32 },  /* ignored */
1489
	[IFLA_PHYS_SWITCH_ID]	= { .type = NLA_BINARY, .len = MAX_PHYS_ITEM_ID_LEN },
1490
	[IFLA_LINK_NETNSID]	= { .type = NLA_S32 },
1491
	[IFLA_PROTO_DOWN]	= { .type = NLA_U8 },
1492
	[IFLA_XDP]		= { .type = NLA_NESTED },
1493
	[IFLA_EVENT]		= { .type = NLA_U32 },
1494
	[IFLA_GROUP]		= { .type = NLA_U32 },
1495 1496
};

P
Patrick McHardy 已提交
1497 1498 1499
static const struct nla_policy ifla_info_policy[IFLA_INFO_MAX+1] = {
	[IFLA_INFO_KIND]	= { .type = NLA_STRING },
	[IFLA_INFO_DATA]	= { .type = NLA_NESTED },
1500 1501
	[IFLA_INFO_SLAVE_KIND]	= { .type = NLA_STRING },
	[IFLA_INFO_SLAVE_DATA]	= { .type = NLA_NESTED },
P
Patrick McHardy 已提交
1502 1503
};

1504
static const struct nla_policy ifla_vf_policy[IFLA_VF_MAX+1] = {
1505 1506
	[IFLA_VF_MAC]		= { .len = sizeof(struct ifla_vf_mac) },
	[IFLA_VF_VLAN]		= { .len = sizeof(struct ifla_vf_vlan) },
1507
	[IFLA_VF_VLAN_LIST]     = { .type = NLA_NESTED },
1508 1509 1510 1511
	[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) },
1512
	[IFLA_VF_RSS_QUERY_EN]	= { .len = sizeof(struct ifla_vf_rss_query_en) },
1513
	[IFLA_VF_STATS]		= { .type = NLA_NESTED },
H
Hiroshi Shimamoto 已提交
1514
	[IFLA_VF_TRUST]		= { .len = sizeof(struct ifla_vf_trust) },
1515 1516
	[IFLA_VF_IB_NODE_GUID]	= { .len = sizeof(struct ifla_vf_guid) },
	[IFLA_VF_IB_PORT_GUID]	= { .len = sizeof(struct ifla_vf_guid) },
1517 1518
};

1519 1520 1521 1522 1523 1524 1525 1526 1527 1528
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, },
1529 1530 1531 1532 1533 1534 1535

	/* 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) },
1536 1537
};

1538 1539 1540
static const struct nla_policy ifla_xdp_policy[IFLA_XDP_MAX + 1] = {
	[IFLA_XDP_FD]		= { .type = NLA_S32 },
	[IFLA_XDP_ATTACHED]	= { .type = NLA_U8 },
1541
	[IFLA_XDP_FLAGS]	= { .type = NLA_U32 },
M
Martin KaFai Lau 已提交
1542
	[IFLA_XDP_PROG_ID]	= { .type = NLA_U32 },
1543 1544
};

1545 1546 1547 1548 1549
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];

1550 1551
	if (nla_parse_nested(linfo, IFLA_INFO_MAX, nla,
			     ifla_info_policy, NULL) < 0)
1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597
		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
1598 1599 1600 1601 1602 1603 1604 1605 1606
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;
1607 1608 1609
	const struct rtnl_link_ops *kind_ops = NULL;
	unsigned int flags = NLM_F_MULTI;
	int master_idx = 0;
1610
	int err;
1611
	int hdrlen;
J
Jiri Pirko 已提交
1612 1613 1614 1615 1616 1617

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

	cb->seq = net->dev_base_seq;

1618 1619 1620 1621 1622 1623 1624 1625 1626 1627
	/* 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);

1628 1629
	if (nlmsg_parse(cb->nlh, hdrlen, tb, IFLA_MAX,
			ifla_policy, NULL) >= 0) {
J
Jiri Pirko 已提交
1630 1631
		if (tb[IFLA_EXT_MASK])
			ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);
1632 1633 1634 1635 1636 1637 1638 1639 1640

		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
1641 1642 1643 1644 1645
	}

	for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) {
		idx = 0;
		head = &net->dev_index_head[h];
1646
		hlist_for_each_entry(dev, head, index_hlist) {
1647
			if (link_dump_filtered(dev, master_idx, kind_ops))
1648
				goto cont;
J
Jiri Pirko 已提交
1649 1650
			if (idx < s_idx)
				goto cont;
1651 1652 1653
			err = rtnl_fill_ifinfo(skb, dev, RTM_NEWLINK,
					       NETLINK_CB(cb->skb).portid,
					       cb->nlh->nlmsg_seq, 0,
1654
					       flags,
1655
					       ext_filter_mask, 0);
1656

1657 1658 1659 1660 1661 1662
			if (err < 0) {
				if (likely(skb->len))
					goto out;

				goto out_err;
			}
J
Jiri Pirko 已提交
1663 1664 1665 1666 1667 1668 1669

			nl_dump_check_consistent(cb, nlmsg_hdr(skb));
cont:
			idx++;
		}
	}
out:
1670 1671
	err = skb->len;
out_err:
J
Jiri Pirko 已提交
1672 1673 1674
	cb->args[1] = idx;
	cb->args[0] = h;

1675
	return err;
J
Jiri Pirko 已提交
1676 1677
}

1678 1679
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
1680
{
1681
	return nla_parse(tb, IFLA_MAX, head, len, ifla_policy, exterr);
J
Jiri Pirko 已提交
1682 1683 1684
}
EXPORT_SYMBOL(rtnl_nla_parse_ifla);

1685 1686 1687 1688 1689 1690 1691 1692
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]));
1693 1694
	else if (tb[IFLA_NET_NS_FD])
		net = get_net_ns_by_fd(nla_get_u32(tb[IFLA_NET_NS_FD]));
1695 1696 1697 1698 1699 1700
	else
		net = get_net(src_net);
	return net;
}
EXPORT_SYMBOL(rtnl_link_get_net);

1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712
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;
	}

1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726
	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) {
1727
				err = af_ops->validate_link_af(dev, af);
1728 1729 1730 1731 1732 1733
				if (err < 0)
					return err;
			}
		}
	}

1734 1735 1736
	return 0;
}

1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752
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);
}

1753
static int do_setvfinfo(struct net_device *dev, struct nlattr **tb)
1754 1755
{
	const struct net_device_ops *ops = dev->netdev_ops;
1756
	int err = -EINVAL;
1757

1758 1759
	if (tb[IFLA_VF_MAC]) {
		struct ifla_vf_mac *ivm = nla_data(tb[IFLA_VF_MAC]);
1760

1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774
		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,
1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800
						   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++;
		}
1801 1802 1803
		if (len == 0)
			return -EINVAL;

1804 1805
		err = ops->ndo_set_vf_vlan(dev, ivvl[0]->vf, ivvl[0]->vlan,
					   ivvl[0]->qos, ivvl[0]->vlan_proto);
1806 1807
		if (err < 0)
			return err;
1808
	}
1809 1810 1811 1812 1813 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

	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
1875 1876 1877 1878 1879 1880 1881 1882 1883 1884
	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;
	}

1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902
	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);
	}

1903 1904 1905
	return err;
}

1906 1907
static int do_set_master(struct net_device *dev, int ifindex)
{
1908
	struct net_device *upper_dev = netdev_master_upper_dev_get(dev);
1909 1910 1911
	const struct net_device_ops *ops;
	int err;

1912 1913
	if (upper_dev) {
		if (upper_dev->ifindex == ifindex)
1914
			return 0;
1915
		ops = upper_dev->netdev_ops;
1916
		if (ops->ndo_del_slave) {
1917
			err = ops->ndo_del_slave(upper_dev, dev);
1918 1919 1920 1921 1922 1923 1924 1925
			if (err)
				return err;
		} else {
			return -EOPNOTSUPP;
		}
	}

	if (ifindex) {
1926 1927
		upper_dev = __dev_get_by_index(dev_net(dev), ifindex);
		if (!upper_dev)
1928
			return -EINVAL;
1929
		ops = upper_dev->netdev_ops;
1930
		if (ops->ndo_add_slave) {
1931
			err = ops->ndo_add_slave(upper_dev, dev);
1932 1933 1934 1935 1936 1937 1938 1939 1940
			if (err)
				return err;
		} else {
			return -EOPNOTSUPP;
		}
	}
	return 0;
}

1941
#define DO_SETLINK_MODIFIED	0x01
1942 1943
/* notify flag means notify + modified. */
#define DO_SETLINK_NOTIFY	0x03
1944 1945
static int do_setlink(const struct sk_buff *skb,
		      struct net_device *dev, struct ifinfomsg *ifm,
1946
		      struct netlink_ext_ack *extack,
1947
		      struct nlattr **tb, char *ifname, int status)
L
Linus Torvalds 已提交
1948
{
1949
	const struct net_device_ops *ops = dev->netdev_ops;
1950
	int err;
L
Linus Torvalds 已提交
1951

1952
	if (tb[IFLA_NET_NS_PID] || tb[IFLA_NET_NS_FD]) {
1953
		struct net *net = rtnl_link_get_net(dev_net(dev), tb);
1954 1955 1956 1957
		if (IS_ERR(net)) {
			err = PTR_ERR(net);
			goto errout;
		}
1958
		if (!netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) {
1959
			put_net(net);
1960 1961 1962
			err = -EPERM;
			goto errout;
		}
1963 1964 1965 1966
		err = dev_change_net_namespace(dev, net, ifname);
		put_net(net);
		if (err)
			goto errout;
1967
		status |= DO_SETLINK_MODIFIED;
1968 1969
	}

1970
	if (tb[IFLA_MAP]) {
L
Linus Torvalds 已提交
1971 1972 1973
		struct rtnl_link_ifmap *u_map;
		struct ifmap k_map;

1974
		if (!ops->ndo_set_config) {
L
Linus Torvalds 已提交
1975
			err = -EOPNOTSUPP;
1976
			goto errout;
L
Linus Torvalds 已提交
1977 1978 1979 1980
		}

		if (!netif_device_present(dev)) {
			err = -ENODEV;
1981
			goto errout;
L
Linus Torvalds 已提交
1982 1983
		}

1984
		u_map = nla_data(tb[IFLA_MAP]);
L
Linus Torvalds 已提交
1985 1986 1987 1988 1989 1990 1991
		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;

1992
		err = ops->ndo_set_config(dev, &k_map);
1993
		if (err < 0)
1994
			goto errout;
L
Linus Torvalds 已提交
1995

1996
		status |= DO_SETLINK_NOTIFY;
L
Linus Torvalds 已提交
1997 1998
	}

1999
	if (tb[IFLA_ADDRESS]) {
2000 2001 2002
		struct sockaddr *sa;
		int len;

2003 2004
		len = sizeof(sa_family_t) + max_t(size_t, dev->addr_len,
						  sizeof(*sa));
2005 2006 2007
		sa = kmalloc(len, GFP_KERNEL);
		if (!sa) {
			err = -ENOMEM;
2008
			goto errout;
2009 2010
		}
		sa->sa_family = dev->type;
2011
		memcpy(sa->sa_data, nla_data(tb[IFLA_ADDRESS]),
2012
		       dev->addr_len);
2013
		err = dev_set_mac_address(dev, sa);
2014
		kfree(sa);
L
Linus Torvalds 已提交
2015
		if (err)
2016
			goto errout;
2017
		status |= DO_SETLINK_MODIFIED;
L
Linus Torvalds 已提交
2018 2019
	}

2020 2021 2022
	if (tb[IFLA_MTU]) {
		err = dev_set_mtu(dev, nla_get_u32(tb[IFLA_MTU]));
		if (err < 0)
2023
			goto errout;
2024
		status |= DO_SETLINK_MODIFIED;
L
Linus Torvalds 已提交
2025 2026
	}

2027 2028
	if (tb[IFLA_GROUP]) {
		dev_set_group(dev, nla_get_u32(tb[IFLA_GROUP]));
2029
		status |= DO_SETLINK_NOTIFY;
2030 2031
	}

2032 2033 2034 2035 2036
	/*
	 * Interface selected by interface index but interface
	 * name provided implies that a name change has been
	 * requested.
	 */
2037
	if (ifm->ifi_index > 0 && ifname[0]) {
2038 2039
		err = dev_change_name(dev, ifname);
		if (err < 0)
2040
			goto errout;
2041
		status |= DO_SETLINK_MODIFIED;
L
Linus Torvalds 已提交
2042 2043
	}

2044 2045 2046 2047 2048
	if (tb[IFLA_IFALIAS]) {
		err = dev_set_alias(dev, nla_data(tb[IFLA_IFALIAS]),
				    nla_len(tb[IFLA_IFALIAS]));
		if (err < 0)
			goto errout;
2049
		status |= DO_SETLINK_NOTIFY;
2050 2051
	}

2052 2053
	if (tb[IFLA_BROADCAST]) {
		nla_memcpy(dev->broadcast, tb[IFLA_BROADCAST], dev->addr_len);
2054
		call_netdevice_notifiers(NETDEV_CHANGEADDR, dev);
L
Linus Torvalds 已提交
2055 2056
	}

2057
	if (ifm->ifi_flags || ifm->ifi_change) {
2058
		err = dev_change_flags(dev, rtnl_dev_combine_flags(dev, ifm));
2059 2060
		if (err < 0)
			goto errout;
2061
	}
L
Linus Torvalds 已提交
2062

2063 2064 2065 2066
	if (tb[IFLA_MASTER]) {
		err = do_set_master(dev, nla_get_u32(tb[IFLA_MASTER]));
		if (err)
			goto errout;
2067
		status |= DO_SETLINK_MODIFIED;
2068 2069
	}

2070 2071 2072 2073
	if (tb[IFLA_CARRIER]) {
		err = dev_change_carrier(dev, nla_get_u8(tb[IFLA_CARRIER]));
		if (err)
			goto errout;
2074
		status |= DO_SETLINK_MODIFIED;
2075 2076
	}

2077
	if (tb[IFLA_TXQLEN]) {
2078 2079
		unsigned int value = nla_get_u32(tb[IFLA_TXQLEN]);
		unsigned int orig_len = dev->tx_queue_len;
2080 2081 2082 2083 2084 2085 2086 2087 2088 2089

		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;
			}
2090
			status |= DO_SETLINK_NOTIFY;
2091
		}
2092
	}
S
Stefan Rompf 已提交
2093

2094
	if (tb[IFLA_OPERSTATE])
2095
		set_operstate(dev, nla_get_u8(tb[IFLA_OPERSTATE]));
S
Stefan Rompf 已提交
2096

2097
	if (tb[IFLA_LINKMODE]) {
2098 2099
		unsigned char value = nla_get_u8(tb[IFLA_LINKMODE]);

2100
		write_lock_bh(&dev_base_lock);
2101
		if (dev->link_mode ^ value)
2102
			status |= DO_SETLINK_NOTIFY;
2103
		dev->link_mode = value;
2104
		write_unlock_bh(&dev_base_lock);
S
Stefan Rompf 已提交
2105 2106
	}

2107
	if (tb[IFLA_VFINFO_LIST]) {
2108
		struct nlattr *vfinfo[IFLA_VF_MAX + 1];
2109 2110
		struct nlattr *attr;
		int rem;
2111

2112
		nla_for_each_nested(attr, tb[IFLA_VFINFO_LIST], rem) {
2113 2114
			if (nla_type(attr) != IFLA_VF_INFO ||
			    nla_len(attr) < NLA_HDRLEN) {
2115
				err = -EINVAL;
2116
				goto errout;
2117
			}
2118
			err = nla_parse_nested(vfinfo, IFLA_VF_MAX, attr,
2119
					       ifla_vf_policy, NULL);
2120 2121 2122
			if (err < 0)
				goto errout;
			err = do_setvfinfo(dev, vfinfo);
2123 2124
			if (err < 0)
				goto errout;
2125
			status |= DO_SETLINK_NOTIFY;
2126
		}
2127
	}
L
Linus Torvalds 已提交
2128 2129
	err = 0;

2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140
	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) {
2141 2142 2143 2144 2145 2146
			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,
2147
					       ifla_port_policy, NULL);
2148 2149 2150 2151 2152 2153 2154 2155 2156 2157
			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;
2158
			status |= DO_SETLINK_NOTIFY;
2159 2160 2161 2162 2163 2164 2165 2166
		}
	}
	err = 0;

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

		err = nla_parse_nested(port, IFLA_PORT_MAX,
2167 2168
				       tb[IFLA_PORT_SELF], ifla_port_policy,
				       NULL);
2169 2170 2171 2172 2173 2174 2175 2176
		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;
2177
		status |= DO_SETLINK_NOTIFY;
2178
	}
T
Thomas Graf 已提交
2179 2180 2181 2182 2183 2184 2185 2186 2187

	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))))
2188
				BUG();
T
Thomas Graf 已提交
2189

2190
			err = af_ops->set_link_af(dev, af);
T
Thomas Graf 已提交
2191 2192 2193
			if (err < 0)
				goto errout;

2194
			status |= DO_SETLINK_NOTIFY;
T
Thomas Graf 已提交
2195 2196
		}
	}
2197 2198
	err = 0;

2199 2200 2201 2202 2203 2204 2205 2206
	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;
	}

2207 2208
	if (tb[IFLA_XDP]) {
		struct nlattr *xdp[IFLA_XDP_MAX + 1];
2209
		u32 xdp_flags = 0;
2210 2211

		err = nla_parse_nested(xdp, IFLA_XDP_MAX, tb[IFLA_XDP],
2212
				       ifla_xdp_policy, NULL);
2213 2214 2215
		if (err < 0)
			goto errout;

M
Martin KaFai Lau 已提交
2216
		if (xdp[IFLA_XDP_ATTACHED] || xdp[IFLA_XDP_PROG_ID]) {
2217 2218 2219
			err = -EINVAL;
			goto errout;
		}
2220 2221 2222 2223 2224 2225 2226

		if (xdp[IFLA_XDP_FLAGS]) {
			xdp_flags = nla_get_u32(xdp[IFLA_XDP_FLAGS]);
			if (xdp_flags & ~XDP_FLAGS_MASK) {
				err = -EINVAL;
				goto errout;
			}
2227
			if (hweight32(xdp_flags & XDP_FLAGS_MODES) > 1) {
2228 2229 2230
				err = -EINVAL;
				goto errout;
			}
2231 2232
		}

2233
		if (xdp[IFLA_XDP_FD]) {
2234
			err = dev_change_xdp_fd(dev, extack,
2235 2236
						nla_get_s32(xdp[IFLA_XDP_FD]),
						xdp_flags);
2237 2238 2239 2240 2241 2242
			if (err)
				goto errout;
			status |= DO_SETLINK_NOTIFY;
		}
	}

2243
errout:
2244 2245 2246 2247 2248 2249 2250 2251
	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);
	}
2252

2253 2254
	return err;
}
L
Linus Torvalds 已提交
2255

2256 2257
static int rtnl_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
2258
{
2259
	struct net *net = sock_net(skb->sk);
2260 2261 2262 2263 2264 2265
	struct ifinfomsg *ifm;
	struct net_device *dev;
	int err;
	struct nlattr *tb[IFLA_MAX+1];
	char ifname[IFNAMSIZ];

2266 2267
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy,
			  extack);
2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278
	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)
2279
		dev = __dev_get_by_index(net, ifm->ifi_index);
2280
	else if (tb[IFLA_IFNAME])
2281
		dev = __dev_get_by_name(net, ifname);
2282 2283 2284 2285 2286 2287 2288 2289
	else
		goto errout;

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

E
Eric Dumazet 已提交
2290 2291
	err = validate_linkmsg(dev, tb);
	if (err < 0)
2292
		goto errout;
2293

2294
	err = do_setlink(skb, dev, ifm, extack, tb, ifname, 0);
2295
errout:
L
Linus Torvalds 已提交
2296 2297 2298
	return err;
}

2299 2300 2301 2302 2303 2304 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
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;
}

2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350
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);

2351 2352
static int rtnl_dellink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
P
Patrick McHardy 已提交
2353
{
2354
	struct net *net = sock_net(skb->sk);
P
Patrick McHardy 已提交
2355 2356 2357 2358 2359 2360
	struct net_device *dev;
	struct ifinfomsg *ifm;
	char ifname[IFNAMSIZ];
	struct nlattr *tb[IFLA_MAX+1];
	int err;

2361
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy, extack);
P
Patrick McHardy 已提交
2362 2363 2364 2365 2366 2367 2368 2369
	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)
2370
		dev = __dev_get_by_index(net, ifm->ifi_index);
P
Patrick McHardy 已提交
2371
	else if (tb[IFLA_IFNAME])
2372
		dev = __dev_get_by_name(net, ifname);
2373 2374
	else if (tb[IFLA_GROUP])
		return rtnl_group_dellink(net, nla_get_u32(tb[IFLA_GROUP]));
P
Patrick McHardy 已提交
2375 2376 2377 2378 2379 2380
	else
		return -EINVAL;

	if (!dev)
		return -ENODEV;

2381
	return rtnl_delete_link(dev);
P
Patrick McHardy 已提交
2382 2383
}

2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397
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;

2398
	__dev_notify_flags(dev, old_flags, ~0U);
2399 2400 2401 2402
	return 0;
}
EXPORT_SYMBOL(rtnl_configure_link);

2403
struct net_device *rtnl_create_link(struct net *net,
2404
	const char *ifname, unsigned char name_assign_type,
2405
	const struct rtnl_link_ops *ops, struct nlattr *tb[])
2406 2407
{
	struct net_device *dev;
2408 2409
	unsigned int num_tx_queues = 1;
	unsigned int num_rx_queues = 1;
2410

2411 2412 2413
	if (tb[IFLA_NUM_TX_QUEUES])
		num_tx_queues = nla_get_u32(tb[IFLA_NUM_TX_QUEUES]);
	else if (ops->get_num_tx_queues)
2414
		num_tx_queues = ops->get_num_tx_queues();
2415 2416 2417 2418

	if (tb[IFLA_NUM_RX_QUEUES])
		num_rx_queues = nla_get_u32(tb[IFLA_NUM_RX_QUEUES]);
	else if (ops->get_num_rx_queues)
2419
		num_rx_queues = ops->get_num_rx_queues();
2420

2421
	dev = alloc_netdev_mqs(ops->priv_size, ifname, name_assign_type,
2422
			       ops->setup, num_tx_queues, num_rx_queues);
2423
	if (!dev)
2424
		return ERR_PTR(-ENOMEM);
2425

2426 2427
	dev_net_set(dev, net);
	dev->rtnl_link_ops = ops;
2428
	dev->rtnl_link_state = RTNL_LINK_INITIALIZING;
2429

2430 2431
	if (tb[IFLA_MTU])
		dev->mtu = nla_get_u32(tb[IFLA_MTU]);
2432
	if (tb[IFLA_ADDRESS]) {
2433 2434
		memcpy(dev->dev_addr, nla_data(tb[IFLA_ADDRESS]),
				nla_len(tb[IFLA_ADDRESS]));
2435 2436
		dev->addr_assign_type = NET_ADDR_SET;
	}
2437 2438 2439 2440 2441 2442
	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])
2443
		set_operstate(dev, nla_get_u8(tb[IFLA_OPERSTATE]));
2444 2445
	if (tb[IFLA_LINKMODE])
		dev->link_mode = nla_get_u8(tb[IFLA_LINKMODE]);
2446 2447
	if (tb[IFLA_GROUP])
		dev_set_group(dev, nla_get_u32(tb[IFLA_GROUP]));
2448 2449 2450

	return dev;
}
E
Eric Dumazet 已提交
2451
EXPORT_SYMBOL(rtnl_create_link);
2452

2453 2454
static int rtnl_group_changelink(const struct sk_buff *skb,
		struct net *net, int group,
2455
		struct ifinfomsg *ifm,
2456
		struct netlink_ext_ack *extack,
2457 2458
		struct nlattr **tb)
{
2459
	struct net_device *dev, *aux;
2460 2461
	int err;

2462
	for_each_netdev_safe(net, dev, aux) {
2463
		if (dev->group == group) {
2464
			err = do_setlink(skb, dev, ifm, extack, tb, NULL, 0);
2465 2466 2467 2468 2469 2470 2471 2472
			if (err < 0)
				return err;
		}
	}

	return 0;
}

2473 2474
static int rtnl_newlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
P
Patrick McHardy 已提交
2475
{
2476
	struct net *net = sock_net(skb->sk);
P
Patrick McHardy 已提交
2477
	const struct rtnl_link_ops *ops;
2478
	const struct rtnl_link_ops *m_ops = NULL;
P
Patrick McHardy 已提交
2479
	struct net_device *dev;
2480
	struct net_device *master_dev = NULL;
P
Patrick McHardy 已提交
2481 2482 2483 2484 2485
	struct ifinfomsg *ifm;
	char kind[MODULE_NAME_LEN];
	char ifname[IFNAMSIZ];
	struct nlattr *tb[IFLA_MAX+1];
	struct nlattr *linkinfo[IFLA_INFO_MAX+1];
2486
	unsigned char name_assign_type = NET_NAME_USER;
P
Patrick McHardy 已提交
2487 2488
	int err;

2489
#ifdef CONFIG_MODULES
P
Patrick McHardy 已提交
2490
replay:
2491
#endif
2492
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy, extack);
P
Patrick McHardy 已提交
2493 2494 2495 2496 2497 2498 2499 2500 2501 2502
	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)
2503
		dev = __dev_get_by_index(net, ifm->ifi_index);
2504 2505 2506 2507 2508 2509
	else {
		if (ifname[0])
			dev = __dev_get_by_name(net, ifname);
		else
			dev = NULL;
	}
P
Patrick McHardy 已提交
2510

2511 2512 2513 2514 2515 2516
	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
2517 2518
	err = validate_linkmsg(dev, tb);
	if (err < 0)
2519 2520
		return err;

P
Patrick McHardy 已提交
2521 2522
	if (tb[IFLA_LINKINFO]) {
		err = nla_parse_nested(linkinfo, IFLA_INFO_MAX,
2523 2524
				       tb[IFLA_LINKINFO], ifla_info_policy,
				       NULL);
P
Patrick McHardy 已提交
2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538
		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
2539 2540
		struct nlattr *attr[ops ? ops->maxtype + 1 : 1];
		struct nlattr *slave_attr[m_ops ? m_ops->slave_maxtype + 1 : 1];
2541 2542
		struct nlattr **data = NULL;
		struct nlattr **slave_data = NULL;
2543
		struct net *dest_net, *link_net = NULL;
P
Patrick McHardy 已提交
2544 2545 2546 2547 2548

		if (ops) {
			if (ops->maxtype && linkinfo[IFLA_INFO_DATA]) {
				err = nla_parse_nested(attr, ops->maxtype,
						       linkinfo[IFLA_INFO_DATA],
2549
						       ops->policy, NULL);
P
Patrick McHardy 已提交
2550 2551 2552 2553 2554
				if (err < 0)
					return err;
				data = attr;
			}
			if (ops->validate) {
2555
				err = ops->validate(tb, data, extack);
P
Patrick McHardy 已提交
2556 2557 2558 2559 2560
				if (err < 0)
					return err;
			}
		}

2561 2562 2563 2564 2565 2566
		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],
2567 2568
						       m_ops->slave_policy,
						       NULL);
2569 2570 2571 2572 2573
				if (err < 0)
					return err;
				slave_data = slave_attr;
			}
			if (m_ops->slave_validate) {
2574 2575
				err = m_ops->slave_validate(tb, slave_data,
							    extack);
2576 2577 2578 2579 2580
				if (err < 0)
					return err;
			}
		}

P
Patrick McHardy 已提交
2581
		if (dev) {
2582
			int status = 0;
P
Patrick McHardy 已提交
2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593

			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;

2594
				err = ops->changelink(dev, tb, data, extack);
P
Patrick McHardy 已提交
2595 2596
				if (err < 0)
					return err;
2597
				status |= DO_SETLINK_NOTIFY;
P
Patrick McHardy 已提交
2598 2599
			}

2600 2601 2602 2603 2604
			if (linkinfo[IFLA_INFO_SLAVE_DATA]) {
				if (!m_ops || !m_ops->slave_changelink)
					return -EOPNOTSUPP;

				err = m_ops->slave_changelink(master_dev, dev,
2605 2606
							      tb, slave_data,
							      extack);
2607 2608
				if (err < 0)
					return err;
2609
				status |= DO_SETLINK_NOTIFY;
2610 2611
			}

2612 2613
			return do_setlink(skb, dev, ifm, extack, tb, ifname,
					  status);
P
Patrick McHardy 已提交
2614 2615
		}

2616 2617
		if (!(nlh->nlmsg_flags & NLM_F_CREATE)) {
			if (ifm->ifi_index == 0 && tb[IFLA_GROUP])
2618
				return rtnl_group_changelink(skb, net,
2619
						nla_get_u32(tb[IFLA_GROUP]),
2620
						ifm, extack, tb);
P
Patrick McHardy 已提交
2621
			return -ENODEV;
2622
		}
P
Patrick McHardy 已提交
2623

2624
		if (tb[IFLA_MAP] || tb[IFLA_PROTINFO])
P
Patrick McHardy 已提交
2625 2626 2627
			return -EOPNOTSUPP;

		if (!ops) {
2628
#ifdef CONFIG_MODULES
P
Patrick McHardy 已提交
2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640
			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;
		}

2641 2642 2643
		if (!ops->setup)
			return -EOPNOTSUPP;

2644
		if (!ifname[0]) {
P
Patrick McHardy 已提交
2645
			snprintf(ifname, IFNAMSIZ, "%s%%d", ops->kind);
2646 2647
			name_assign_type = NET_NAME_ENUM;
		}
2648

2649
		dest_net = rtnl_link_get_net(net, tb);
2650 2651 2652
		if (IS_ERR(dest_net))
			return PTR_ERR(dest_net);

2653 2654 2655 2656
		err = -EPERM;
		if (!netlink_ns_capable(skb, dest_net->user_ns, CAP_NET_ADMIN))
			goto out;

2657 2658 2659 2660 2661 2662 2663 2664
		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;
			}
2665 2666 2667
			err = -EPERM;
			if (!netlink_ns_capable(skb, link_net->user_ns, CAP_NET_ADMIN))
				goto out;
2668 2669 2670 2671
		}

		dev = rtnl_create_link(link_net ? : dest_net, ifname,
				       name_assign_type, ops, tb);
2672
		if (IS_ERR(dev)) {
2673
			err = PTR_ERR(dev);
2674 2675 2676 2677 2678
			goto out;
		}

		dev->ifindex = ifm->ifi_index;

2679
		if (ops->newlink) {
2680 2681
			err = ops->newlink(link_net ? : net, dev, tb, data,
					   extack);
2682
			/* Drivers should call free_netdev() in ->destructor
2683 2684
			 * and unregister it on failure after registration
			 * so that device could be finally freed in rtnl_unlock.
2685
			 */
2686 2687 2688 2689
			if (err < 0) {
				/* If device is not registered at all, free it now */
				if (dev->reg_state == NETREG_UNINITIALIZED)
					free_netdev(dev);
2690
				goto out;
2691
			}
2692
		} else {
2693
			err = register_netdevice(dev);
2694 2695 2696 2697
			if (err < 0) {
				free_netdev(dev);
				goto out;
			}
2698
		}
2699
		err = rtnl_configure_link(dev, ifm);
2700 2701
		if (err < 0)
			goto out_unregister;
2702
		if (link_net) {
2703
			err = dev_change_net_namespace(dev, dest_net, ifname);
2704
			if (err < 0)
2705
				goto out_unregister;
2706
		}
2707 2708 2709 2710 2711
		if (tb[IFLA_MASTER]) {
			err = do_set_master(dev, nla_get_u32(tb[IFLA_MASTER]));
			if (err)
				goto out_unregister;
		}
2712
out:
2713 2714
		if (link_net)
			put_net(link_net);
2715
		put_net(dest_net);
P
Patrick McHardy 已提交
2716
		return err;
2717 2718 2719 2720 2721 2722 2723 2724 2725 2726
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
2727 2728 2729
	}
}

2730 2731
static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
2732
{
2733
	struct net *net = sock_net(skb->sk);
2734
	struct ifinfomsg *ifm;
2735
	char ifname[IFNAMSIZ];
2736 2737 2738
	struct nlattr *tb[IFLA_MAX+1];
	struct net_device *dev = NULL;
	struct sk_buff *nskb;
2739
	int err;
2740
	u32 ext_filter_mask = 0;
2741

2742
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy, extack);
2743
	if (err < 0)
2744
		return err;
2745

2746 2747 2748
	if (tb[IFLA_IFNAME])
		nla_strlcpy(ifname, tb[IFLA_IFNAME], IFNAMSIZ);

2749 2750 2751
	if (tb[IFLA_EXT_MASK])
		ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);

2752
	ifm = nlmsg_data(nlh);
2753 2754 2755 2756 2757
	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
2758 2759
		return -EINVAL;

2760 2761 2762
	if (dev == NULL)
		return -ENODEV;

2763
	nskb = nlmsg_new(if_nlmsg_size(dev, ext_filter_mask), GFP_KERNEL);
2764 2765
	if (nskb == NULL)
		return -ENOBUFS;
2766

2767
	err = rtnl_fill_ifinfo(nskb, dev, RTM_NEWLINK, NETLINK_CB(skb).portid,
2768
			       nlh->nlmsg_seq, 0, 0, ext_filter_mask, 0);
2769 2770 2771 2772
	if (err < 0) {
		/* -EMSGSIZE implies BUG in if_nlmsg_size */
		WARN_ON(err == -EMSGSIZE);
		kfree_skb(nskb);
2773
	} else
2774
		err = rtnl_unicast(nskb, net, NETLINK_CB(skb).portid);
2775

2776
	return err;
2777 2778
}

2779
static u16 rtnl_calcit(struct sk_buff *skb, struct nlmsghdr *nlh)
2780
{
2781 2782 2783 2784 2785
	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;
2786 2787 2788 2789 2790
	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);
2791

2792
	if (nlmsg_parse(nlh, hdrlen, tb, IFLA_MAX, ifla_policy, NULL) >= 0) {
2793 2794 2795
		if (tb[IFLA_EXT_MASK])
			ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);
	}
2796 2797 2798 2799 2800 2801 2802

	if (!ext_filter_mask)
		return NLMSG_GOODSIZE;
	/*
	 * traverse the list of net devices and compute the minimum
	 * buffer size based upon the filter mask.
	 */
2803 2804
	rcu_read_lock();
	for_each_netdev_rcu(net, dev) {
2805 2806 2807 2808
		min_ifinfo_dump_size = max_t(u16, min_ifinfo_dump_size,
					     if_nlmsg_size(dev,
						           ext_filter_mask));
	}
2809
	rcu_read_unlock();
2810

2811
	return nlmsg_total_size(min_ifinfo_dump_size);
2812 2813
}

A
Adrian Bunk 已提交
2814
static int rtnl_dump_all(struct sk_buff *skb, struct netlink_callback *cb)
L
Linus Torvalds 已提交
2815 2816 2817 2818 2819 2820
{
	int idx;
	int s_idx = cb->family;

	if (s_idx == 0)
		s_idx = 1;
2821

2822
	for (idx = 1; idx <= RTNL_FAMILY_MAX; idx++) {
L
Linus Torvalds 已提交
2823
		int type = cb->nlh->nlmsg_type-RTM_BASE;
2824 2825 2826
		struct rtnl_link *handlers;
		rtnl_dumpit_func dumpit;

L
Linus Torvalds 已提交
2827 2828
		if (idx < s_idx || idx == PF_PACKET)
			continue;
2829 2830 2831

		handlers = rtnl_dereference(rtnl_msg_handlers[idx]);
		if (!handlers)
L
Linus Torvalds 已提交
2832
			continue;
2833 2834 2835 2836 2837

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

2838
		if (idx > s_idx) {
L
Linus Torvalds 已提交
2839
			memset(&cb->args[0], 0, sizeof(cb->args));
2840 2841 2842
			cb->prev_seq = 0;
			cb->seq = 0;
		}
2843
		if (dumpit(skb, cb))
L
Linus Torvalds 已提交
2844 2845 2846 2847 2848 2849 2850
			break;
	}
	cb->family = idx;

	return skb->len;
}

2851
struct sk_buff *rtmsg_ifinfo_build_skb(int type, struct net_device *dev,
2852 2853
				       unsigned int change,
				       u32 event, gfp_t flags)
L
Linus Torvalds 已提交
2854
{
2855
	struct net *net = dev_net(dev);
L
Linus Torvalds 已提交
2856
	struct sk_buff *skb;
2857
	int err = -ENOBUFS;
2858
	size_t if_info_size;
L
Linus Torvalds 已提交
2859

2860
	skb = nlmsg_new((if_info_size = if_nlmsg_size(dev, 0)), flags);
2861 2862
	if (skb == NULL)
		goto errout;
L
Linus Torvalds 已提交
2863

2864
	err = rtnl_fill_ifinfo(skb, dev, type, 0, 0, change, 0, 0, event);
2865 2866 2867 2868 2869 2870
	if (err < 0) {
		/* -EMSGSIZE implies BUG in if_nlmsg_size() */
		WARN_ON(err == -EMSGSIZE);
		kfree_skb(skb);
		goto errout;
	}
2871
	return skb;
2872 2873
errout:
	if (err < 0)
2874
		rtnl_set_sk_err(net, RTNLGRP_LINK, err);
2875 2876 2877 2878 2879 2880 2881 2882 2883 2884
	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);
}

2885 2886 2887
static void rtmsg_ifinfo_event(int type, struct net_device *dev,
			       unsigned int change, u32 event,
			       gfp_t flags)
2888 2889 2890
{
	struct sk_buff *skb;

2891 2892 2893
	if (dev->reg_state != NETREG_REGISTERED)
		return;

2894
	skb = rtmsg_ifinfo_build_skb(type, dev, change, event, flags);
2895 2896
	if (skb)
		rtmsg_ifinfo_send(skb, dev, flags);
L
Linus Torvalds 已提交
2897
}
2898 2899 2900 2901

void rtmsg_ifinfo(int type, struct net_device *dev, unsigned int change,
		  gfp_t flags)
{
2902
	rtmsg_ifinfo_event(type, dev, change, rtnl_get_event(0), flags);
2903
}
2904
EXPORT_SYMBOL(rtmsg_ifinfo);
L
Linus Torvalds 已提交
2905

2906 2907
static int nlmsg_populate_fdb_fill(struct sk_buff *skb,
				   struct net_device *dev,
2908
				   u8 *addr, u16 vid, u32 pid, u32 seq,
2909
				   int type, unsigned int flags,
2910
				   int nlflags, u16 ndm_state)
2911 2912 2913 2914
{
	struct nlmsghdr *nlh;
	struct ndmsg *ndm;

2915
	nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ndm), nlflags);
2916 2917 2918 2919 2920 2921 2922 2923 2924 2925
	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;
2926
	ndm->ndm_state   = ndm_state;
2927 2928 2929

	if (nla_put(skb, NDA_LLADDR, ETH_ALEN, addr))
		goto nla_put_failure;
2930 2931 2932
	if (vid)
		if (nla_put(skb, NDA_VLAN, sizeof(u16), &vid))
			goto nla_put_failure;
2933

2934 2935
	nlmsg_end(skb, nlh);
	return 0;
2936 2937 2938 2939 2940 2941

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

2942 2943
static inline size_t rtnl_fdb_nlmsg_size(void)
{
2944 2945 2946 2947
	return NLMSG_ALIGN(sizeof(struct ndmsg)) +
	       nla_total_size(ETH_ALEN) +	/* NDA_LLADDR */
	       nla_total_size(sizeof(u16)) +	/* NDA_VLAN */
	       0;
2948 2949
}

2950 2951
static void rtnl_fdb_notify(struct net_device *dev, u8 *addr, u16 vid, int type,
			    u16 ndm_state)
2952 2953 2954 2955 2956 2957 2958 2959 2960
{
	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;

2961
	err = nlmsg_populate_fdb_fill(skb, dev, addr, vid,
2962
				      0, 0, type, NTF_SELF, 0, ndm_state);
2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973
	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);
}

2974 2975 2976 2977 2978 2979
/**
 * 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,
2980
		     const unsigned char *addr, u16 vid,
2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992
		     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;
	}

2993 2994 2995 2996 2997
	if (vid) {
		pr_info("%s: vlans aren't supported yet for dev_uc|mc_add()\n", dev->name);
		return err;
	}

2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010
	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);

3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032
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;
}

3033 3034
static int rtnl_fdb_add(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
3035 3036 3037 3038 3039 3040
{
	struct net *net = sock_net(skb->sk);
	struct ndmsg *ndm;
	struct nlattr *tb[NDA_MAX+1];
	struct net_device *dev;
	u8 *addr;
3041
	u16 vid;
3042 3043
	int err;

3044
	err = nlmsg_parse(nlh, sizeof(*ndm), tb, NDA_MAX, NULL, extack);
3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066
	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]);

3067 3068 3069 3070
	err = fdb_vid_parse(tb[NDA_VLAN], &vid);
	if (err)
		return err;

3071 3072 3073 3074 3075
	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)) {
3076 3077 3078
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);
		const struct net_device_ops *ops = br_dev->netdev_ops;

3079 3080
		err = ops->ndo_fdb_add(ndm, tb, dev, addr, vid,
				       nlh->nlmsg_flags);
3081 3082 3083 3084 3085 3086 3087
		if (err)
			goto out;
		else
			ndm->ndm_flags &= ~NTF_MASTER;
	}

	/* Embedded bridge, macvlan, and any other device support */
3088 3089 3090
	if ((ndm->ndm_flags & NTF_SELF)) {
		if (dev->netdev_ops->ndo_fdb_add)
			err = dev->netdev_ops->ndo_fdb_add(ndm, tb, dev, addr,
3091
							   vid,
3092 3093
							   nlh->nlmsg_flags);
		else
3094
			err = ndo_dflt_fdb_add(ndm, tb, dev, addr, vid,
3095
					       nlh->nlmsg_flags);
3096

3097
		if (!err) {
3098 3099
			rtnl_fdb_notify(dev, addr, vid, RTM_NEWNEIGH,
					ndm->ndm_state);
3100
			ndm->ndm_flags &= ~NTF_SELF;
3101
		}
3102 3103 3104 3105 3106
	}
out:
	return err;
}

3107 3108 3109 3110 3111 3112
/**
 * 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,
3113
		     const unsigned char *addr, u16 vid)
3114
{
3115
	int err = -EINVAL;
3116 3117 3118 3119

	/* If aging addresses are supported device will need to
	 * implement its own handler for this.
	 */
3120
	if (!(ndm->ndm_state & NUD_PERMANENT)) {
3121
		pr_info("%s: FDB only supports static addresses\n", dev->name);
3122
		return err;
3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133
	}

	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);

3134 3135
static int rtnl_fdb_del(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
3136 3137 3138
{
	struct net *net = sock_net(skb->sk);
	struct ndmsg *ndm;
3139
	struct nlattr *tb[NDA_MAX+1];
3140 3141 3142
	struct net_device *dev;
	int err = -EINVAL;
	__u8 *addr;
3143
	u16 vid;
3144

3145
	if (!netlink_capable(skb, CAP_NET_ADMIN))
3146 3147
		return -EPERM;

3148
	err = nlmsg_parse(nlh, sizeof(*ndm), tb, NDA_MAX, NULL, extack);
3149 3150
	if (err < 0)
		return err;
3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163

	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;
	}

3164 3165 3166 3167 3168 3169
	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]);
3170

3171 3172 3173 3174
	err = fdb_vid_parse(tb[NDA_VLAN], &vid);
	if (err)
		return err;

3175 3176 3177 3178 3179
	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)) {
3180 3181
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);
		const struct net_device_ops *ops = br_dev->netdev_ops;
3182

3183
		if (ops->ndo_fdb_del)
3184
			err = ops->ndo_fdb_del(ndm, tb, dev, addr, vid);
3185 3186 3187 3188 3189 3190 3191 3192

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

	/* Embedded bridge, macvlan, and any other device support */
3193 3194
	if (ndm->ndm_flags & NTF_SELF) {
		if (dev->netdev_ops->ndo_fdb_del)
3195 3196
			err = dev->netdev_ops->ndo_fdb_del(ndm, tb, dev, addr,
							   vid);
3197
		else
3198
			err = ndo_dflt_fdb_del(ndm, tb, dev, addr, vid);
3199

3200
		if (!err) {
3201 3202
			rtnl_fdb_notify(dev, addr, vid, RTM_DELNEIGH,
					ndm->ndm_state);
3203
			ndm->ndm_flags &= ~NTF_SELF;
3204
		}
3205 3206 3207 3208 3209
	}
out:
	return err;
}

3210 3211 3212 3213 3214 3215 3216 3217
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;
3218
	u32 portid, seq;
3219

3220
	portid = NETLINK_CB(cb->skb).portid;
3221 3222 3223
	seq = cb->nlh->nlmsg_seq;

	list_for_each_entry(ha, &list->list, list) {
3224
		if (*idx < cb->args[2])
3225 3226
			goto skip;

3227
		err = nlmsg_populate_fdb_fill(skb, dev, ha->addr, 0,
3228
					      portid, seq,
3229
					      RTM_NEWNEIGH, NTF_SELF,
3230
					      NLM_F_MULTI, NUD_PERMANENT);
3231 3232 3233 3234 3235 3236 3237 3238 3239
		if (err < 0)
			return err;
skip:
		*idx += 1;
	}
	return 0;
}

/**
3240
 * ndo_dflt_fdb_dump - default netdevice operation to dump an FDB table.
3241 3242 3243 3244
 * @nlh: netlink message header
 * @dev: netdevice
 *
 * Default netdevice operation to dump the existing unicast address list.
3245
 * Returns number of addresses from list put in skb.
3246 3247 3248 3249
 */
int ndo_dflt_fdb_dump(struct sk_buff *skb,
		      struct netlink_callback *cb,
		      struct net_device *dev,
3250
		      struct net_device *filter_dev,
3251
		      int *idx)
3252 3253 3254 3255
{
	int err;

	netif_addr_lock_bh(dev);
3256
	err = nlmsg_populate_fdb(skb, cb, dev, idx, &dev->uc);
3257 3258
	if (err)
		goto out;
3259
	err = nlmsg_populate_fdb(skb, cb, dev, idx, &dev->mc);
3260 3261
out:
	netif_addr_unlock_bh(dev);
3262
	return err;
3263 3264 3265
}
EXPORT_SYMBOL(ndo_dflt_fdb_dump);

3266 3267 3268
static int rtnl_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb)
{
	struct net_device *dev;
3269 3270 3271 3272 3273 3274
	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);
3275
	struct hlist_head *head;
3276 3277
	int brport_idx = 0;
	int br_idx = 0;
3278 3279 3280 3281
	int h, s_h;
	int idx = 0, s_idx;
	int err = 0;
	int fidx = 0;
3282

3283 3284 3285 3286 3287
	err = nlmsg_parse(cb->nlh, sizeof(struct ifinfomsg), tb,
			  IFLA_MAX, ifla_policy, NULL);
	if (err < 0) {
		return -EINVAL;
	} else if (err == 0) {
3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301
		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;
	}

3302 3303
	s_h = cb->args[0];
	s_idx = cb->args[1];
3304

3305 3306 3307 3308
	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) {
3309

3310
			if (brport_idx && (dev->ifindex != brport_idx))
3311 3312
				continue;

3313 3314 3315 3316 3317 3318 3319 3320 3321
			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;
3322

3323 3324 3325 3326 3327
				if (br_dev != netdev_master_upper_dev_get(dev) &&
				    !(dev->priv_flags & IFF_EBRIDGE))
					continue;
				cops = ops;
			}
3328

3329 3330
			if (idx < s_idx)
				goto cont;
3331

3332 3333 3334 3335 3336 3337 3338 3339 3340
			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;
				}
			}
3341

3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359
			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++;
		}
3360 3361
	}

3362 3363 3364 3365 3366
out:
	cb->args[0] = h;
	cb->args[1] = idx;
	cb->args[2] = fidx;

3367 3368 3369
	return skb->len;
}

3370 3371 3372 3373 3374 3375 3376 3377
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;
}

3378
int ndo_dflt_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
3379
			    struct net_device *dev, u16 mode,
3380 3381 3382 3383 3384
			    u32 flags, u32 mask, int nlflags,
			    u32 filter_mask,
			    int (*vlan_fill)(struct sk_buff *skb,
					     struct net_device *dev,
					     u32 filter_mask))
3385 3386 3387 3388
{
	struct nlmsghdr *nlh;
	struct ifinfomsg *ifm;
	struct nlattr *br_afspec;
3389
	struct nlattr *protinfo;
3390
	u8 operstate = netif_running(dev) ? dev->operstate : IF_OPER_DOWN;
3391
	struct net_device *br_dev = netdev_master_upper_dev_get(dev);
3392
	int err = 0;
3393

3394
	nlh = nlmsg_put(skb, pid, seq, RTM_NEWLINK, sizeof(*ifm), nlflags);
3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409
	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) ||
3410 3411
	    (br_dev &&
	     nla_put_u32(skb, IFLA_MASTER, br_dev->ifindex)) ||
3412 3413
	    (dev->addr_len &&
	     nla_put(skb, IFLA_ADDRESS, dev->addr_len, dev->dev_addr)) ||
3414 3415
	    (dev->ifindex != dev_get_iflink(dev) &&
	     nla_put_u32(skb, IFLA_LINK, dev_get_iflink(dev))))
3416 3417 3418 3419 3420 3421
		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
3422
	if (nla_put_u16(skb, IFLA_BRIDGE_FLAGS, BRIDGE_FLAGS_SELF)) {
3423 3424 3425
		nla_nest_cancel(skb, br_afspec);
		goto nla_put_failure;
	}
R
Roopa Prabhu 已提交
3426 3427 3428 3429 3430 3431 3432

	if (mode != BRIDGE_MODE_UNDEF) {
		if (nla_put_u16(skb, IFLA_BRIDGE_MODE, mode)) {
			nla_nest_cancel(skb, br_afspec);
			goto nla_put_failure;
		}
	}
3433 3434 3435 3436 3437 3438 3439
	if (vlan_fill) {
		err = vlan_fill(skb, dev, filter_mask);
		if (err) {
			nla_nest_cancel(skb, br_afspec);
			goto nla_put_failure;
		}
	}
3440 3441
	nla_nest_end(skb, br_afspec);

3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468
	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);

3469 3470
	nlmsg_end(skb, nlh);
	return 0;
3471 3472
nla_put_failure:
	nlmsg_cancel(skb, nlh);
3473
	return err ? err : -EMSGSIZE;
3474
}
3475
EXPORT_SYMBOL_GPL(ndo_dflt_bridge_getlink);
3476

J
John Fastabend 已提交
3477 3478 3479 3480 3481 3482 3483
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;
3484
	u32 filter_mask = 0;
3485
	int err;
3486

3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498
	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
3499 3500 3501 3502

	rcu_read_lock();
	for_each_netdev_rcu(net, dev) {
		const struct net_device_ops *ops = dev->netdev_ops;
3503
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);
J
John Fastabend 已提交
3504

3505
		if (br_dev && br_dev->netdev_ops->ndo_bridge_getlink) {
3506 3507 3508 3509
			if (idx >= cb->args[0]) {
				err = br_dev->netdev_ops->ndo_bridge_getlink(
						skb, portid, seq, dev,
						filter_mask, NLM_F_MULTI);
3510 3511 3512 3513 3514 3515
				if (err < 0 && err != -EOPNOTSUPP) {
					if (likely(skb->len))
						break;

					goto out_err;
				}
3516
			}
3517
			idx++;
J
John Fastabend 已提交
3518 3519 3520
		}

		if (ops->ndo_bridge_getlink) {
3521 3522 3523 3524 3525
			if (idx >= cb->args[0]) {
				err = ops->ndo_bridge_getlink(skb, portid,
							      seq, dev,
							      filter_mask,
							      NLM_F_MULTI);
3526 3527 3528 3529 3530 3531
				if (err < 0 && err != -EOPNOTSUPP) {
					if (likely(skb->len))
						break;

					goto out_err;
				}
3532
			}
3533
			idx++;
J
John Fastabend 已提交
3534 3535
		}
	}
3536 3537
	err = skb->len;
out_err:
J
John Fastabend 已提交
3538 3539 3540
	rcu_read_unlock();
	cb->args[0] = idx;

3541
	return err;
J
John Fastabend 已提交
3542 3543
}

3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558
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 */
}

3559
static int rtnl_bridge_notify(struct net_device *dev)
3560 3561 3562 3563 3564
{
	struct net *net = dev_net(dev);
	struct sk_buff *skb;
	int err = -EOPNOTSUPP;

3565 3566 3567
	if (!dev->netdev_ops->ndo_bridge_getlink)
		return 0;

3568 3569 3570 3571 3572 3573
	skb = nlmsg_new(bridge_nlmsg_size(), GFP_ATOMIC);
	if (!skb) {
		err = -ENOMEM;
		goto errout;
	}

3574
	err = dev->netdev_ops->ndo_bridge_getlink(skb, 0, 0, dev, 0, 0);
3575 3576
	if (err < 0)
		goto errout;
3577

3578 3579 3580
	if (!skb->len)
		goto errout;

3581 3582 3583 3584 3585
	rtnl_notify(skb, net, 0, RTNLGRP_LINK, NULL, GFP_ATOMIC);
	return 0;
errout:
	WARN_ON(err == -EMSGSIZE);
	kfree_skb(skb);
3586 3587
	if (err)
		rtnl_set_sk_err(net, RTNLGRP_LINK, err);
3588 3589 3590
	return err;
}

3591 3592
static int rtnl_bridge_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			       struct netlink_ext_ack *extack)
J
John Fastabend 已提交
3593 3594 3595 3596
{
	struct net *net = sock_net(skb->sk);
	struct ifinfomsg *ifm;
	struct net_device *dev;
3597 3598
	struct nlattr *br_spec, *attr = NULL;
	int rem, err = -EOPNOTSUPP;
3599
	u16 flags = 0;
3600
	bool have_flags = false;
J
John Fastabend 已提交
3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611 3612 3613 3614

	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;
	}

3615 3616 3617 3618
	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) {
3619 3620 3621
				if (nla_len(attr) < sizeof(flags))
					return -EINVAL;

3622
				have_flags = true;
3623 3624 3625 3626 3627 3628 3629
				flags = nla_get_u16(attr);
				break;
			}
		}
	}

	if (!flags || (flags & BRIDGE_FLAGS_MASTER)) {
3630 3631 3632
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);

		if (!br_dev || !br_dev->netdev_ops->ndo_bridge_setlink) {
3633 3634 3635 3636
			err = -EOPNOTSUPP;
			goto out;
		}

3637
		err = br_dev->netdev_ops->ndo_bridge_setlink(dev, nlh, flags);
J
John Fastabend 已提交
3638 3639
		if (err)
			goto out;
3640 3641

		flags &= ~BRIDGE_FLAGS_MASTER;
J
John Fastabend 已提交
3642 3643
	}

3644 3645 3646 3647
	if ((flags & BRIDGE_FLAGS_SELF)) {
		if (!dev->netdev_ops->ndo_bridge_setlink)
			err = -EOPNOTSUPP;
		else
3648 3649
			err = dev->netdev_ops->ndo_bridge_setlink(dev, nlh,
								  flags);
3650
		if (!err) {
3651
			flags &= ~BRIDGE_FLAGS_SELF;
3652 3653 3654 3655 3656 3657

			/* Generate event to notify upper layer of bridge
			 * change
			 */
			err = rtnl_bridge_notify(dev);
		}
3658
	}
J
John Fastabend 已提交
3659

3660
	if (have_flags)
3661
		memcpy(nla_data(attr), &flags, sizeof(flags));
J
John Fastabend 已提交
3662 3663 3664 3665
out:
	return err;
}

3666 3667
static int rtnl_bridge_dellink(struct sk_buff *skb, struct nlmsghdr *nlh,
			       struct netlink_ext_ack *extack)
3668 3669 3670 3671 3672 3673
{
	struct net *net = sock_net(skb->sk);
	struct ifinfomsg *ifm;
	struct net_device *dev;
	struct nlattr *br_spec, *attr = NULL;
	int rem, err = -EOPNOTSUPP;
3674
	u16 flags = 0;
3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693
	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) {
3694 3695 3696
				if (nla_len(attr) < sizeof(flags))
					return -EINVAL;

3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711
				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;
		}

3712
		err = br_dev->netdev_ops->ndo_bridge_dellink(dev, nlh, flags);
3713 3714 3715 3716 3717 3718 3719 3720 3721 3722
		if (err)
			goto out;

		flags &= ~BRIDGE_FLAGS_MASTER;
	}

	if ((flags & BRIDGE_FLAGS_SELF)) {
		if (!dev->netdev_ops->ndo_bridge_dellink)
			err = -EOPNOTSUPP;
		else
3723 3724
			err = dev->netdev_ops->ndo_bridge_dellink(dev, nlh,
								  flags);
3725

3726
		if (!err) {
3727
			flags &= ~BRIDGE_FLAGS_SELF;
3728 3729 3730 3731 3732 3733

			/* Generate event to notify upper layer of bridge
			 * change
			 */
			err = rtnl_bridge_notify(dev);
		}
3734 3735 3736 3737 3738 3739 3740 3741
	}

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

3742 3743 3744 3745 3746 3747
static bool stats_attr_valid(unsigned int mask, int attrid, int idxattr)
{
	return (mask & IFLA_STATS_FILTER_BIT(attrid)) &&
	       (!idxattr || idxattr == attrid);
}

3748 3749 3750 3751 3752 3753 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
#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;

3780
		if (!dev->netdev_ops->ndo_has_offload_stats(dev, attr_id))
3781 3782 3783 3784 3785 3786 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
			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++) {
3821
		if (!dev->netdev_ops->ndo_has_offload_stats(dev, attr_id))
3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832
			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;
}

3833 3834
static int rtnl_fill_statsinfo(struct sk_buff *skb, struct net_device *dev,
			       int type, u32 pid, u32 seq, u32 change,
3835 3836
			       unsigned int flags, unsigned int filter_mask,
			       int *idxattr, int *prividx)
3837 3838 3839 3840
{
	struct if_stats_msg *ifsm;
	struct nlmsghdr *nlh;
	struct nlattr *attr;
3841
	int s_prividx = *prividx;
3842
	int err;
3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853

	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;

3854
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_64, *idxattr)) {
3855 3856
		struct rtnl_link_stats64 *sp;

3857 3858 3859
		attr = nla_reserve_64bit(skb, IFLA_STATS_LINK_64,
					 sizeof(struct rtnl_link_stats64),
					 IFLA_STATS_UNSPEC);
3860 3861 3862 3863 3864 3865 3866
		if (!attr)
			goto nla_put_failure;

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

3867 3868 3869 3870 3871 3872 3873 3874 3875 3876
	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;

3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900
			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);
3901 3902 3903 3904 3905 3906 3907
			nla_nest_end(skb, attr);
			if (err)
				goto nla_put_failure;
			*idxattr = 0;
		}
	}

3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925
	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;
	}

3926 3927 3928 3929 3930 3931 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
	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;
	}

3959 3960 3961 3962 3963
	nlmsg_end(skb, nlh);

	return 0;

nla_put_failure:
3964 3965 3966 3967 3968
	/* 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);
3969 3970 3971 3972 3973 3974 3975 3976 3977

	return -EMSGSIZE;
}

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

3978
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_64, 0))
3979 3980
		size += nla_total_size_64bit(sizeof(struct rtnl_link_stats64));

3981 3982
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_XSTATS, 0)) {
		const struct rtnl_link_ops *ops = dev->rtnl_link_ops;
3983
		int attr = IFLA_STATS_LINK_XSTATS;
3984 3985

		if (ops && ops->get_linkxstats_size) {
3986 3987
			size += nla_total_size(ops->get_linkxstats_size(dev,
									attr));
3988 3989 3990 3991 3992
			/* for IFLA_STATS_LINK_XSTATS */
			size += nla_total_size(0);
		}
	}

3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007 4008 4009 4010 4011
	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);
		}
	}

4012 4013 4014
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_OFFLOAD_XSTATS, 0))
		size += rtnl_get_offload_stats_size(dev);

4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031
	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);
			}
		}
	}

4032 4033 4034
	return size;
}

4035 4036
static int rtnl_stats_get(struct sk_buff *skb, struct nlmsghdr *nlh,
			  struct netlink_ext_ack *extack)
4037 4038 4039
{
	struct net *net = sock_net(skb->sk);
	struct net_device *dev = NULL;
4040 4041
	int idxattr = 0, prividx = 0;
	struct if_stats_msg *ifsm;
4042 4043 4044 4045
	struct sk_buff *nskb;
	u32 filter_mask;
	int err;

4046 4047 4048
	if (nlmsg_len(nlh) < sizeof(*ifsm))
		return -EINVAL;

4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067
	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,
4068
				  0, filter_mask, &idxattr, &prividx);
4069 4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081
	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)
{
4082
	int h, s_h, err, s_idx, s_idxattr, s_prividx;
4083
	struct net *net = sock_net(skb->sk);
4084
	unsigned int flags = NLM_F_MULTI;
4085 4086
	struct if_stats_msg *ifsm;
	struct hlist_head *head;
4087
	struct net_device *dev;
4088
	u32 filter_mask = 0;
4089
	int idx = 0;
4090 4091 4092

	s_h = cb->args[0];
	s_idx = cb->args[1];
4093 4094
	s_idxattr = cb->args[2];
	s_prividx = cb->args[3];
4095 4096 4097

	cb->seq = net->dev_base_seq;

4098 4099 4100
	if (nlmsg_len(cb->nlh) < sizeof(*ifsm))
		return -EINVAL;

4101 4102 4103 4104 4105 4106 4107 4108 4109 4110 4111 4112 4113 4114
	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,
4115 4116
						  flags, filter_mask,
						  &s_idxattr, &s_prividx);
4117 4118 4119 4120 4121 4122 4123
			/* 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;
4124 4125
			s_prividx = 0;
			s_idxattr = 0;
4126 4127 4128 4129 4130 4131
			nl_dump_check_consistent(cb, nlmsg_hdr(skb));
cont:
			idx++;
		}
	}
out:
4132 4133
	cb->args[3] = s_prividx;
	cb->args[2] = s_idxattr;
4134 4135 4136 4137 4138 4139
	cb->args[1] = idx;
	cb->args[0] = h;

	return skb->len;
}

L
Linus Torvalds 已提交
4140 4141
/* Process one rtnetlink message. */

J
Johannes Berg 已提交
4142 4143
static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
			     struct netlink_ext_ack *extack)
L
Linus Torvalds 已提交
4144
{
4145
	struct net *net = sock_net(skb->sk);
4146 4147
	struct rtnl_link *handlers;
	int err = -EOPNOTSUPP;
4148
	rtnl_doit_func doit;
4149
	unsigned int flags;
4150
	int kind;
L
Linus Torvalds 已提交
4151 4152 4153 4154 4155
	int family;
	int type;

	type = nlh->nlmsg_type;
	if (type > RTM_MAX)
4156
		return -EOPNOTSUPP;
L
Linus Torvalds 已提交
4157 4158 4159 4160

	type -= RTM_BASE;

	/* All the messages must have at least 1 byte length */
4161
	if (nlmsg_len(nlh) < sizeof(struct rtgenmsg))
L
Linus Torvalds 已提交
4162 4163
		return 0;

4164
	family = ((struct rtgenmsg *)nlmsg_data(nlh))->rtgen_family;
L
Linus Torvalds 已提交
4165 4166
	kind = type&3;

4167
	if (kind != 2 && !netlink_net_capable(skb, CAP_NET_ADMIN))
4168
		return -EPERM;
L
Linus Torvalds 已提交
4169

4170
	if (family >= ARRAY_SIZE(rtnl_msg_handlers))
4171 4172 4173 4174 4175 4176 4177 4178 4179
		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]);
	}

4180
	if (kind == 2 && nlh->nlmsg_flags&NLM_F_DUMP) {
4181
		struct sock *rtnl;
4182
		rtnl_dumpit_func dumpit;
4183
		u16 min_dump_alloc = 0;
L
Linus Torvalds 已提交
4184

4185 4186 4187 4188 4189 4190
		dumpit = READ_ONCE(handlers[type].dumpit);
		if (!dumpit) {
			family = PF_UNSPEC;
			handlers = rcu_dereference(rtnl_msg_handlers[PF_UNSPEC]);
			if (!handlers)
				goto err_unlock;
4191

4192 4193 4194 4195
			dumpit = READ_ONCE(handlers[type].dumpit);
			if (!dumpit)
				goto err_unlock;
		}
4196

4197 4198
		refcount_inc(&rtnl_msg_handlers_ref[family]);

4199
		if (type == RTM_GETLINK - RTM_BASE)
4200
			min_dump_alloc = rtnl_calcit(skb, nlh);
4201

4202 4203
		rcu_read_unlock();

4204
		rtnl = net->rtnl;
4205 4206 4207 4208 4209 4210 4211
		{
			struct netlink_dump_control c = {
				.dump		= dumpit,
				.min_dump_alloc	= min_dump_alloc,
			};
			err = netlink_dump_start(rtnl, skb, nlh, &c);
		}
4212
		refcount_dec(&rtnl_msg_handlers_ref[family]);
4213
		return err;
L
Linus Torvalds 已提交
4214 4215
	}

4216 4217 4218 4219 4220 4221 4222 4223 4224 4225 4226
	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;
	}

4227
	rcu_read_unlock();
L
Linus Torvalds 已提交
4228

4229 4230 4231 4232 4233 4234 4235
	rtnl_lock();
	handlers = rtnl_dereference(rtnl_msg_handlers[family]);
	if (handlers) {
		doit = READ_ONCE(handlers[type].doit);
		if (doit)
			err = doit(skb, nlh, extack);
	}
4236 4237 4238 4239
	rtnl_unlock();
	return err;

err_unlock:
4240
	rcu_read_unlock();
4241
	return -EOPNOTSUPP;
L
Linus Torvalds 已提交
4242 4243
}

4244
static void rtnetlink_rcv(struct sk_buff *skb)
L
Linus Torvalds 已提交
4245
{
4246
	netlink_rcv_skb(skb, &rtnetlink_rcv_msg);
L
Linus Torvalds 已提交
4247 4248
}

4249 4250 4251 4252 4253 4254 4255 4256 4257 4258 4259 4260
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
4261 4262
static int rtnetlink_event(struct notifier_block *this, unsigned long event, void *ptr)
{
4263
	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
4264

L
Linus Torvalds 已提交
4265
	switch (event) {
4266
	case NETDEV_REBOOT:
4267
	case NETDEV_CHANGEADDR:
4268 4269 4270 4271 4272 4273
	case NETDEV_CHANGENAME:
	case NETDEV_FEAT_CHANGE:
	case NETDEV_BONDING_FAILOVER:
	case NETDEV_NOTIFY_PEERS:
	case NETDEV_RESEND_IGMP:
	case NETDEV_CHANGEINFODATA:
4274 4275
		rtmsg_ifinfo_event(RTM_NEWLINK, dev, 0, rtnl_get_event(event),
				   GFP_KERNEL);
L
Linus Torvalds 已提交
4276 4277 4278 4279 4280 4281 4282 4283 4284 4285 4286
		break;
	default:
		break;
	}
	return NOTIFY_DONE;
}

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

4287

4288
static int __net_init rtnetlink_net_init(struct net *net)
4289 4290
{
	struct sock *sk;
4291 4292 4293 4294
	struct netlink_kernel_cfg cfg = {
		.groups		= RTNLGRP_MAX,
		.input		= rtnetlink_rcv,
		.cb_mutex	= &rtnl_mutex,
4295
		.flags		= NL_CFG_F_NONROOT_RECV,
4296
		.bind		= rtnetlink_bind,
4297 4298
	};

4299
	sk = netlink_kernel_create(net, NETLINK_ROUTE, &cfg);
4300 4301 4302 4303 4304 4305
	if (!sk)
		return -ENOMEM;
	net->rtnl = sk;
	return 0;
}

4306
static void __net_exit rtnetlink_net_exit(struct net *net)
4307
{
4308 4309
	netlink_kernel_release(net->rtnl);
	net->rtnl = NULL;
4310 4311 4312 4313 4314 4315 4316
}

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

L
Linus Torvalds 已提交
4317 4318
void __init rtnetlink_init(void)
{
4319
	if (register_pernet_subsys(&rtnetlink_net_ops))
L
Linus Torvalds 已提交
4320
		panic("rtnetlink_init: cannot initialize rtnetlink\n");
4321

L
Linus Torvalds 已提交
4322
	register_netdevice_notifier(&rtnetlink_dev_notifier);
4323

4324
	rtnl_register(PF_UNSPEC, RTM_GETLINK, rtnl_getlink,
4325 4326 4327 4328
		      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);
4329

4330 4331 4332
	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);
4333

4334 4335 4336
	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
4337

4338 4339 4340
	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);
4341 4342

	rtnl_register(PF_UNSPEC, RTM_GETSTATS, rtnl_stats_get, rtnl_stats_dump,
4343
		      0);
L
Linus Torvalds 已提交
4344
}