rtnetlink.c 104.1 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 65 66
	rtnl_doit_func		doit;
	rtnl_dumpit_func	dumpit;
};

67
static DEFINE_MUTEX(rtnl_mutex);
L
Linus Torvalds 已提交
68 69 70

void rtnl_lock(void)
{
71
	mutex_lock(&rtnl_mutex);
L
Linus Torvalds 已提交
72
}
E
Eric Dumazet 已提交
73
EXPORT_SYMBOL(rtnl_lock);
L
Linus Torvalds 已提交
74

75 76 77 78 79 80 81 82 83 84
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);

85
void __rtnl_unlock(void)
L
Linus Torvalds 已提交
86
{
87 88 89 90
	struct sk_buff *head = defer_kfree_skb_list;

	defer_kfree_skb_list = NULL;

91
	mutex_unlock(&rtnl_mutex);
92 93 94 95 96 97 98 99

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

		kfree_skb(head);
		cond_resched();
		head = next;
	}
L
Linus Torvalds 已提交
100
}
101

L
Linus Torvalds 已提交
102 103
void rtnl_unlock(void)
{
H
Herbert Xu 已提交
104
	/* This fellow will unlock it for us. */
L
Linus Torvalds 已提交
105 106
	netdev_run_todo();
}
E
Eric Dumazet 已提交
107
EXPORT_SYMBOL(rtnl_unlock);
L
Linus Torvalds 已提交
108

109 110 111 112
int rtnl_trylock(void)
{
	return mutex_trylock(&rtnl_mutex);
}
E
Eric Dumazet 已提交
113
EXPORT_SYMBOL(rtnl_trylock);
114

115 116 117 118
int rtnl_is_locked(void)
{
	return mutex_is_locked(&rtnl_mutex);
}
E
Eric Dumazet 已提交
119
EXPORT_SYMBOL(rtnl_is_locked);
120

121
#ifdef CONFIG_PROVE_LOCKING
122
bool lockdep_rtnl_is_held(void)
123 124 125 126 127 128
{
	return lockdep_is_held(&rtnl_mutex);
}
EXPORT_SYMBOL(lockdep_rtnl_is_held);
#endif /* #ifdef CONFIG_PROVE_LOCKING */

129
static struct rtnl_link __rcu *rtnl_msg_handlers[RTNL_FAMILY_MAX + 1];
130
static refcount_t rtnl_msg_handlers_ref[RTNL_FAMILY_MAX + 1];
131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151

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
152
 * @flags: rtnl_link_flags to modifiy behaviour of doit/dumpit functions
153 154 155 156 157 158 159 160 161 162 163 164
 *
 * 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,
165
		    rtnl_doit_func doit, rtnl_dumpit_func dumpit,
166
		    unsigned int flags)
167 168 169 170
{
	struct rtnl_link *tab;
	int msgindex;

171
	BUG_ON(protocol < 0 || protocol > RTNL_FAMILY_MAX);
172 173
	msgindex = rtm_msgindex(msgtype);

174
	tab = rcu_dereference(rtnl_msg_handlers[protocol]);
175 176 177 178 179
	if (tab == NULL) {
		tab = kcalloc(RTM_NR_MSGTYPES, sizeof(*tab), GFP_KERNEL);
		if (tab == NULL)
			return -ENOBUFS;

180
		rcu_assign_pointer(rtnl_msg_handlers[protocol], tab);
181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198
	}

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

	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
199
 * of memory implies no sense in continuing.
200 201
 */
void rtnl_register(int protocol, int msgtype,
202
		   rtnl_doit_func doit, rtnl_dumpit_func dumpit,
203
		   unsigned int flags)
204
{
205
	if (__rtnl_register(protocol, msgtype, doit, dumpit, flags) < 0)
206 207 208 209 210 211 212 213 214 215 216 217 218 219 220
		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)
{
221
	struct rtnl_link *handlers;
222 223
	int msgindex;

224
	BUG_ON(protocol < 0 || protocol > RTNL_FAMILY_MAX);
225 226
	msgindex = rtm_msgindex(msgtype);

227 228 229 230
	rtnl_lock();
	handlers = rtnl_dereference(rtnl_msg_handlers[protocol]);
	if (!handlers) {
		rtnl_unlock();
231
		return -ENOENT;
232
	}
233

234 235 236
	handlers[msgindex].doit = NULL;
	handlers[msgindex].dumpit = NULL;
	rtnl_unlock();
237 238 239 240 241 242 243 244 245 246 247 248 249 250

	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)
{
251 252
	struct rtnl_link *handlers;

253
	BUG_ON(protocol < 0 || protocol > RTNL_FAMILY_MAX);
254

255
	rtnl_lock();
256 257
	handlers = rtnl_dereference(rtnl_msg_handlers[protocol]);
	RCU_INIT_POINTER(rtnl_msg_handlers[protocol], NULL);
258 259
	rtnl_unlock();

260 261
	synchronize_net();

262 263 264
	while (refcount_read(&rtnl_msg_handlers_ref[protocol]) > 0)
		schedule();
	kfree(handlers);
265 266
}
EXPORT_SYMBOL_GPL(rtnl_unregister_all);
L
Linus Torvalds 已提交
267

P
Patrick McHardy 已提交
268 269
static LIST_HEAD(link_ops);

270 271 272 273 274 275 276 277 278 279 280
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
281 282 283 284 285 286 287 288 289 290 291 292
/**
 * __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)
{
293 294 295
	if (rtnl_link_ops_get(ops->kind))
		return -EEXIST;

296 297 298 299 300 301
	/* 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)
302
		ops->dellink = unregister_netdevice_queue;
303

P
Patrick McHardy 已提交
304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325
	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);

326 327 328
static void __rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops)
{
	struct net_device *dev;
329 330
	LIST_HEAD(list_kill);

331
	for_each_netdev(net, dev) {
332 333
		if (dev->rtnl_link_ops == ops)
			ops->dellink(dev, &list_kill);
334
	}
335
	unregister_netdevice_many(&list_kill);
336 337
}

P
Patrick McHardy 已提交
338 339 340 341
/**
 * __rtnl_link_unregister - Unregister rtnl_link_ops from rtnetlink.
 * @ops: struct rtnl_link_ops * to unregister
 *
342
 * The caller must hold the rtnl_mutex.
P
Patrick McHardy 已提交
343 344 345
 */
void __rtnl_link_unregister(struct rtnl_link_ops *ops)
{
346
	struct net *net;
347

348
	for_each_net(net) {
349
		__rtnl_kill_links(net, ops);
350
	}
P
Patrick McHardy 已提交
351 352 353 354
	list_del(&ops->list);
}
EXPORT_SYMBOL_GPL(__rtnl_link_unregister);

355 356 357 358 359 360 361
/* 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;
362
	DEFINE_WAIT_FUNC(wait, woken_wake_function);
363

364
	add_wait_queue(&netdev_unregistering_wq, &wait);
365 366 367 368 369 370 371 372 373 374 375 376
	for (;;) {
		unregistering = false;
		rtnl_lock();
		for_each_net(net) {
			if (net->dev_unreg_count > 0) {
				unregistering = true;
				break;
			}
		}
		if (!unregistering)
			break;
		__rtnl_unlock();
377 378

		wait_woken(&wait, TASK_UNINTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT);
379
	}
380
	remove_wait_queue(&netdev_unregistering_wq, &wait);
381 382
}

P
Patrick McHardy 已提交
383 384 385 386 387 388
/**
 * 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)
{
389 390 391
	/* Close the race with cleanup_net() */
	mutex_lock(&net_mutex);
	rtnl_lock_unregistering_all();
P
Patrick McHardy 已提交
392 393
	__rtnl_link_unregister(ops);
	rtnl_unlock();
394
	mutex_unlock(&net_mutex);
P
Patrick McHardy 已提交
395 396 397
}
EXPORT_SYMBOL_GPL(rtnl_link_unregister);

398 399 400 401 402 403 404 405 406
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;
407
	if (!ops || !ops->get_slave_size)
408 409 410 411 412 413
		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
414 415 416 417 418 419 420 421
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;

422 423
	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
424 425 426

	if (ops->get_size)
		/* IFLA_INFO_DATA + nested data */
427
		size += nla_total_size(sizeof(struct nlattr)) +
P
Patrick McHardy 已提交
428 429 430
			ops->get_size(dev);

	if (ops->get_xstats_size)
431 432
		/* IFLA_INFO_XSTATS */
		size += nla_total_size(ops->get_xstats_size(dev));
P
Patrick McHardy 已提交
433

434 435
	size += rtnl_link_get_slave_info_data_size(dev);

P
Patrick McHardy 已提交
436 437 438
	return size;
}

T
Thomas Graf 已提交
439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458
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.
 */
459
void rtnl_af_register(struct rtnl_af_ops *ops)
T
Thomas Graf 已提交
460 461
{
	rtnl_lock();
462
	list_add_tail(&ops->list, &rtnl_af_ops);
T
Thomas Graf 已提交
463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490
	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);

491 492
static size_t rtnl_link_get_af_size(const struct net_device *dev,
				    u32 ext_filter_mask)
T
Thomas Graf 已提交
493 494 495 496 497 498 499 500 501 502 503
{
	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)) +
504
				af_ops->get_link_af_size(dev, ext_filter_mask);
T
Thomas Graf 已提交
505 506 507 508 509 510
		}
	}

	return size;
}

511
static bool rtnl_have_link_slave_info(const struct net_device *dev)
P
Patrick McHardy 已提交
512
{
513
	struct net_device *master_dev;
P
Patrick McHardy 已提交
514

515
	master_dev = netdev_master_upper_dev_get((struct net_device *) dev);
516
	if (master_dev && master_dev->rtnl_link_ops)
517 518 519 520 521 522 523 524 525 526 527
		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
528

529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561
	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
562
	if (nla_put_string(skb, IFLA_INFO_KIND, ops->kind) < 0)
563
		return -EMSGSIZE;
P
Patrick McHardy 已提交
564 565 566
	if (ops->fill_xstats) {
		err = ops->fill_xstats(skb, dev);
		if (err < 0)
567
			return err;
P
Patrick McHardy 已提交
568 569 570
	}
	if (ops->fill_info) {
		data = nla_nest_start(skb, IFLA_INFO_DATA);
571 572
		if (data == NULL)
			return -EMSGSIZE;
P
Patrick McHardy 已提交
573 574 575 576 577 578 579 580 581
		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);
582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604
	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
605 606 607 608 609 610
err_cancel_link:
	nla_nest_cancel(skb, linkinfo);
out:
	return err;
}

611
int rtnetlink_send(struct sk_buff *skb, struct net *net, u32 pid, unsigned int group, int echo)
L
Linus Torvalds 已提交
612
{
613
	struct sock *rtnl = net->rtnl;
L
Linus Torvalds 已提交
614 615
	int err = 0;

616
	NETLINK_CB(skb).dst_group = group;
L
Linus Torvalds 已提交
617
	if (echo)
618
		refcount_inc(&skb->users);
L
Linus Torvalds 已提交
619 620 621 622 623 624
	netlink_broadcast(rtnl, skb, pid, group, GFP_KERNEL);
	if (echo)
		err = netlink_unicast(rtnl, skb, pid, MSG_DONTWAIT);
	return err;
}

625
int rtnl_unicast(struct sk_buff *skb, struct net *net, u32 pid)
626
{
627 628
	struct sock *rtnl = net->rtnl;

629 630
	return nlmsg_unicast(rtnl, skb, pid);
}
E
Eric Dumazet 已提交
631
EXPORT_SYMBOL(rtnl_unicast);
632

633 634
void rtnl_notify(struct sk_buff *skb, struct net *net, u32 pid, u32 group,
		 struct nlmsghdr *nlh, gfp_t flags)
635
{
636
	struct sock *rtnl = net->rtnl;
637 638 639 640 641
	int report = 0;

	if (nlh)
		report = nlmsg_report(nlh);

642
	nlmsg_notify(rtnl, skb, pid, group, report, flags);
643
}
E
Eric Dumazet 已提交
644
EXPORT_SYMBOL(rtnl_notify);
645

646
void rtnl_set_sk_err(struct net *net, u32 group, int error)
647
{
648 649
	struct sock *rtnl = net->rtnl;

650 651
	netlink_set_err(rtnl, 0, group, error);
}
E
Eric Dumazet 已提交
652
EXPORT_SYMBOL(rtnl_set_sk_err);
653

L
Linus Torvalds 已提交
654 655
int rtnetlink_put_metrics(struct sk_buff *skb, u32 *metrics)
{
656 657 658 659 660 661 662 663 664
	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]) {
665 666 667 668 669 670 671 672
			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;
673 674 675
			} else if (i == RTAX_FEATURES - 1) {
				u32 user_features = metrics[i] & RTAX_FEATURE_MASK;

676 677
				if (!user_features)
					continue;
678 679 680
				BUILD_BUG_ON(RTAX_FEATURE_MASK & DST_FEATURE_MASK);
				if (nla_put_u32(skb, i + 1, user_features))
					goto nla_put_failure;
681 682 683 684
			} else {
				if (nla_put_u32(skb, i + 1, metrics[i]))
					goto nla_put_failure;
			}
685 686
			valid++;
		}
L
Linus Torvalds 已提交
687 688
	}

689 690 691 692
	if (!valid) {
		nla_nest_cancel(skb, mx);
		return 0;
	}
693 694 695 696

	return nla_nest_end(skb, mx);

nla_put_failure:
697 698
	nla_nest_cancel(skb, mx);
	return -EMSGSIZE;
L
Linus Torvalds 已提交
699
}
E
Eric Dumazet 已提交
700
EXPORT_SYMBOL(rtnetlink_put_metrics);
L
Linus Torvalds 已提交
701

702
int rtnl_put_cacheinfo(struct sk_buff *skb, struct dst_entry *dst, u32 id,
703
		       long expires, u32 error)
704 705
{
	struct rta_cacheinfo ci = {
706
		.rta_lastuse = jiffies_delta_to_clock_t(jiffies - dst->lastuse),
707 708 709 710 711 712
		.rta_used = dst->__use,
		.rta_clntref = atomic_read(&(dst->__refcnt)),
		.rta_error = error,
		.rta_id =  id,
	};

713 714
	if (expires) {
		unsigned long clock;
715

716 717 718 719
		clock = jiffies_to_clock_t(abs(expires));
		clock = min_t(unsigned long, clock, INT_MAX);
		ci.rta_expires = (expires > 0) ? clock : -clock;
	}
720 721 722
	return nla_put(skb, RTA_CACHEINFO, sizeof(ci), &ci);
}
EXPORT_SYMBOL_GPL(rtnl_put_cacheinfo);
L
Linus Torvalds 已提交
723

724
static void set_operstate(struct net_device *dev, unsigned char transition)
S
Stefan Rompf 已提交
725 726 727
{
	unsigned char operstate = dev->operstate;

E
Eric Dumazet 已提交
728
	switch (transition) {
S
Stefan Rompf 已提交
729 730 731 732 733 734 735 736 737 738 739 740
	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;
741
	}
S
Stefan Rompf 已提交
742 743 744 745 746

	if (dev->operstate != operstate) {
		write_lock_bh(&dev_base_lock);
		dev->operstate = operstate;
		write_unlock_bh(&dev_base_lock);
747 748
		netdev_state_change(dev);
	}
S
Stefan Rompf 已提交
749 750
}

751 752 753 754 755 756
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));
}

757 758 759 760 761 762 763 764
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) |
765
			(rtnl_dev_get_flags(dev) & ~ifm->ifi_change);
766 767 768 769

	return flags;
}

770
static void copy_rtnl_link_stats(struct rtnl_link_stats *a,
771
				 const struct rtnl_link_stats64 *b)
L
Linus Torvalds 已提交
772
{
773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799
	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;
800 801

	a->rx_nohandler = b->rx_nohandler;
802 803
}

804
/* All VF info */
805 806
static inline int rtnl_vfinfo_size(const struct net_device *dev,
				   u32 ext_filter_mask)
807
{
808
	if (dev->dev.parent && (ext_filter_mask & RTEXT_FILTER_VF)) {
809
		int num_vfs = dev_num_vf(dev->dev.parent);
S
Sabrina Dubroca 已提交
810
		size_t size = nla_total_size(0);
811
		size += num_vfs *
S
Sabrina Dubroca 已提交
812 813 814 815
			(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 */
816 817
			 nla_total_size(MAX_VLAN_LIST_LEN *
					sizeof(struct ifla_vf_vlan_info)) +
818
			 nla_total_size(sizeof(struct ifla_vf_spoofchk)) +
S
Sabrina Dubroca 已提交
819
			 nla_total_size(sizeof(struct ifla_vf_tx_rate)) +
J
Jiri Benc 已提交
820
			 nla_total_size(sizeof(struct ifla_vf_rate)) +
821
			 nla_total_size(sizeof(struct ifla_vf_link_state)) +
822
			 nla_total_size(sizeof(struct ifla_vf_rss_query_en)) +
S
Sabrina Dubroca 已提交
823
			 nla_total_size(0) + /* nest IFLA_VF_STATS */
824
			 /* IFLA_VF_STATS_RX_PACKETS */
N
Nicolas Dichtel 已提交
825
			 nla_total_size_64bit(sizeof(__u64)) +
826
			 /* IFLA_VF_STATS_TX_PACKETS */
N
Nicolas Dichtel 已提交
827
			 nla_total_size_64bit(sizeof(__u64)) +
828
			 /* IFLA_VF_STATS_RX_BYTES */
N
Nicolas Dichtel 已提交
829
			 nla_total_size_64bit(sizeof(__u64)) +
830
			 /* IFLA_VF_STATS_TX_BYTES */
N
Nicolas Dichtel 已提交
831
			 nla_total_size_64bit(sizeof(__u64)) +
832
			 /* IFLA_VF_STATS_BROADCAST */
N
Nicolas Dichtel 已提交
833
			 nla_total_size_64bit(sizeof(__u64)) +
834
			 /* IFLA_VF_STATS_MULTICAST */
N
Nicolas Dichtel 已提交
835
			 nla_total_size_64bit(sizeof(__u64)) +
H
Hiroshi Shimamoto 已提交
836
			 nla_total_size(sizeof(struct ifla_vf_trust)));
837 838
		return size;
	} else
839 840 841
		return 0;
}

842 843
static size_t rtnl_port_size(const struct net_device *dev,
			     u32 ext_filter_mask)
844 845 846 847 848 849 850 851 852 853 854 855 856
{
	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;

857 858
	if (!dev->netdev_ops->ndo_get_vf_port || !dev->dev.parent ||
	    !(ext_filter_mask & RTEXT_FILTER_VF))
859 860 861 862 863 864 865 866
		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
867
static size_t rtnl_xdp_size(void)
868
{
869
	size_t xdp_size = nla_total_size(0) +	/* nest IFLA_XDP */
M
Martin KaFai Lau 已提交
870 871
			  nla_total_size(1) +	/* XDP_ATTACHED */
			  nla_total_size(4);	/* XDP_PROG_ID */
872

D
David S. Miller 已提交
873
	return xdp_size;
874 875
}

876 877
static noinline size_t if_nlmsg_size(const struct net_device *dev,
				     u32 ext_filter_mask)
878 879 880
{
	return NLMSG_ALIGN(sizeof(struct ifinfomsg))
	       + nla_total_size(IFNAMSIZ) /* IFLA_IFNAME */
881
	       + nla_total_size(IFALIASZ) /* IFLA_IFALIAS */
882
	       + nla_total_size(IFNAMSIZ) /* IFLA_QDISC */
883
	       + nla_total_size_64bit(sizeof(struct rtnl_link_ifmap))
884
	       + nla_total_size(sizeof(struct rtnl_link_stats))
885
	       + nla_total_size_64bit(sizeof(struct rtnl_link_stats64))
886 887 888 889 890 891 892
	       + 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 */
893
	       + nla_total_size(1) /* IFLA_CARRIER */
894
	       + nla_total_size(4) /* IFLA_PROMISCUITY */
895 896
	       + nla_total_size(4) /* IFLA_NUM_TX_QUEUES */
	       + nla_total_size(4) /* IFLA_NUM_RX_QUEUES */
897 898
	       + nla_total_size(4) /* IFLA_GSO_MAX_SEGS */
	       + nla_total_size(4) /* IFLA_GSO_MAX_SIZE */
899
	       + nla_total_size(1) /* IFLA_OPERSTATE */
P
Patrick McHardy 已提交
900
	       + nla_total_size(1) /* IFLA_LINKMODE */
901
	       + nla_total_size(4) /* IFLA_CARRIER_CHANGES */
902
	       + nla_total_size(4) /* IFLA_LINK_NETNSID */
903
	       + nla_total_size(4) /* IFLA_GROUP */
904 905 906
	       + nla_total_size(ext_filter_mask
			        & RTEXT_FILTER_VF ? 4 : 0) /* IFLA_NUM_VF */
	       + rtnl_vfinfo_size(dev, ext_filter_mask) /* IFLA_VFINFO_LIST */
907
	       + rtnl_port_size(dev, ext_filter_mask) /* IFLA_VF_PORTS + IFLA_PORT_SELF */
T
Thomas Graf 已提交
908
	       + rtnl_link_get_size(dev) /* IFLA_LINKINFO */
909
	       + rtnl_link_get_af_size(dev, ext_filter_mask) /* IFLA_AF_SPEC */
910
	       + nla_total_size(MAX_PHYS_ITEM_ID_LEN) /* IFLA_PHYS_PORT_ID */
911
	       + nla_total_size(MAX_PHYS_ITEM_ID_LEN) /* IFLA_PHYS_SWITCH_ID */
912
	       + nla_total_size(IFNAMSIZ) /* IFLA_PHYS_PORT_NAME */
D
David S. Miller 已提交
913
	       + rtnl_xdp_size() /* IFLA_XDP */
914
	       + nla_total_size(4)  /* IFLA_EVENT */
915 916
	       + nla_total_size(1); /* IFLA_PROTO_DOWN */

917 918
}

919 920 921 922 923 924 925 926 927 928 929 930 931
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);
932 933
		if (!vf_port)
			goto nla_put_failure;
934 935
		if (nla_put_u32(skb, IFLA_PORT_VF, vf))
			goto nla_put_failure;
936
		err = dev->netdev_ops->ndo_get_vf_port(dev, vf, skb);
937 938
		if (err == -EMSGSIZE)
			goto nla_put_failure;
939 940 941 942 943 944 945 946 947 948
		if (err) {
			nla_nest_cancel(skb, vf_port);
			continue;
		}
		nla_nest_end(skb, vf_port);
	}

	nla_nest_end(skb, vf_ports);

	return 0;
949 950 951 952

nla_put_failure:
	nla_nest_cancel(skb, vf_ports);
	return -EMSGSIZE;
953 954 955 956 957 958 959 960 961 962 963 964 965 966
}

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);
967
		return (err == -EMSGSIZE) ? err : 0;
968 969 970 971 972 973 974
	}

	nla_nest_end(skb, port_self);

	return 0;
}

975 976
static int rtnl_port_fill(struct sk_buff *skb, struct net_device *dev,
			  u32 ext_filter_mask)
977 978 979
{
	int err;

980 981
	if (!dev->netdev_ops->ndo_get_vf_port || !dev->dev.parent ||
	    !(ext_filter_mask & RTEXT_FILTER_VF))
982 983 984 985 986 987 988 989 990 991 992 993 994 995 996
		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;
}

997 998 999
static int rtnl_phys_port_id_fill(struct sk_buff *skb, struct net_device *dev)
{
	int err;
1000
	struct netdev_phys_item_id ppid;
1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014

	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;
}

1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026
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;
	}

1027
	if (nla_put_string(skb, IFLA_PHYS_PORT_NAME, name))
1028 1029 1030 1031 1032
		return -EMSGSIZE;

	return 0;
}

1033 1034 1035
static int rtnl_phys_switch_id_fill(struct sk_buff *skb, struct net_device *dev)
{
	int err;
1036
	struct switchdev_attr attr = {
1037
		.orig_dev = dev,
1038
		.id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID,
1039 1040
		.flags = SWITCHDEV_F_NO_RECURSE,
	};
1041

1042
	err = switchdev_port_attr_get(dev, &attr);
1043 1044 1045 1046 1047 1048
	if (err) {
		if (err == -EOPNOTSUPP)
			return 0;
		return err;
	}

1049 1050
	if (nla_put(skb, IFLA_PHYS_SWITCH_ID, attr.u.ppid.id_len,
		    attr.u.ppid.id))
1051 1052 1053 1054 1055
		return -EMSGSIZE;

	return 0;
}

1056 1057 1058
static noinline_for_stack int rtnl_fill_stats(struct sk_buff *skb,
					      struct net_device *dev)
{
1059
	struct rtnl_link_stats64 *sp;
1060
	struct nlattr *attr;
1061

1062 1063
	attr = nla_reserve_64bit(skb, IFLA_STATS64,
				 sizeof(struct rtnl_link_stats64), IFLA_PAD);
1064 1065 1066
	if (!attr)
		return -EMSGSIZE;

1067 1068
	sp = nla_data(attr);
	dev_get_stats(dev, sp);
1069

1070 1071
	attr = nla_reserve(skb, IFLA_STATS,
			   sizeof(struct rtnl_link_stats));
1072 1073 1074
	if (!attr)
		return -EMSGSIZE;

1075
	copy_rtnl_link_stats(nla_data(attr), sp);
1076 1077 1078 1079 1080 1081 1082 1083 1084 1085

	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;
1086
	struct nlattr *vf, *vfstats, *vfvlanlist;
1087
	struct ifla_vf_link_state vf_linkstate;
1088
	struct ifla_vf_vlan_info vf_vlan_info;
1089 1090 1091 1092 1093 1094 1095 1096 1097
	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;

1098 1099
	memset(&ivi, 0, sizeof(ivi));

1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111
	/* 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;
1112 1113
	/* VLAN Protocol by default is 802.1Q */
	ivi.vlan_proto = htons(ETH_P_8021Q);
1114 1115 1116
	if (dev->netdev_ops->ndo_get_vf_config(dev, vfs_num, &ivi))
		return 0;

1117 1118
	memset(&vf_vlan_info, 0, sizeof(vf_vlan_info));

1119 1120
	vf_mac.vf =
		vf_vlan.vf =
1121
		vf_vlan_info.vf =
1122 1123 1124 1125 1126 1127 1128 1129 1130 1131
		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;
1132 1133 1134
	vf_vlan_info.vlan = ivi.vlan;
	vf_vlan_info.qos = ivi.qos;
	vf_vlan_info.vlan_proto = ivi.vlan_proto;
1135 1136 1137 1138 1139 1140 1141 1142
	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);
1143 1144
	if (!vf)
		goto nla_put_vfinfo_failure;
1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159
	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))
1160 1161 1162 1163 1164 1165 1166 1167 1168 1169
		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);
1170 1171 1172 1173 1174
	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);
1175 1176
	if (!vfstats)
		goto nla_put_vf_failure;
N
Nicolas Dichtel 已提交
1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187
	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,
1188 1189 1190 1191
			      vf_stats.multicast, IFLA_VF_STATS_PAD)) {
		nla_nest_cancel(skb, vfstats);
		goto nla_put_vf_failure;
	}
1192 1193 1194
	nla_nest_end(skb, vfstats);
	nla_nest_end(skb, vf);
	return 0;
1195 1196 1197 1198 1199 1200

nla_put_vf_failure:
	nla_nest_cancel(skb, vf);
nla_put_vfinfo_failure:
	nla_nest_cancel(skb, vfinfo);
	return -EMSGSIZE;
1201 1202 1203 1204
}

static int rtnl_fill_link_ifmap(struct sk_buff *skb, struct net_device *dev)
{
K
Kangjie Lu 已提交
1205 1206 1207 1208 1209 1210 1211 1212 1213 1214
	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;

1215
	if (nla_put_64bit(skb, IFLA_MAP, sizeof(map), &map, IFLA_PAD))
1216 1217 1218 1219 1220
		return -EMSGSIZE;

	return 0;
}

M
Martin KaFai Lau 已提交
1221
static u8 rtnl_xdp_attached_mode(struct net_device *dev, u32 *prog_id)
1222 1223
{
	const struct net_device_ops *ops = dev->netdev_ops;
M
Martin KaFai Lau 已提交
1224
	const struct bpf_prog *generic_xdp_prog;
1225 1226 1227

	ASSERT_RTNL();

M
Martin KaFai Lau 已提交
1228 1229 1230 1231
	*prog_id = 0;
	generic_xdp_prog = rtnl_dereference(dev->xdp_prog);
	if (generic_xdp_prog) {
		*prog_id = generic_xdp_prog->aux->id;
1232
		return XDP_ATTACHED_SKB;
M
Martin KaFai Lau 已提交
1233
	}
1234 1235
	if (!ops->ndo_xdp)
		return XDP_ATTACHED_NONE;
1236

1237
	return __dev_xdp_attached(dev, ops->ndo_xdp, prog_id);
1238 1239
}

1240 1241 1242
static int rtnl_xdp_fill(struct sk_buff *skb, struct net_device *dev)
{
	struct nlattr *xdp;
M
Martin KaFai Lau 已提交
1243
	u32 prog_id;
1244 1245 1246 1247 1248
	int err;

	xdp = nla_nest_start(skb, IFLA_XDP);
	if (!xdp)
		return -EMSGSIZE;
1249 1250

	err = nla_put_u8(skb, IFLA_XDP_ATTACHED,
M
Martin KaFai Lau 已提交
1251
			 rtnl_xdp_attached_mode(dev, &prog_id));
1252 1253 1254
	if (err)
		goto err_cancel;

M
Martin KaFai Lau 已提交
1255 1256 1257 1258 1259 1260
	if (prog_id) {
		err = nla_put_u32(skb, IFLA_XDP_PROG_ID, prog_id);
		if (err)
			goto err_cancel;
	}

1261 1262 1263 1264 1265 1266 1267 1268
	nla_nest_end(skb, xdp);
	return 0;

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

1269 1270 1271 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
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;
}

1299
static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
1300
			    int type, u32 pid, u32 seq, u32 change,
1301 1302
			    unsigned int flags, u32 ext_filter_mask,
			    u32 event)
1303 1304 1305
{
	struct ifinfomsg *ifm;
	struct nlmsghdr *nlh;
1306
	struct nlattr *af_spec;
T
Thomas Graf 已提交
1307
	struct rtnl_af_ops *af_ops;
1308
	struct net_device *upper_dev = netdev_master_upper_dev_get(dev);
L
Linus Torvalds 已提交
1309

1310
	ASSERT_RTNL();
1311 1312
	nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ifm), flags);
	if (nlh == NULL)
1313
		return -EMSGSIZE;
L
Linus Torvalds 已提交
1314

1315 1316 1317 1318 1319 1320 1321 1322
	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;

1323 1324 1325 1326 1327 1328 1329
	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) ||
1330
	    nla_put_u32(skb, IFLA_PROMISCUITY, dev->promiscuity) ||
1331
	    nla_put_u32(skb, IFLA_NUM_TX_QUEUES, dev->num_tx_queues) ||
1332 1333
	    nla_put_u32(skb, IFLA_GSO_MAX_SEGS, dev->gso_max_segs) ||
	    nla_put_u32(skb, IFLA_GSO_MAX_SIZE, dev->gso_max_size) ||
1334
#ifdef CONFIG_RPS
1335
	    nla_put_u32(skb, IFLA_NUM_RX_QUEUES, dev->num_rx_queues) ||
1336
#endif
1337 1338
	    (dev->ifindex != dev_get_iflink(dev) &&
	     nla_put_u32(skb, IFLA_LINK, dev_get_iflink(dev))) ||
1339 1340
	    (upper_dev &&
	     nla_put_u32(skb, IFLA_MASTER, upper_dev->ifindex)) ||
1341
	    nla_put_u8(skb, IFLA_CARRIER, netif_carrier_ok(dev)) ||
1342 1343 1344
	    (dev->qdisc &&
	     nla_put_string(skb, IFLA_QDISC, dev->qdisc->ops->id)) ||
	    (dev->ifalias &&
1345 1346
	     nla_put_string(skb, IFLA_IFALIAS, dev->ifalias)) ||
	    nla_put_u32(skb, IFLA_CARRIER_CHANGES,
1347 1348
			atomic_read(&dev->carrier_changes)) ||
	    nla_put_u8(skb, IFLA_PROTO_DOWN, dev->proto_down))
1349
		goto nla_put_failure;
1350

1351 1352 1353 1354 1355
	if (event != IFLA_EVENT_NONE) {
		if (nla_put_u32(skb, IFLA_EVENT, event))
			goto nla_put_failure;
	}

1356 1357
	if (rtnl_fill_link_ifmap(skb, dev))
		goto nla_put_failure;
L
Linus Torvalds 已提交
1358 1359

	if (dev->addr_len) {
1360 1361 1362
		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
1363 1364
	}

1365 1366 1367
	if (rtnl_phys_port_id_fill(skb, dev))
		goto nla_put_failure;

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

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

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

1377 1378 1379
	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;
1380

1381 1382
	if (dev->netdev_ops->ndo_get_vf_config && dev->dev.parent &&
	    ext_filter_mask & RTEXT_FILTER_VF) {
1383
		int i;
1384
		struct nlattr *vfinfo;
1385 1386 1387 1388 1389 1390
		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++) {
1391
			if (rtnl_fill_vfinfo(skb, dev, i, vfinfo))
1392
				goto nla_put_failure;
1393
		}
1394

1395
		nla_nest_end(skb, vfinfo);
1396
	}
1397

1398
	if (rtnl_port_fill(skb, dev, ext_filter_mask))
1399 1400
		goto nla_put_failure;

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

1404
	if (dev->rtnl_link_ops || rtnl_have_link_slave_info(dev)) {
P
Patrick McHardy 已提交
1405 1406 1407 1408
		if (rtnl_link_fill(skb, dev) < 0)
			goto nla_put_failure;
	}

1409 1410 1411 1412 1413
	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)) {
1414
			int id = peernet2id_alloc(dev_net(dev), link_net);
1415 1416 1417 1418 1419 1420

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

T
Thomas Graf 已提交
1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431
	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;

1432
			err = af_ops->fill_link_af(skb, dev, ext_filter_mask);
T
Thomas Graf 已提交
1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450

			/*
			 * 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);

1451 1452
	nlmsg_end(skb, nlh);
	return 0;
1453 1454

nla_put_failure:
1455 1456
	nlmsg_cancel(skb, nlh);
	return -EMSGSIZE;
L
Linus Torvalds 已提交
1457 1458
}

J
Jiri Pirko 已提交
1459
static const struct nla_policy ifla_policy[IFLA_MAX+1] = {
1460
	[IFLA_IFNAME]		= { .type = NLA_STRING, .len = IFNAMSIZ-1 },
P
Patrick McHardy 已提交
1461 1462
	[IFLA_ADDRESS]		= { .type = NLA_BINARY, .len = MAX_ADDR_LEN },
	[IFLA_BROADCAST]	= { .type = NLA_BINARY, .len = MAX_ADDR_LEN },
1463
	[IFLA_MAP]		= { .len = sizeof(struct rtnl_link_ifmap) },
1464
	[IFLA_MTU]		= { .type = NLA_U32 },
1465
	[IFLA_LINK]		= { .type = NLA_U32 },
1466
	[IFLA_MASTER]		= { .type = NLA_U32 },
1467
	[IFLA_CARRIER]		= { .type = NLA_U8 },
1468 1469 1470 1471
	[IFLA_TXQLEN]		= { .type = NLA_U32 },
	[IFLA_WEIGHT]		= { .type = NLA_U32 },
	[IFLA_OPERSTATE]	= { .type = NLA_U8 },
	[IFLA_LINKMODE]		= { .type = NLA_U8 },
1472
	[IFLA_LINKINFO]		= { .type = NLA_NESTED },
1473
	[IFLA_NET_NS_PID]	= { .type = NLA_U32 },
1474
	[IFLA_NET_NS_FD]	= { .type = NLA_U32 },
1475
	[IFLA_IFALIAS]	        = { .type = NLA_STRING, .len = IFALIASZ-1 },
1476
	[IFLA_VFINFO_LIST]	= {. type = NLA_NESTED },
1477 1478
	[IFLA_VF_PORTS]		= { .type = NLA_NESTED },
	[IFLA_PORT_SELF]	= { .type = NLA_NESTED },
T
Thomas Graf 已提交
1479
	[IFLA_AF_SPEC]		= { .type = NLA_NESTED },
1480
	[IFLA_EXT_MASK]		= { .type = NLA_U32 },
1481
	[IFLA_PROMISCUITY]	= { .type = NLA_U32 },
1482 1483
	[IFLA_NUM_TX_QUEUES]	= { .type = NLA_U32 },
	[IFLA_NUM_RX_QUEUES]	= { .type = NLA_U32 },
1484
	[IFLA_PHYS_PORT_ID]	= { .type = NLA_BINARY, .len = MAX_PHYS_ITEM_ID_LEN },
1485
	[IFLA_CARRIER_CHANGES]	= { .type = NLA_U32 },  /* ignored */
1486
	[IFLA_PHYS_SWITCH_ID]	= { .type = NLA_BINARY, .len = MAX_PHYS_ITEM_ID_LEN },
1487
	[IFLA_LINK_NETNSID]	= { .type = NLA_S32 },
1488
	[IFLA_PROTO_DOWN]	= { .type = NLA_U8 },
1489
	[IFLA_XDP]		= { .type = NLA_NESTED },
1490
	[IFLA_EVENT]		= { .type = NLA_U32 },
1491
	[IFLA_GROUP]		= { .type = NLA_U32 },
1492 1493
};

P
Patrick McHardy 已提交
1494 1495 1496
static const struct nla_policy ifla_info_policy[IFLA_INFO_MAX+1] = {
	[IFLA_INFO_KIND]	= { .type = NLA_STRING },
	[IFLA_INFO_DATA]	= { .type = NLA_NESTED },
1497 1498
	[IFLA_INFO_SLAVE_KIND]	= { .type = NLA_STRING },
	[IFLA_INFO_SLAVE_DATA]	= { .type = NLA_NESTED },
P
Patrick McHardy 已提交
1499 1500
};

1501
static const struct nla_policy ifla_vf_policy[IFLA_VF_MAX+1] = {
1502 1503
	[IFLA_VF_MAC]		= { .len = sizeof(struct ifla_vf_mac) },
	[IFLA_VF_VLAN]		= { .len = sizeof(struct ifla_vf_vlan) },
1504
	[IFLA_VF_VLAN_LIST]     = { .type = NLA_NESTED },
1505 1506 1507 1508
	[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) },
1509
	[IFLA_VF_RSS_QUERY_EN]	= { .len = sizeof(struct ifla_vf_rss_query_en) },
1510
	[IFLA_VF_STATS]		= { .type = NLA_NESTED },
H
Hiroshi Shimamoto 已提交
1511
	[IFLA_VF_TRUST]		= { .len = sizeof(struct ifla_vf_trust) },
1512 1513
	[IFLA_VF_IB_NODE_GUID]	= { .len = sizeof(struct ifla_vf_guid) },
	[IFLA_VF_IB_PORT_GUID]	= { .len = sizeof(struct ifla_vf_guid) },
1514 1515
};

1516 1517 1518 1519 1520 1521 1522 1523 1524 1525
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, },
1526 1527 1528 1529 1530 1531 1532

	/* 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) },
1533 1534
};

1535 1536 1537
static const struct nla_policy ifla_xdp_policy[IFLA_XDP_MAX + 1] = {
	[IFLA_XDP_FD]		= { .type = NLA_S32 },
	[IFLA_XDP_ATTACHED]	= { .type = NLA_U8 },
1538
	[IFLA_XDP_FLAGS]	= { .type = NLA_U32 },
M
Martin KaFai Lau 已提交
1539
	[IFLA_XDP_PROG_ID]	= { .type = NLA_U32 },
1540 1541
};

1542 1543 1544 1545 1546
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];

1547 1548
	if (nla_parse_nested(linfo, IFLA_INFO_MAX, nla,
			     ifla_info_policy, NULL) < 0)
1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594
		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
1595 1596 1597 1598 1599 1600 1601 1602 1603
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;
1604 1605 1606
	const struct rtnl_link_ops *kind_ops = NULL;
	unsigned int flags = NLM_F_MULTI;
	int master_idx = 0;
1607
	int err;
1608
	int hdrlen;
J
Jiri Pirko 已提交
1609 1610 1611 1612 1613 1614

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

	cb->seq = net->dev_base_seq;

1615 1616 1617 1618 1619 1620 1621 1622 1623 1624
	/* 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);

1625 1626
	if (nlmsg_parse(cb->nlh, hdrlen, tb, IFLA_MAX,
			ifla_policy, NULL) >= 0) {
J
Jiri Pirko 已提交
1627 1628
		if (tb[IFLA_EXT_MASK])
			ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);
1629 1630 1631 1632 1633 1634 1635 1636 1637

		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
1638 1639 1640 1641 1642
	}

	for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) {
		idx = 0;
		head = &net->dev_index_head[h];
1643
		hlist_for_each_entry(dev, head, index_hlist) {
1644
			if (link_dump_filtered(dev, master_idx, kind_ops))
1645
				goto cont;
J
Jiri Pirko 已提交
1646 1647
			if (idx < s_idx)
				goto cont;
1648 1649 1650
			err = rtnl_fill_ifinfo(skb, dev, RTM_NEWLINK,
					       NETLINK_CB(cb->skb).portid,
					       cb->nlh->nlmsg_seq, 0,
1651
					       flags,
1652
					       ext_filter_mask, 0);
1653

1654 1655 1656 1657 1658 1659
			if (err < 0) {
				if (likely(skb->len))
					goto out;

				goto out_err;
			}
J
Jiri Pirko 已提交
1660 1661 1662 1663 1664 1665 1666

			nl_dump_check_consistent(cb, nlmsg_hdr(skb));
cont:
			idx++;
		}
	}
out:
1667 1668
	err = skb->len;
out_err:
J
Jiri Pirko 已提交
1669 1670 1671
	cb->args[1] = idx;
	cb->args[0] = h;

1672
	return err;
J
Jiri Pirko 已提交
1673 1674
}

1675 1676
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
1677
{
1678
	return nla_parse(tb, IFLA_MAX, head, len, ifla_policy, exterr);
J
Jiri Pirko 已提交
1679 1680 1681
}
EXPORT_SYMBOL(rtnl_nla_parse_ifla);

1682 1683 1684 1685 1686 1687 1688 1689
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]));
1690 1691
	else if (tb[IFLA_NET_NS_FD])
		net = get_net_ns_by_fd(nla_get_u32(tb[IFLA_NET_NS_FD]));
1692 1693 1694 1695 1696 1697
	else
		net = get_net(src_net);
	return net;
}
EXPORT_SYMBOL(rtnl_link_get_net);

1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709
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;
	}

1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723
	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) {
1724
				err = af_ops->validate_link_af(dev, af);
1725 1726 1727 1728 1729 1730
				if (err < 0)
					return err;
			}
		}
	}

1731 1732 1733
	return 0;
}

1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749
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);
}

1750
static int do_setvfinfo(struct net_device *dev, struct nlattr **tb)
1751 1752
{
	const struct net_device_ops *ops = dev->netdev_ops;
1753
	int err = -EINVAL;
1754

1755 1756
	if (tb[IFLA_VF_MAC]) {
		struct ifla_vf_mac *ivm = nla_data(tb[IFLA_VF_MAC]);
1757

1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771
		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,
1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797
						   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++;
		}
1798 1799 1800
		if (len == 0)
			return -EINVAL;

1801 1802
		err = ops->ndo_set_vf_vlan(dev, ivvl[0]->vf, ivvl[0]->vlan,
					   ivvl[0]->qos, ivvl[0]->vlan_proto);
1803 1804
		if (err < 0)
			return err;
1805
	}
1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 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

	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
1872 1873 1874 1875 1876 1877 1878 1879 1880 1881
	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;
	}

1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899
	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);
	}

1900 1901 1902
	return err;
}

1903 1904
static int do_set_master(struct net_device *dev, int ifindex)
{
1905
	struct net_device *upper_dev = netdev_master_upper_dev_get(dev);
1906 1907 1908
	const struct net_device_ops *ops;
	int err;

1909 1910
	if (upper_dev) {
		if (upper_dev->ifindex == ifindex)
1911
			return 0;
1912
		ops = upper_dev->netdev_ops;
1913
		if (ops->ndo_del_slave) {
1914
			err = ops->ndo_del_slave(upper_dev, dev);
1915 1916 1917 1918 1919 1920 1921 1922
			if (err)
				return err;
		} else {
			return -EOPNOTSUPP;
		}
	}

	if (ifindex) {
1923 1924
		upper_dev = __dev_get_by_index(dev_net(dev), ifindex);
		if (!upper_dev)
1925
			return -EINVAL;
1926
		ops = upper_dev->netdev_ops;
1927
		if (ops->ndo_add_slave) {
1928
			err = ops->ndo_add_slave(upper_dev, dev);
1929 1930 1931 1932 1933 1934 1935 1936 1937
			if (err)
				return err;
		} else {
			return -EOPNOTSUPP;
		}
	}
	return 0;
}

1938
#define DO_SETLINK_MODIFIED	0x01
1939 1940
/* notify flag means notify + modified. */
#define DO_SETLINK_NOTIFY	0x03
1941 1942
static int do_setlink(const struct sk_buff *skb,
		      struct net_device *dev, struct ifinfomsg *ifm,
1943
		      struct netlink_ext_ack *extack,
1944
		      struct nlattr **tb, char *ifname, int status)
L
Linus Torvalds 已提交
1945
{
1946
	const struct net_device_ops *ops = dev->netdev_ops;
1947
	int err;
L
Linus Torvalds 已提交
1948

1949
	if (tb[IFLA_NET_NS_PID] || tb[IFLA_NET_NS_FD]) {
1950
		struct net *net = rtnl_link_get_net(dev_net(dev), tb);
1951 1952 1953 1954
		if (IS_ERR(net)) {
			err = PTR_ERR(net);
			goto errout;
		}
1955
		if (!netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) {
1956
			put_net(net);
1957 1958 1959
			err = -EPERM;
			goto errout;
		}
1960 1961 1962 1963
		err = dev_change_net_namespace(dev, net, ifname);
		put_net(net);
		if (err)
			goto errout;
1964
		status |= DO_SETLINK_MODIFIED;
1965 1966
	}

1967
	if (tb[IFLA_MAP]) {
L
Linus Torvalds 已提交
1968 1969 1970
		struct rtnl_link_ifmap *u_map;
		struct ifmap k_map;

1971
		if (!ops->ndo_set_config) {
L
Linus Torvalds 已提交
1972
			err = -EOPNOTSUPP;
1973
			goto errout;
L
Linus Torvalds 已提交
1974 1975 1976 1977
		}

		if (!netif_device_present(dev)) {
			err = -ENODEV;
1978
			goto errout;
L
Linus Torvalds 已提交
1979 1980
		}

1981
		u_map = nla_data(tb[IFLA_MAP]);
L
Linus Torvalds 已提交
1982 1983 1984 1985 1986 1987 1988
		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;

1989
		err = ops->ndo_set_config(dev, &k_map);
1990
		if (err < 0)
1991
			goto errout;
L
Linus Torvalds 已提交
1992

1993
		status |= DO_SETLINK_NOTIFY;
L
Linus Torvalds 已提交
1994 1995
	}

1996
	if (tb[IFLA_ADDRESS]) {
1997 1998 1999
		struct sockaddr *sa;
		int len;

2000 2001
		len = sizeof(sa_family_t) + max_t(size_t, dev->addr_len,
						  sizeof(*sa));
2002 2003 2004
		sa = kmalloc(len, GFP_KERNEL);
		if (!sa) {
			err = -ENOMEM;
2005
			goto errout;
2006 2007
		}
		sa->sa_family = dev->type;
2008
		memcpy(sa->sa_data, nla_data(tb[IFLA_ADDRESS]),
2009
		       dev->addr_len);
2010
		err = dev_set_mac_address(dev, sa);
2011
		kfree(sa);
L
Linus Torvalds 已提交
2012
		if (err)
2013
			goto errout;
2014
		status |= DO_SETLINK_MODIFIED;
L
Linus Torvalds 已提交
2015 2016
	}

2017 2018 2019
	if (tb[IFLA_MTU]) {
		err = dev_set_mtu(dev, nla_get_u32(tb[IFLA_MTU]));
		if (err < 0)
2020
			goto errout;
2021
		status |= DO_SETLINK_MODIFIED;
L
Linus Torvalds 已提交
2022 2023
	}

2024 2025
	if (tb[IFLA_GROUP]) {
		dev_set_group(dev, nla_get_u32(tb[IFLA_GROUP]));
2026
		status |= DO_SETLINK_NOTIFY;
2027 2028
	}

2029 2030 2031 2032 2033
	/*
	 * Interface selected by interface index but interface
	 * name provided implies that a name change has been
	 * requested.
	 */
2034
	if (ifm->ifi_index > 0 && ifname[0]) {
2035 2036
		err = dev_change_name(dev, ifname);
		if (err < 0)
2037
			goto errout;
2038
		status |= DO_SETLINK_MODIFIED;
L
Linus Torvalds 已提交
2039 2040
	}

2041 2042 2043 2044 2045
	if (tb[IFLA_IFALIAS]) {
		err = dev_set_alias(dev, nla_data(tb[IFLA_IFALIAS]),
				    nla_len(tb[IFLA_IFALIAS]));
		if (err < 0)
			goto errout;
2046
		status |= DO_SETLINK_NOTIFY;
2047 2048
	}

2049 2050
	if (tb[IFLA_BROADCAST]) {
		nla_memcpy(dev->broadcast, tb[IFLA_BROADCAST], dev->addr_len);
2051
		call_netdevice_notifiers(NETDEV_CHANGEADDR, dev);
L
Linus Torvalds 已提交
2052 2053
	}

2054
	if (ifm->ifi_flags || ifm->ifi_change) {
2055
		err = dev_change_flags(dev, rtnl_dev_combine_flags(dev, ifm));
2056 2057
		if (err < 0)
			goto errout;
2058
	}
L
Linus Torvalds 已提交
2059

2060 2061 2062 2063
	if (tb[IFLA_MASTER]) {
		err = do_set_master(dev, nla_get_u32(tb[IFLA_MASTER]));
		if (err)
			goto errout;
2064
		status |= DO_SETLINK_MODIFIED;
2065 2066
	}

2067 2068 2069 2070
	if (tb[IFLA_CARRIER]) {
		err = dev_change_carrier(dev, nla_get_u8(tb[IFLA_CARRIER]));
		if (err)
			goto errout;
2071
		status |= DO_SETLINK_MODIFIED;
2072 2073
	}

2074
	if (tb[IFLA_TXQLEN]) {
2075 2076
		unsigned int value = nla_get_u32(tb[IFLA_TXQLEN]);
		unsigned int orig_len = dev->tx_queue_len;
2077 2078 2079 2080 2081 2082 2083 2084 2085 2086

		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;
			}
2087
			status |= DO_SETLINK_NOTIFY;
2088
		}
2089
	}
S
Stefan Rompf 已提交
2090

2091
	if (tb[IFLA_OPERSTATE])
2092
		set_operstate(dev, nla_get_u8(tb[IFLA_OPERSTATE]));
S
Stefan Rompf 已提交
2093

2094
	if (tb[IFLA_LINKMODE]) {
2095 2096
		unsigned char value = nla_get_u8(tb[IFLA_LINKMODE]);

2097
		write_lock_bh(&dev_base_lock);
2098
		if (dev->link_mode ^ value)
2099
			status |= DO_SETLINK_NOTIFY;
2100
		dev->link_mode = value;
2101
		write_unlock_bh(&dev_base_lock);
S
Stefan Rompf 已提交
2102 2103
	}

2104
	if (tb[IFLA_VFINFO_LIST]) {
2105
		struct nlattr *vfinfo[IFLA_VF_MAX + 1];
2106 2107
		struct nlattr *attr;
		int rem;
2108

2109
		nla_for_each_nested(attr, tb[IFLA_VFINFO_LIST], rem) {
2110 2111
			if (nla_type(attr) != IFLA_VF_INFO ||
			    nla_len(attr) < NLA_HDRLEN) {
2112
				err = -EINVAL;
2113
				goto errout;
2114
			}
2115
			err = nla_parse_nested(vfinfo, IFLA_VF_MAX, attr,
2116
					       ifla_vf_policy, NULL);
2117 2118 2119
			if (err < 0)
				goto errout;
			err = do_setvfinfo(dev, vfinfo);
2120 2121
			if (err < 0)
				goto errout;
2122
			status |= DO_SETLINK_NOTIFY;
2123
		}
2124
	}
L
Linus Torvalds 已提交
2125 2126
	err = 0;

2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137
	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) {
2138 2139 2140 2141 2142 2143
			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,
2144
					       ifla_port_policy, NULL);
2145 2146 2147 2148 2149 2150 2151 2152 2153 2154
			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;
2155
			status |= DO_SETLINK_NOTIFY;
2156 2157 2158 2159 2160 2161 2162 2163
		}
	}
	err = 0;

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

		err = nla_parse_nested(port, IFLA_PORT_MAX,
2164 2165
				       tb[IFLA_PORT_SELF], ifla_port_policy,
				       NULL);
2166 2167 2168 2169 2170 2171 2172 2173
		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;
2174
		status |= DO_SETLINK_NOTIFY;
2175
	}
T
Thomas Graf 已提交
2176 2177 2178 2179 2180 2181 2182 2183 2184

	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))))
2185
				BUG();
T
Thomas Graf 已提交
2186

2187
			err = af_ops->set_link_af(dev, af);
T
Thomas Graf 已提交
2188 2189 2190
			if (err < 0)
				goto errout;

2191
			status |= DO_SETLINK_NOTIFY;
T
Thomas Graf 已提交
2192 2193
		}
	}
2194 2195
	err = 0;

2196 2197 2198 2199 2200 2201 2202 2203
	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;
	}

2204 2205
	if (tb[IFLA_XDP]) {
		struct nlattr *xdp[IFLA_XDP_MAX + 1];
2206
		u32 xdp_flags = 0;
2207 2208

		err = nla_parse_nested(xdp, IFLA_XDP_MAX, tb[IFLA_XDP],
2209
				       ifla_xdp_policy, NULL);
2210 2211 2212
		if (err < 0)
			goto errout;

M
Martin KaFai Lau 已提交
2213
		if (xdp[IFLA_XDP_ATTACHED] || xdp[IFLA_XDP_PROG_ID]) {
2214 2215 2216
			err = -EINVAL;
			goto errout;
		}
2217 2218 2219 2220 2221 2222 2223

		if (xdp[IFLA_XDP_FLAGS]) {
			xdp_flags = nla_get_u32(xdp[IFLA_XDP_FLAGS]);
			if (xdp_flags & ~XDP_FLAGS_MASK) {
				err = -EINVAL;
				goto errout;
			}
2224
			if (hweight32(xdp_flags & XDP_FLAGS_MODES) > 1) {
2225 2226 2227
				err = -EINVAL;
				goto errout;
			}
2228 2229
		}

2230
		if (xdp[IFLA_XDP_FD]) {
2231
			err = dev_change_xdp_fd(dev, extack,
2232 2233
						nla_get_s32(xdp[IFLA_XDP_FD]),
						xdp_flags);
2234 2235 2236 2237 2238 2239
			if (err)
				goto errout;
			status |= DO_SETLINK_NOTIFY;
		}
	}

2240
errout:
2241 2242 2243 2244 2245 2246 2247 2248
	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);
	}
2249

2250 2251
	return err;
}
L
Linus Torvalds 已提交
2252

2253 2254
static int rtnl_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
2255
{
2256
	struct net *net = sock_net(skb->sk);
2257 2258 2259 2260 2261 2262
	struct ifinfomsg *ifm;
	struct net_device *dev;
	int err;
	struct nlattr *tb[IFLA_MAX+1];
	char ifname[IFNAMSIZ];

2263 2264
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy,
			  extack);
2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275
	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)
2276
		dev = __dev_get_by_index(net, ifm->ifi_index);
2277
	else if (tb[IFLA_IFNAME])
2278
		dev = __dev_get_by_name(net, ifname);
2279 2280 2281 2282 2283 2284 2285 2286
	else
		goto errout;

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

E
Eric Dumazet 已提交
2287 2288
	err = validate_linkmsg(dev, tb);
	if (err < 0)
2289
		goto errout;
2290

2291
	err = do_setlink(skb, dev, ifm, extack, tb, ifname, 0);
2292
errout:
L
Linus Torvalds 已提交
2293 2294 2295
	return err;
}

2296 2297 2298 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
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;
}

2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347
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);

2348 2349
static int rtnl_dellink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
P
Patrick McHardy 已提交
2350
{
2351
	struct net *net = sock_net(skb->sk);
P
Patrick McHardy 已提交
2352 2353 2354 2355 2356 2357
	struct net_device *dev;
	struct ifinfomsg *ifm;
	char ifname[IFNAMSIZ];
	struct nlattr *tb[IFLA_MAX+1];
	int err;

2358
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy, extack);
P
Patrick McHardy 已提交
2359 2360 2361 2362 2363 2364 2365 2366
	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)
2367
		dev = __dev_get_by_index(net, ifm->ifi_index);
P
Patrick McHardy 已提交
2368
	else if (tb[IFLA_IFNAME])
2369
		dev = __dev_get_by_name(net, ifname);
2370 2371
	else if (tb[IFLA_GROUP])
		return rtnl_group_dellink(net, nla_get_u32(tb[IFLA_GROUP]));
P
Patrick McHardy 已提交
2372 2373 2374 2375 2376 2377
	else
		return -EINVAL;

	if (!dev)
		return -ENODEV;

2378
	return rtnl_delete_link(dev);
P
Patrick McHardy 已提交
2379 2380
}

2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394
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;

2395
	__dev_notify_flags(dev, old_flags, ~0U);
2396 2397 2398 2399
	return 0;
}
EXPORT_SYMBOL(rtnl_configure_link);

2400
struct net_device *rtnl_create_link(struct net *net,
2401
	const char *ifname, unsigned char name_assign_type,
2402
	const struct rtnl_link_ops *ops, struct nlattr *tb[])
2403 2404
{
	struct net_device *dev;
2405 2406
	unsigned int num_tx_queues = 1;
	unsigned int num_rx_queues = 1;
2407

2408 2409 2410
	if (tb[IFLA_NUM_TX_QUEUES])
		num_tx_queues = nla_get_u32(tb[IFLA_NUM_TX_QUEUES]);
	else if (ops->get_num_tx_queues)
2411
		num_tx_queues = ops->get_num_tx_queues();
2412 2413 2414 2415

	if (tb[IFLA_NUM_RX_QUEUES])
		num_rx_queues = nla_get_u32(tb[IFLA_NUM_RX_QUEUES]);
	else if (ops->get_num_rx_queues)
2416
		num_rx_queues = ops->get_num_rx_queues();
2417

2418
	dev = alloc_netdev_mqs(ops->priv_size, ifname, name_assign_type,
2419
			       ops->setup, num_tx_queues, num_rx_queues);
2420
	if (!dev)
2421
		return ERR_PTR(-ENOMEM);
2422

2423 2424
	dev_net_set(dev, net);
	dev->rtnl_link_ops = ops;
2425
	dev->rtnl_link_state = RTNL_LINK_INITIALIZING;
2426

2427 2428
	if (tb[IFLA_MTU])
		dev->mtu = nla_get_u32(tb[IFLA_MTU]);
2429
	if (tb[IFLA_ADDRESS]) {
2430 2431
		memcpy(dev->dev_addr, nla_data(tb[IFLA_ADDRESS]),
				nla_len(tb[IFLA_ADDRESS]));
2432 2433
		dev->addr_assign_type = NET_ADDR_SET;
	}
2434 2435 2436 2437 2438 2439
	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])
2440
		set_operstate(dev, nla_get_u8(tb[IFLA_OPERSTATE]));
2441 2442
	if (tb[IFLA_LINKMODE])
		dev->link_mode = nla_get_u8(tb[IFLA_LINKMODE]);
2443 2444
	if (tb[IFLA_GROUP])
		dev_set_group(dev, nla_get_u32(tb[IFLA_GROUP]));
2445 2446 2447

	return dev;
}
E
Eric Dumazet 已提交
2448
EXPORT_SYMBOL(rtnl_create_link);
2449

2450 2451
static int rtnl_group_changelink(const struct sk_buff *skb,
		struct net *net, int group,
2452
		struct ifinfomsg *ifm,
2453
		struct netlink_ext_ack *extack,
2454 2455
		struct nlattr **tb)
{
2456
	struct net_device *dev, *aux;
2457 2458
	int err;

2459
	for_each_netdev_safe(net, dev, aux) {
2460
		if (dev->group == group) {
2461
			err = do_setlink(skb, dev, ifm, extack, tb, NULL, 0);
2462 2463 2464 2465 2466 2467 2468 2469
			if (err < 0)
				return err;
		}
	}

	return 0;
}

2470 2471
static int rtnl_newlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
P
Patrick McHardy 已提交
2472
{
2473
	struct net *net = sock_net(skb->sk);
P
Patrick McHardy 已提交
2474
	const struct rtnl_link_ops *ops;
2475
	const struct rtnl_link_ops *m_ops = NULL;
P
Patrick McHardy 已提交
2476
	struct net_device *dev;
2477
	struct net_device *master_dev = NULL;
P
Patrick McHardy 已提交
2478 2479 2480 2481 2482
	struct ifinfomsg *ifm;
	char kind[MODULE_NAME_LEN];
	char ifname[IFNAMSIZ];
	struct nlattr *tb[IFLA_MAX+1];
	struct nlattr *linkinfo[IFLA_INFO_MAX+1];
2483
	unsigned char name_assign_type = NET_NAME_USER;
P
Patrick McHardy 已提交
2484 2485
	int err;

2486
#ifdef CONFIG_MODULES
P
Patrick McHardy 已提交
2487
replay:
2488
#endif
2489
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy, extack);
P
Patrick McHardy 已提交
2490 2491 2492 2493 2494 2495 2496 2497 2498 2499
	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)
2500
		dev = __dev_get_by_index(net, ifm->ifi_index);
2501 2502 2503 2504 2505 2506
	else {
		if (ifname[0])
			dev = __dev_get_by_name(net, ifname);
		else
			dev = NULL;
	}
P
Patrick McHardy 已提交
2507

2508 2509 2510 2511 2512 2513
	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
2514 2515
	err = validate_linkmsg(dev, tb);
	if (err < 0)
2516 2517
		return err;

P
Patrick McHardy 已提交
2518 2519
	if (tb[IFLA_LINKINFO]) {
		err = nla_parse_nested(linkinfo, IFLA_INFO_MAX,
2520 2521
				       tb[IFLA_LINKINFO], ifla_info_policy,
				       NULL);
P
Patrick McHardy 已提交
2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535
		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
2536 2537
		struct nlattr *attr[ops ? ops->maxtype + 1 : 1];
		struct nlattr *slave_attr[m_ops ? m_ops->slave_maxtype + 1 : 1];
2538 2539
		struct nlattr **data = NULL;
		struct nlattr **slave_data = NULL;
2540
		struct net *dest_net, *link_net = NULL;
P
Patrick McHardy 已提交
2541 2542 2543 2544 2545

		if (ops) {
			if (ops->maxtype && linkinfo[IFLA_INFO_DATA]) {
				err = nla_parse_nested(attr, ops->maxtype,
						       linkinfo[IFLA_INFO_DATA],
2546
						       ops->policy, NULL);
P
Patrick McHardy 已提交
2547 2548 2549 2550 2551
				if (err < 0)
					return err;
				data = attr;
			}
			if (ops->validate) {
2552
				err = ops->validate(tb, data, extack);
P
Patrick McHardy 已提交
2553 2554 2555 2556 2557
				if (err < 0)
					return err;
			}
		}

2558 2559 2560 2561 2562 2563
		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],
2564 2565
						       m_ops->slave_policy,
						       NULL);
2566 2567 2568 2569 2570
				if (err < 0)
					return err;
				slave_data = slave_attr;
			}
			if (m_ops->slave_validate) {
2571 2572
				err = m_ops->slave_validate(tb, slave_data,
							    extack);
2573 2574 2575 2576 2577
				if (err < 0)
					return err;
			}
		}

P
Patrick McHardy 已提交
2578
		if (dev) {
2579
			int status = 0;
P
Patrick McHardy 已提交
2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590

			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;

2591
				err = ops->changelink(dev, tb, data, extack);
P
Patrick McHardy 已提交
2592 2593
				if (err < 0)
					return err;
2594
				status |= DO_SETLINK_NOTIFY;
P
Patrick McHardy 已提交
2595 2596
			}

2597 2598 2599 2600 2601
			if (linkinfo[IFLA_INFO_SLAVE_DATA]) {
				if (!m_ops || !m_ops->slave_changelink)
					return -EOPNOTSUPP;

				err = m_ops->slave_changelink(master_dev, dev,
2602 2603
							      tb, slave_data,
							      extack);
2604 2605
				if (err < 0)
					return err;
2606
				status |= DO_SETLINK_NOTIFY;
2607 2608
			}

2609 2610
			return do_setlink(skb, dev, ifm, extack, tb, ifname,
					  status);
P
Patrick McHardy 已提交
2611 2612
		}

2613 2614
		if (!(nlh->nlmsg_flags & NLM_F_CREATE)) {
			if (ifm->ifi_index == 0 && tb[IFLA_GROUP])
2615
				return rtnl_group_changelink(skb, net,
2616
						nla_get_u32(tb[IFLA_GROUP]),
2617
						ifm, extack, tb);
P
Patrick McHardy 已提交
2618
			return -ENODEV;
2619
		}
P
Patrick McHardy 已提交
2620

2621
		if (tb[IFLA_MAP] || tb[IFLA_PROTINFO])
P
Patrick McHardy 已提交
2622 2623 2624
			return -EOPNOTSUPP;

		if (!ops) {
2625
#ifdef CONFIG_MODULES
P
Patrick McHardy 已提交
2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637
			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;
		}

2638 2639 2640
		if (!ops->setup)
			return -EOPNOTSUPP;

2641
		if (!ifname[0]) {
P
Patrick McHardy 已提交
2642
			snprintf(ifname, IFNAMSIZ, "%s%%d", ops->kind);
2643 2644
			name_assign_type = NET_NAME_ENUM;
		}
2645

2646
		dest_net = rtnl_link_get_net(net, tb);
2647 2648 2649
		if (IS_ERR(dest_net))
			return PTR_ERR(dest_net);

2650 2651 2652 2653
		err = -EPERM;
		if (!netlink_ns_capable(skb, dest_net->user_ns, CAP_NET_ADMIN))
			goto out;

2654 2655 2656 2657 2658 2659 2660 2661
		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;
			}
2662 2663 2664
			err = -EPERM;
			if (!netlink_ns_capable(skb, link_net->user_ns, CAP_NET_ADMIN))
				goto out;
2665 2666 2667 2668
		}

		dev = rtnl_create_link(link_net ? : dest_net, ifname,
				       name_assign_type, ops, tb);
2669
		if (IS_ERR(dev)) {
2670
			err = PTR_ERR(dev);
2671 2672 2673 2674 2675
			goto out;
		}

		dev->ifindex = ifm->ifi_index;

2676
		if (ops->newlink) {
2677 2678
			err = ops->newlink(link_net ? : net, dev, tb, data,
					   extack);
2679
			/* Drivers should call free_netdev() in ->destructor
2680 2681
			 * and unregister it on failure after registration
			 * so that device could be finally freed in rtnl_unlock.
2682
			 */
2683 2684 2685 2686
			if (err < 0) {
				/* If device is not registered at all, free it now */
				if (dev->reg_state == NETREG_UNINITIALIZED)
					free_netdev(dev);
2687
				goto out;
2688
			}
2689
		} else {
2690
			err = register_netdevice(dev);
2691 2692 2693 2694
			if (err < 0) {
				free_netdev(dev);
				goto out;
			}
2695
		}
2696
		err = rtnl_configure_link(dev, ifm);
2697 2698
		if (err < 0)
			goto out_unregister;
2699
		if (link_net) {
2700
			err = dev_change_net_namespace(dev, dest_net, ifname);
2701
			if (err < 0)
2702
				goto out_unregister;
2703
		}
2704 2705 2706 2707 2708
		if (tb[IFLA_MASTER]) {
			err = do_set_master(dev, nla_get_u32(tb[IFLA_MASTER]));
			if (err)
				goto out_unregister;
		}
2709
out:
2710 2711
		if (link_net)
			put_net(link_net);
2712
		put_net(dest_net);
P
Patrick McHardy 已提交
2713
		return err;
2714 2715 2716 2717 2718 2719 2720 2721 2722 2723
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
2724 2725 2726
	}
}

2727 2728
static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
2729
{
2730
	struct net *net = sock_net(skb->sk);
2731
	struct ifinfomsg *ifm;
2732
	char ifname[IFNAMSIZ];
2733 2734 2735
	struct nlattr *tb[IFLA_MAX+1];
	struct net_device *dev = NULL;
	struct sk_buff *nskb;
2736
	int err;
2737
	u32 ext_filter_mask = 0;
2738

2739
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy, extack);
2740
	if (err < 0)
2741
		return err;
2742

2743 2744 2745
	if (tb[IFLA_IFNAME])
		nla_strlcpy(ifname, tb[IFLA_IFNAME], IFNAMSIZ);

2746 2747 2748
	if (tb[IFLA_EXT_MASK])
		ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);

2749
	ifm = nlmsg_data(nlh);
2750 2751 2752 2753 2754
	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
2755 2756
		return -EINVAL;

2757 2758 2759
	if (dev == NULL)
		return -ENODEV;

2760
	nskb = nlmsg_new(if_nlmsg_size(dev, ext_filter_mask), GFP_KERNEL);
2761 2762
	if (nskb == NULL)
		return -ENOBUFS;
2763

2764
	err = rtnl_fill_ifinfo(nskb, dev, RTM_NEWLINK, NETLINK_CB(skb).portid,
2765
			       nlh->nlmsg_seq, 0, 0, ext_filter_mask, 0);
2766 2767 2768 2769
	if (err < 0) {
		/* -EMSGSIZE implies BUG in if_nlmsg_size */
		WARN_ON(err == -EMSGSIZE);
		kfree_skb(nskb);
2770
	} else
2771
		err = rtnl_unicast(nskb, net, NETLINK_CB(skb).portid);
2772

2773
	return err;
2774 2775
}

2776
static u16 rtnl_calcit(struct sk_buff *skb, struct nlmsghdr *nlh)
2777
{
2778 2779 2780 2781 2782
	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;
2783 2784 2785 2786 2787
	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);
2788

2789
	if (nlmsg_parse(nlh, hdrlen, tb, IFLA_MAX, ifla_policy, NULL) >= 0) {
2790 2791 2792
		if (tb[IFLA_EXT_MASK])
			ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);
	}
2793 2794 2795 2796 2797 2798 2799

	if (!ext_filter_mask)
		return NLMSG_GOODSIZE;
	/*
	 * traverse the list of net devices and compute the minimum
	 * buffer size based upon the filter mask.
	 */
2800 2801
	rcu_read_lock();
	for_each_netdev_rcu(net, dev) {
2802 2803 2804 2805
		min_ifinfo_dump_size = max_t(u16, min_ifinfo_dump_size,
					     if_nlmsg_size(dev,
						           ext_filter_mask));
	}
2806
	rcu_read_unlock();
2807

2808
	return nlmsg_total_size(min_ifinfo_dump_size);
2809 2810
}

A
Adrian Bunk 已提交
2811
static int rtnl_dump_all(struct sk_buff *skb, struct netlink_callback *cb)
L
Linus Torvalds 已提交
2812 2813 2814 2815 2816 2817
{
	int idx;
	int s_idx = cb->family;

	if (s_idx == 0)
		s_idx = 1;
2818

2819
	for (idx = 1; idx <= RTNL_FAMILY_MAX; idx++) {
L
Linus Torvalds 已提交
2820
		int type = cb->nlh->nlmsg_type-RTM_BASE;
2821 2822 2823
		struct rtnl_link *handlers;
		rtnl_dumpit_func dumpit;

L
Linus Torvalds 已提交
2824 2825
		if (idx < s_idx || idx == PF_PACKET)
			continue;
2826 2827 2828

		handlers = rtnl_dereference(rtnl_msg_handlers[idx]);
		if (!handlers)
L
Linus Torvalds 已提交
2829
			continue;
2830 2831 2832 2833 2834

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

2835
		if (idx > s_idx) {
L
Linus Torvalds 已提交
2836
			memset(&cb->args[0], 0, sizeof(cb->args));
2837 2838 2839
			cb->prev_seq = 0;
			cb->seq = 0;
		}
2840
		if (dumpit(skb, cb))
L
Linus Torvalds 已提交
2841 2842 2843 2844 2845 2846 2847
			break;
	}
	cb->family = idx;

	return skb->len;
}

2848
struct sk_buff *rtmsg_ifinfo_build_skb(int type, struct net_device *dev,
2849 2850
				       unsigned int change,
				       u32 event, gfp_t flags)
L
Linus Torvalds 已提交
2851
{
2852
	struct net *net = dev_net(dev);
L
Linus Torvalds 已提交
2853
	struct sk_buff *skb;
2854
	int err = -ENOBUFS;
2855
	size_t if_info_size;
L
Linus Torvalds 已提交
2856

2857
	skb = nlmsg_new((if_info_size = if_nlmsg_size(dev, 0)), flags);
2858 2859
	if (skb == NULL)
		goto errout;
L
Linus Torvalds 已提交
2860

2861
	err = rtnl_fill_ifinfo(skb, dev, type, 0, 0, change, 0, 0, event);
2862 2863 2864 2865 2866 2867
	if (err < 0) {
		/* -EMSGSIZE implies BUG in if_nlmsg_size() */
		WARN_ON(err == -EMSGSIZE);
		kfree_skb(skb);
		goto errout;
	}
2868
	return skb;
2869 2870
errout:
	if (err < 0)
2871
		rtnl_set_sk_err(net, RTNLGRP_LINK, err);
2872 2873 2874 2875 2876 2877 2878 2879 2880 2881
	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);
}

2882 2883 2884
static void rtmsg_ifinfo_event(int type, struct net_device *dev,
			       unsigned int change, u32 event,
			       gfp_t flags)
2885 2886 2887
{
	struct sk_buff *skb;

2888 2889 2890
	if (dev->reg_state != NETREG_REGISTERED)
		return;

2891
	skb = rtmsg_ifinfo_build_skb(type, dev, change, event, flags);
2892 2893
	if (skb)
		rtmsg_ifinfo_send(skb, dev, flags);
L
Linus Torvalds 已提交
2894
}
2895 2896 2897 2898

void rtmsg_ifinfo(int type, struct net_device *dev, unsigned int change,
		  gfp_t flags)
{
2899
	rtmsg_ifinfo_event(type, dev, change, rtnl_get_event(0), flags);
2900
}
2901
EXPORT_SYMBOL(rtmsg_ifinfo);
L
Linus Torvalds 已提交
2902

2903 2904
static int nlmsg_populate_fdb_fill(struct sk_buff *skb,
				   struct net_device *dev,
2905
				   u8 *addr, u16 vid, u32 pid, u32 seq,
2906
				   int type, unsigned int flags,
2907
				   int nlflags, u16 ndm_state)
2908 2909 2910 2911
{
	struct nlmsghdr *nlh;
	struct ndmsg *ndm;

2912
	nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ndm), nlflags);
2913 2914 2915 2916 2917 2918 2919 2920 2921 2922
	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;
2923
	ndm->ndm_state   = ndm_state;
2924 2925 2926

	if (nla_put(skb, NDA_LLADDR, ETH_ALEN, addr))
		goto nla_put_failure;
2927 2928 2929
	if (vid)
		if (nla_put(skb, NDA_VLAN, sizeof(u16), &vid))
			goto nla_put_failure;
2930

2931 2932
	nlmsg_end(skb, nlh);
	return 0;
2933 2934 2935 2936 2937 2938

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

2939 2940
static inline size_t rtnl_fdb_nlmsg_size(void)
{
2941 2942 2943 2944
	return NLMSG_ALIGN(sizeof(struct ndmsg)) +
	       nla_total_size(ETH_ALEN) +	/* NDA_LLADDR */
	       nla_total_size(sizeof(u16)) +	/* NDA_VLAN */
	       0;
2945 2946
}

2947 2948
static void rtnl_fdb_notify(struct net_device *dev, u8 *addr, u16 vid, int type,
			    u16 ndm_state)
2949 2950 2951 2952 2953 2954 2955 2956 2957
{
	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;

2958
	err = nlmsg_populate_fdb_fill(skb, dev, addr, vid,
2959
				      0, 0, type, NTF_SELF, 0, ndm_state);
2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970
	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);
}

2971 2972 2973 2974 2975 2976
/**
 * 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,
2977
		     const unsigned char *addr, u16 vid,
2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989
		     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;
	}

2990 2991 2992 2993 2994
	if (vid) {
		pr_info("%s: vlans aren't supported yet for dev_uc|mc_add()\n", dev->name);
		return err;
	}

2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007
	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);

3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029
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;
}

3030 3031
static int rtnl_fdb_add(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
3032 3033 3034 3035 3036 3037
{
	struct net *net = sock_net(skb->sk);
	struct ndmsg *ndm;
	struct nlattr *tb[NDA_MAX+1];
	struct net_device *dev;
	u8 *addr;
3038
	u16 vid;
3039 3040
	int err;

3041
	err = nlmsg_parse(nlh, sizeof(*ndm), tb, NDA_MAX, NULL, extack);
3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063
	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]);

3064 3065 3066 3067
	err = fdb_vid_parse(tb[NDA_VLAN], &vid);
	if (err)
		return err;

3068 3069 3070 3071 3072
	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)) {
3073 3074 3075
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);
		const struct net_device_ops *ops = br_dev->netdev_ops;

3076 3077
		err = ops->ndo_fdb_add(ndm, tb, dev, addr, vid,
				       nlh->nlmsg_flags);
3078 3079 3080 3081 3082 3083 3084
		if (err)
			goto out;
		else
			ndm->ndm_flags &= ~NTF_MASTER;
	}

	/* Embedded bridge, macvlan, and any other device support */
3085 3086 3087
	if ((ndm->ndm_flags & NTF_SELF)) {
		if (dev->netdev_ops->ndo_fdb_add)
			err = dev->netdev_ops->ndo_fdb_add(ndm, tb, dev, addr,
3088
							   vid,
3089 3090
							   nlh->nlmsg_flags);
		else
3091
			err = ndo_dflt_fdb_add(ndm, tb, dev, addr, vid,
3092
					       nlh->nlmsg_flags);
3093

3094
		if (!err) {
3095 3096
			rtnl_fdb_notify(dev, addr, vid, RTM_NEWNEIGH,
					ndm->ndm_state);
3097
			ndm->ndm_flags &= ~NTF_SELF;
3098
		}
3099 3100 3101 3102 3103
	}
out:
	return err;
}

3104 3105 3106 3107 3108 3109
/**
 * 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,
3110
		     const unsigned char *addr, u16 vid)
3111
{
3112
	int err = -EINVAL;
3113 3114 3115 3116

	/* If aging addresses are supported device will need to
	 * implement its own handler for this.
	 */
3117
	if (!(ndm->ndm_state & NUD_PERMANENT)) {
3118
		pr_info("%s: FDB only supports static addresses\n", dev->name);
3119
		return err;
3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130
	}

	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);

3131 3132
static int rtnl_fdb_del(struct sk_buff *skb, struct nlmsghdr *nlh,
			struct netlink_ext_ack *extack)
3133 3134 3135
{
	struct net *net = sock_net(skb->sk);
	struct ndmsg *ndm;
3136
	struct nlattr *tb[NDA_MAX+1];
3137 3138 3139
	struct net_device *dev;
	int err = -EINVAL;
	__u8 *addr;
3140
	u16 vid;
3141

3142
	if (!netlink_capable(skb, CAP_NET_ADMIN))
3143 3144
		return -EPERM;

3145
	err = nlmsg_parse(nlh, sizeof(*ndm), tb, NDA_MAX, NULL, extack);
3146 3147
	if (err < 0)
		return err;
3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160

	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;
	}

3161 3162 3163 3164 3165 3166
	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]);
3167

3168 3169 3170 3171
	err = fdb_vid_parse(tb[NDA_VLAN], &vid);
	if (err)
		return err;

3172 3173 3174 3175 3176
	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)) {
3177 3178
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);
		const struct net_device_ops *ops = br_dev->netdev_ops;
3179

3180
		if (ops->ndo_fdb_del)
3181
			err = ops->ndo_fdb_del(ndm, tb, dev, addr, vid);
3182 3183 3184 3185 3186 3187 3188 3189

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

	/* Embedded bridge, macvlan, and any other device support */
3190 3191
	if (ndm->ndm_flags & NTF_SELF) {
		if (dev->netdev_ops->ndo_fdb_del)
3192 3193
			err = dev->netdev_ops->ndo_fdb_del(ndm, tb, dev, addr,
							   vid);
3194
		else
3195
			err = ndo_dflt_fdb_del(ndm, tb, dev, addr, vid);
3196

3197
		if (!err) {
3198 3199
			rtnl_fdb_notify(dev, addr, vid, RTM_DELNEIGH,
					ndm->ndm_state);
3200
			ndm->ndm_flags &= ~NTF_SELF;
3201
		}
3202 3203 3204 3205 3206
	}
out:
	return err;
}

3207 3208 3209 3210 3211 3212 3213 3214
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;
3215
	u32 portid, seq;
3216

3217
	portid = NETLINK_CB(cb->skb).portid;
3218 3219 3220
	seq = cb->nlh->nlmsg_seq;

	list_for_each_entry(ha, &list->list, list) {
3221
		if (*idx < cb->args[2])
3222 3223
			goto skip;

3224
		err = nlmsg_populate_fdb_fill(skb, dev, ha->addr, 0,
3225
					      portid, seq,
3226
					      RTM_NEWNEIGH, NTF_SELF,
3227
					      NLM_F_MULTI, NUD_PERMANENT);
3228 3229 3230 3231 3232 3233 3234 3235 3236
		if (err < 0)
			return err;
skip:
		*idx += 1;
	}
	return 0;
}

/**
3237
 * ndo_dflt_fdb_dump - default netdevice operation to dump an FDB table.
3238 3239 3240 3241
 * @nlh: netlink message header
 * @dev: netdevice
 *
 * Default netdevice operation to dump the existing unicast address list.
3242
 * Returns number of addresses from list put in skb.
3243 3244 3245 3246
 */
int ndo_dflt_fdb_dump(struct sk_buff *skb,
		      struct netlink_callback *cb,
		      struct net_device *dev,
3247
		      struct net_device *filter_dev,
3248
		      int *idx)
3249 3250 3251 3252
{
	int err;

	netif_addr_lock_bh(dev);
3253
	err = nlmsg_populate_fdb(skb, cb, dev, idx, &dev->uc);
3254 3255
	if (err)
		goto out;
3256
	err = nlmsg_populate_fdb(skb, cb, dev, idx, &dev->mc);
3257 3258
out:
	netif_addr_unlock_bh(dev);
3259
	return err;
3260 3261 3262
}
EXPORT_SYMBOL(ndo_dflt_fdb_dump);

3263 3264 3265
static int rtnl_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb)
{
	struct net_device *dev;
3266 3267 3268 3269 3270 3271
	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);
3272
	struct hlist_head *head;
3273 3274
	int brport_idx = 0;
	int br_idx = 0;
3275 3276 3277 3278
	int h, s_h;
	int idx = 0, s_idx;
	int err = 0;
	int fidx = 0;
3279

3280 3281 3282 3283 3284
	err = nlmsg_parse(cb->nlh, sizeof(struct ifinfomsg), tb,
			  IFLA_MAX, ifla_policy, NULL);
	if (err < 0) {
		return -EINVAL;
	} else if (err == 0) {
3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298
		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;
	}

3299 3300
	s_h = cb->args[0];
	s_idx = cb->args[1];
3301

3302 3303 3304 3305
	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) {
3306

3307
			if (brport_idx && (dev->ifindex != brport_idx))
3308 3309
				continue;

3310 3311 3312 3313 3314 3315 3316 3317 3318
			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;
3319

3320 3321 3322 3323 3324
				if (br_dev != netdev_master_upper_dev_get(dev) &&
				    !(dev->priv_flags & IFF_EBRIDGE))
					continue;
				cops = ops;
			}
3325

3326 3327
			if (idx < s_idx)
				goto cont;
3328

3329 3330 3331 3332 3333 3334 3335 3336 3337
			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;
				}
			}
3338

3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356
			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++;
		}
3357 3358
	}

3359 3360 3361 3362 3363
out:
	cb->args[0] = h;
	cb->args[1] = idx;
	cb->args[2] = fidx;

3364 3365 3366
	return skb->len;
}

3367 3368 3369 3370 3371 3372 3373 3374
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;
}

3375
int ndo_dflt_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
3376
			    struct net_device *dev, u16 mode,
3377 3378 3379 3380 3381
			    u32 flags, u32 mask, int nlflags,
			    u32 filter_mask,
			    int (*vlan_fill)(struct sk_buff *skb,
					     struct net_device *dev,
					     u32 filter_mask))
3382 3383 3384 3385
{
	struct nlmsghdr *nlh;
	struct ifinfomsg *ifm;
	struct nlattr *br_afspec;
3386
	struct nlattr *protinfo;
3387
	u8 operstate = netif_running(dev) ? dev->operstate : IF_OPER_DOWN;
3388
	struct net_device *br_dev = netdev_master_upper_dev_get(dev);
3389
	int err = 0;
3390

3391
	nlh = nlmsg_put(skb, pid, seq, RTM_NEWLINK, sizeof(*ifm), nlflags);
3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406
	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) ||
3407 3408
	    (br_dev &&
	     nla_put_u32(skb, IFLA_MASTER, br_dev->ifindex)) ||
3409 3410
	    (dev->addr_len &&
	     nla_put(skb, IFLA_ADDRESS, dev->addr_len, dev->dev_addr)) ||
3411 3412
	    (dev->ifindex != dev_get_iflink(dev) &&
	     nla_put_u32(skb, IFLA_LINK, dev_get_iflink(dev))))
3413 3414 3415 3416 3417 3418
		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
3419
	if (nla_put_u16(skb, IFLA_BRIDGE_FLAGS, BRIDGE_FLAGS_SELF)) {
3420 3421 3422
		nla_nest_cancel(skb, br_afspec);
		goto nla_put_failure;
	}
R
Roopa Prabhu 已提交
3423 3424 3425 3426 3427 3428 3429

	if (mode != BRIDGE_MODE_UNDEF) {
		if (nla_put_u16(skb, IFLA_BRIDGE_MODE, mode)) {
			nla_nest_cancel(skb, br_afspec);
			goto nla_put_failure;
		}
	}
3430 3431 3432 3433 3434 3435 3436
	if (vlan_fill) {
		err = vlan_fill(skb, dev, filter_mask);
		if (err) {
			nla_nest_cancel(skb, br_afspec);
			goto nla_put_failure;
		}
	}
3437 3438
	nla_nest_end(skb, br_afspec);

3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465
	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);

3466 3467
	nlmsg_end(skb, nlh);
	return 0;
3468 3469
nla_put_failure:
	nlmsg_cancel(skb, nlh);
3470
	return err ? err : -EMSGSIZE;
3471
}
3472
EXPORT_SYMBOL_GPL(ndo_dflt_bridge_getlink);
3473

J
John Fastabend 已提交
3474 3475 3476 3477 3478 3479 3480
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;
3481
	u32 filter_mask = 0;
3482
	int err;
3483

3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495
	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
3496 3497 3498 3499

	rcu_read_lock();
	for_each_netdev_rcu(net, dev) {
		const struct net_device_ops *ops = dev->netdev_ops;
3500
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);
J
John Fastabend 已提交
3501

3502
		if (br_dev && br_dev->netdev_ops->ndo_bridge_getlink) {
3503 3504 3505 3506
			if (idx >= cb->args[0]) {
				err = br_dev->netdev_ops->ndo_bridge_getlink(
						skb, portid, seq, dev,
						filter_mask, NLM_F_MULTI);
3507 3508 3509 3510 3511 3512
				if (err < 0 && err != -EOPNOTSUPP) {
					if (likely(skb->len))
						break;

					goto out_err;
				}
3513
			}
3514
			idx++;
J
John Fastabend 已提交
3515 3516 3517
		}

		if (ops->ndo_bridge_getlink) {
3518 3519 3520 3521 3522
			if (idx >= cb->args[0]) {
				err = ops->ndo_bridge_getlink(skb, portid,
							      seq, dev,
							      filter_mask,
							      NLM_F_MULTI);
3523 3524 3525 3526 3527 3528
				if (err < 0 && err != -EOPNOTSUPP) {
					if (likely(skb->len))
						break;

					goto out_err;
				}
3529
			}
3530
			idx++;
J
John Fastabend 已提交
3531 3532
		}
	}
3533 3534
	err = skb->len;
out_err:
J
John Fastabend 已提交
3535 3536 3537
	rcu_read_unlock();
	cb->args[0] = idx;

3538
	return err;
J
John Fastabend 已提交
3539 3540
}

3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555
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 */
}

3556
static int rtnl_bridge_notify(struct net_device *dev)
3557 3558 3559 3560 3561
{
	struct net *net = dev_net(dev);
	struct sk_buff *skb;
	int err = -EOPNOTSUPP;

3562 3563 3564
	if (!dev->netdev_ops->ndo_bridge_getlink)
		return 0;

3565 3566 3567 3568 3569 3570
	skb = nlmsg_new(bridge_nlmsg_size(), GFP_ATOMIC);
	if (!skb) {
		err = -ENOMEM;
		goto errout;
	}

3571
	err = dev->netdev_ops->ndo_bridge_getlink(skb, 0, 0, dev, 0, 0);
3572 3573
	if (err < 0)
		goto errout;
3574

3575 3576 3577
	if (!skb->len)
		goto errout;

3578 3579 3580 3581 3582
	rtnl_notify(skb, net, 0, RTNLGRP_LINK, NULL, GFP_ATOMIC);
	return 0;
errout:
	WARN_ON(err == -EMSGSIZE);
	kfree_skb(skb);
3583 3584
	if (err)
		rtnl_set_sk_err(net, RTNLGRP_LINK, err);
3585 3586 3587
	return err;
}

3588 3589
static int rtnl_bridge_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
			       struct netlink_ext_ack *extack)
J
John Fastabend 已提交
3590 3591 3592 3593
{
	struct net *net = sock_net(skb->sk);
	struct ifinfomsg *ifm;
	struct net_device *dev;
3594 3595
	struct nlattr *br_spec, *attr = NULL;
	int rem, err = -EOPNOTSUPP;
3596
	u16 flags = 0;
3597
	bool have_flags = false;
J
John Fastabend 已提交
3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611

	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;
	}

3612 3613 3614 3615
	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) {
3616 3617 3618
				if (nla_len(attr) < sizeof(flags))
					return -EINVAL;

3619
				have_flags = true;
3620 3621 3622 3623 3624 3625 3626
				flags = nla_get_u16(attr);
				break;
			}
		}
	}

	if (!flags || (flags & BRIDGE_FLAGS_MASTER)) {
3627 3628 3629
		struct net_device *br_dev = netdev_master_upper_dev_get(dev);

		if (!br_dev || !br_dev->netdev_ops->ndo_bridge_setlink) {
3630 3631 3632 3633
			err = -EOPNOTSUPP;
			goto out;
		}

3634
		err = br_dev->netdev_ops->ndo_bridge_setlink(dev, nlh, flags);
J
John Fastabend 已提交
3635 3636
		if (err)
			goto out;
3637 3638

		flags &= ~BRIDGE_FLAGS_MASTER;
J
John Fastabend 已提交
3639 3640
	}

3641 3642 3643 3644
	if ((flags & BRIDGE_FLAGS_SELF)) {
		if (!dev->netdev_ops->ndo_bridge_setlink)
			err = -EOPNOTSUPP;
		else
3645 3646
			err = dev->netdev_ops->ndo_bridge_setlink(dev, nlh,
								  flags);
3647
		if (!err) {
3648
			flags &= ~BRIDGE_FLAGS_SELF;
3649 3650 3651 3652 3653 3654

			/* Generate event to notify upper layer of bridge
			 * change
			 */
			err = rtnl_bridge_notify(dev);
		}
3655
	}
J
John Fastabend 已提交
3656

3657
	if (have_flags)
3658
		memcpy(nla_data(attr), &flags, sizeof(flags));
J
John Fastabend 已提交
3659 3660 3661 3662
out:
	return err;
}

3663 3664
static int rtnl_bridge_dellink(struct sk_buff *skb, struct nlmsghdr *nlh,
			       struct netlink_ext_ack *extack)
3665 3666 3667 3668 3669 3670
{
	struct net *net = sock_net(skb->sk);
	struct ifinfomsg *ifm;
	struct net_device *dev;
	struct nlattr *br_spec, *attr = NULL;
	int rem, err = -EOPNOTSUPP;
3671
	u16 flags = 0;
3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690
	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) {
3691 3692 3693
				if (nla_len(attr) < sizeof(flags))
					return -EINVAL;

3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708
				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;
		}

3709
		err = br_dev->netdev_ops->ndo_bridge_dellink(dev, nlh, flags);
3710 3711 3712 3713 3714 3715 3716 3717 3718 3719
		if (err)
			goto out;

		flags &= ~BRIDGE_FLAGS_MASTER;
	}

	if ((flags & BRIDGE_FLAGS_SELF)) {
		if (!dev->netdev_ops->ndo_bridge_dellink)
			err = -EOPNOTSUPP;
		else
3720 3721
			err = dev->netdev_ops->ndo_bridge_dellink(dev, nlh,
								  flags);
3722

3723
		if (!err) {
3724
			flags &= ~BRIDGE_FLAGS_SELF;
3725 3726 3727 3728 3729 3730

			/* Generate event to notify upper layer of bridge
			 * change
			 */
			err = rtnl_bridge_notify(dev);
		}
3731 3732 3733 3734 3735 3736 3737 3738
	}

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

3739 3740 3741 3742 3743 3744
static bool stats_attr_valid(unsigned int mask, int attrid, int idxattr)
{
	return (mask & IFLA_STATS_FILTER_BIT(attrid)) &&
	       (!idxattr || idxattr == attrid);
}

3745 3746 3747 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
#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;

3777
		if (!dev->netdev_ops->ndo_has_offload_stats(dev, attr_id))
3778 3779 3780 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
			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++) {
3818
		if (!dev->netdev_ops->ndo_has_offload_stats(dev, attr_id))
3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829
			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;
}

3830 3831
static int rtnl_fill_statsinfo(struct sk_buff *skb, struct net_device *dev,
			       int type, u32 pid, u32 seq, u32 change,
3832 3833
			       unsigned int flags, unsigned int filter_mask,
			       int *idxattr, int *prividx)
3834 3835 3836 3837
{
	struct if_stats_msg *ifsm;
	struct nlmsghdr *nlh;
	struct nlattr *attr;
3838
	int s_prividx = *prividx;
3839
	int err;
3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850

	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;

3851
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_64, *idxattr)) {
3852 3853
		struct rtnl_link_stats64 *sp;

3854 3855 3856
		attr = nla_reserve_64bit(skb, IFLA_STATS_LINK_64,
					 sizeof(struct rtnl_link_stats64),
					 IFLA_STATS_UNSPEC);
3857 3858 3859 3860 3861 3862 3863
		if (!attr)
			goto nla_put_failure;

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

3864 3865 3866 3867 3868 3869 3870 3871 3872 3873
	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;

3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897
			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);
3898 3899 3900 3901 3902 3903 3904
			nla_nest_end(skb, attr);
			if (err)
				goto nla_put_failure;
			*idxattr = 0;
		}
	}

3905 3906 3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922
	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;
	}

3923 3924 3925 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
	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;
	}

3956 3957 3958 3959 3960
	nlmsg_end(skb, nlh);

	return 0;

nla_put_failure:
3961 3962 3963 3964 3965
	/* 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);
3966 3967 3968 3969 3970 3971 3972 3973 3974

	return -EMSGSIZE;
}

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

3975
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_64, 0))
3976 3977
		size += nla_total_size_64bit(sizeof(struct rtnl_link_stats64));

3978 3979
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_XSTATS, 0)) {
		const struct rtnl_link_ops *ops = dev->rtnl_link_ops;
3980
		int attr = IFLA_STATS_LINK_XSTATS;
3981 3982

		if (ops && ops->get_linkxstats_size) {
3983 3984
			size += nla_total_size(ops->get_linkxstats_size(dev,
									attr));
3985 3986 3987 3988 3989
			/* for IFLA_STATS_LINK_XSTATS */
			size += nla_total_size(0);
		}
	}

3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007 4008
	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);
		}
	}

4009 4010 4011
	if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_OFFLOAD_XSTATS, 0))
		size += rtnl_get_offload_stats_size(dev);

4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028
	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);
			}
		}
	}

4029 4030 4031
	return size;
}

4032 4033
static int rtnl_stats_get(struct sk_buff *skb, struct nlmsghdr *nlh,
			  struct netlink_ext_ack *extack)
4034 4035 4036
{
	struct net *net = sock_net(skb->sk);
	struct net_device *dev = NULL;
4037 4038
	int idxattr = 0, prividx = 0;
	struct if_stats_msg *ifsm;
4039 4040 4041 4042
	struct sk_buff *nskb;
	u32 filter_mask;
	int err;

4043 4044 4045
	if (nlmsg_len(nlh) < sizeof(*ifsm))
		return -EINVAL;

4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064
	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,
4065
				  0, filter_mask, &idxattr, &prividx);
4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077 4078
	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)
{
4079
	int h, s_h, err, s_idx, s_idxattr, s_prividx;
4080
	struct net *net = sock_net(skb->sk);
4081
	unsigned int flags = NLM_F_MULTI;
4082 4083
	struct if_stats_msg *ifsm;
	struct hlist_head *head;
4084
	struct net_device *dev;
4085
	u32 filter_mask = 0;
4086
	int idx = 0;
4087 4088 4089

	s_h = cb->args[0];
	s_idx = cb->args[1];
4090 4091
	s_idxattr = cb->args[2];
	s_prividx = cb->args[3];
4092 4093 4094

	cb->seq = net->dev_base_seq;

4095 4096 4097
	if (nlmsg_len(cb->nlh) < sizeof(*ifsm))
		return -EINVAL;

4098 4099 4100 4101 4102 4103 4104 4105 4106 4107 4108 4109 4110 4111
	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,
4112 4113
						  flags, filter_mask,
						  &s_idxattr, &s_prividx);
4114 4115 4116 4117 4118 4119 4120
			/* 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;
4121 4122
			s_prividx = 0;
			s_idxattr = 0;
4123 4124 4125 4126 4127 4128
			nl_dump_check_consistent(cb, nlmsg_hdr(skb));
cont:
			idx++;
		}
	}
out:
4129 4130
	cb->args[3] = s_prividx;
	cb->args[2] = s_idxattr;
4131 4132 4133 4134 4135 4136
	cb->args[1] = idx;
	cb->args[0] = h;

	return skb->len;
}

L
Linus Torvalds 已提交
4137 4138
/* Process one rtnetlink message. */

J
Johannes Berg 已提交
4139 4140
static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
			     struct netlink_ext_ack *extack)
L
Linus Torvalds 已提交
4141
{
4142
	struct net *net = sock_net(skb->sk);
4143 4144
	struct rtnl_link *handlers;
	int err = -EOPNOTSUPP;
4145
	rtnl_doit_func doit;
4146
	int kind;
L
Linus Torvalds 已提交
4147 4148 4149 4150 4151
	int family;
	int type;

	type = nlh->nlmsg_type;
	if (type > RTM_MAX)
4152
		return -EOPNOTSUPP;
L
Linus Torvalds 已提交
4153 4154 4155 4156

	type -= RTM_BASE;

	/* All the messages must have at least 1 byte length */
4157
	if (nlmsg_len(nlh) < sizeof(struct rtgenmsg))
L
Linus Torvalds 已提交
4158 4159
		return 0;

4160
	family = ((struct rtgenmsg *)nlmsg_data(nlh))->rtgen_family;
L
Linus Torvalds 已提交
4161 4162
	kind = type&3;

4163
	if (kind != 2 && !netlink_net_capable(skb, CAP_NET_ADMIN))
4164
		return -EPERM;
L
Linus Torvalds 已提交
4165

4166 4167 4168 4169 4170 4171 4172 4173 4174 4175
	if (family > ARRAY_SIZE(rtnl_msg_handlers))
		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]);
	}

4176
	if (kind == 2 && nlh->nlmsg_flags&NLM_F_DUMP) {
4177
		struct sock *rtnl;
4178
		rtnl_dumpit_func dumpit;
4179
		u16 min_dump_alloc = 0;
L
Linus Torvalds 已提交
4180

4181 4182 4183 4184 4185 4186
		dumpit = READ_ONCE(handlers[type].dumpit);
		if (!dumpit) {
			family = PF_UNSPEC;
			handlers = rcu_dereference(rtnl_msg_handlers[PF_UNSPEC]);
			if (!handlers)
				goto err_unlock;
4187

4188 4189 4190 4191
			dumpit = READ_ONCE(handlers[type].dumpit);
			if (!dumpit)
				goto err_unlock;
		}
4192

4193 4194
		refcount_inc(&rtnl_msg_handlers_ref[family]);

4195 4196
		if (type == RTM_GETLINK)
			min_dump_alloc = rtnl_calcit(skb, nlh);
4197

4198 4199
		rcu_read_unlock();

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

4212
	rcu_read_unlock();
L
Linus Torvalds 已提交
4213

4214 4215 4216 4217 4218 4219 4220
	rtnl_lock();
	handlers = rtnl_dereference(rtnl_msg_handlers[family]);
	if (handlers) {
		doit = READ_ONCE(handlers[type].doit);
		if (doit)
			err = doit(skb, nlh, extack);
	}
4221 4222 4223 4224
	rtnl_unlock();
	return err;

err_unlock:
4225
	rcu_read_unlock();
4226
	return -EOPNOTSUPP;
L
Linus Torvalds 已提交
4227 4228
}

4229
static void rtnetlink_rcv(struct sk_buff *skb)
L
Linus Torvalds 已提交
4230
{
4231
	netlink_rcv_skb(skb, &rtnetlink_rcv_msg);
L
Linus Torvalds 已提交
4232 4233
}

4234 4235 4236 4237 4238 4239 4240 4241 4242 4243 4244 4245
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
4246 4247
static int rtnetlink_event(struct notifier_block *this, unsigned long event, void *ptr)
{
4248
	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
4249

L
Linus Torvalds 已提交
4250
	switch (event) {
4251
	case NETDEV_REBOOT:
4252
	case NETDEV_CHANGEADDR:
4253 4254 4255 4256 4257 4258
	case NETDEV_CHANGENAME:
	case NETDEV_FEAT_CHANGE:
	case NETDEV_BONDING_FAILOVER:
	case NETDEV_NOTIFY_PEERS:
	case NETDEV_RESEND_IGMP:
	case NETDEV_CHANGEINFODATA:
4259 4260
		rtmsg_ifinfo_event(RTM_NEWLINK, dev, 0, rtnl_get_event(event),
				   GFP_KERNEL);
L
Linus Torvalds 已提交
4261 4262 4263 4264 4265 4266 4267 4268 4269 4270 4271
		break;
	default:
		break;
	}
	return NOTIFY_DONE;
}

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

4272

4273
static int __net_init rtnetlink_net_init(struct net *net)
4274 4275
{
	struct sock *sk;
4276 4277 4278 4279
	struct netlink_kernel_cfg cfg = {
		.groups		= RTNLGRP_MAX,
		.input		= rtnetlink_rcv,
		.cb_mutex	= &rtnl_mutex,
4280
		.flags		= NL_CFG_F_NONROOT_RECV,
4281
		.bind		= rtnetlink_bind,
4282 4283
	};

4284
	sk = netlink_kernel_create(net, NETLINK_ROUTE, &cfg);
4285 4286 4287 4288 4289 4290
	if (!sk)
		return -ENOMEM;
	net->rtnl = sk;
	return 0;
}

4291
static void __net_exit rtnetlink_net_exit(struct net *net)
4292
{
4293 4294
	netlink_kernel_release(net->rtnl);
	net->rtnl = NULL;
4295 4296 4297 4298 4299 4300 4301
}

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

L
Linus Torvalds 已提交
4302 4303
void __init rtnetlink_init(void)
{
4304
	if (register_pernet_subsys(&rtnetlink_net_ops))
L
Linus Torvalds 已提交
4305
		panic("rtnetlink_init: cannot initialize rtnetlink\n");
4306

L
Linus Torvalds 已提交
4307
	register_netdevice_notifier(&rtnetlink_dev_notifier);
4308

4309
	rtnl_register(PF_UNSPEC, RTM_GETLINK, rtnl_getlink,
4310 4311 4312 4313
		      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);
4314

4315 4316 4317
	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);
4318

4319 4320 4321
	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
4322

4323 4324 4325
	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);
4326 4327

	rtnl_register(PF_UNSPEC, RTM_GETSTATS, rtnl_stats_get, rtnl_stats_dump,
4328
		      0);
L
Linus Torvalds 已提交
4329
}