rtnetlink.c 33.7 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 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
/*
 * 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.
 */

#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>
36
#include <linux/mutex.h>
37
#include <linux/if_addr.h>
L
Linus Torvalds 已提交
38 39 40 41 42 43 44 45 46 47 48 49 50

#include <asm/uaccess.h>
#include <asm/system.h>

#include <linux/inet.h>
#include <linux/netdevice.h>
#include <net/ip.h>
#include <net/protocol.h>
#include <net/arp.h>
#include <net/route.h>
#include <net/udp.h>
#include <net/sock.h>
#include <net/pkt_sched.h>
51
#include <net/fib_rules.h>
52
#include <net/rtnetlink.h>
53
#include <net/net_namespace.h>
L
Linus Torvalds 已提交
54

E
Eric Dumazet 已提交
55
struct rtnl_link {
56 57 58 59
	rtnl_doit_func		doit;
	rtnl_dumpit_func	dumpit;
};

60
static DEFINE_MUTEX(rtnl_mutex);
L
Linus Torvalds 已提交
61 62 63

void rtnl_lock(void)
{
64
	mutex_lock(&rtnl_mutex);
L
Linus Torvalds 已提交
65
}
E
Eric Dumazet 已提交
66
EXPORT_SYMBOL(rtnl_lock);
L
Linus Torvalds 已提交
67

68
void __rtnl_unlock(void)
L
Linus Torvalds 已提交
69
{
70
	mutex_unlock(&rtnl_mutex);
L
Linus Torvalds 已提交
71
}
72

L
Linus Torvalds 已提交
73 74
void rtnl_unlock(void)
{
H
Herbert Xu 已提交
75
	/* This fellow will unlock it for us. */
L
Linus Torvalds 已提交
76 77
	netdev_run_todo();
}
E
Eric Dumazet 已提交
78
EXPORT_SYMBOL(rtnl_unlock);
L
Linus Torvalds 已提交
79

80 81 82 83
int rtnl_trylock(void)
{
	return mutex_trylock(&rtnl_mutex);
}
E
Eric Dumazet 已提交
84
EXPORT_SYMBOL(rtnl_trylock);
85

86 87 88 89
int rtnl_is_locked(void)
{
	return mutex_is_locked(&rtnl_mutex);
}
E
Eric Dumazet 已提交
90
EXPORT_SYMBOL(rtnl_is_locked);
91

A
Adrian Bunk 已提交
92
static struct rtnl_link *rtnl_msg_handlers[NPROTO];
93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112

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

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

	return msgindex;
}

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

	tab = rtnl_msg_handlers[protocol];
113
	if (tab == NULL || tab[msgindex].doit == NULL)
114 115
		tab = rtnl_msg_handlers[PF_UNSPEC];

116
	return tab ? tab[msgindex].doit : NULL;
117 118 119 120 121 122 123
}

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

	tab = rtnl_msg_handlers[protocol];
124
	if (tab == NULL || tab[msgindex].dumpit == NULL)
125 126
		tab = rtnl_msg_handlers[PF_UNSPEC];

127
	return tab ? tab[msgindex].dumpit : NULL;
128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232
}

/**
 * __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
 *
 * 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,
		    rtnl_doit_func doit, rtnl_dumpit_func dumpit)
{
	struct rtnl_link *tab;
	int msgindex;

	BUG_ON(protocol < 0 || protocol >= NPROTO);
	msgindex = rtm_msgindex(msgtype);

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

		rtnl_msg_handlers[protocol] = tab;
	}

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

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

	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
 * of memory implies no sense in continueing.
 */
void rtnl_register(int protocol, int msgtype,
		   rtnl_doit_func doit, rtnl_dumpit_func dumpit)
{
	if (__rtnl_register(protocol, msgtype, doit, dumpit) < 0)
		panic("Unable to register rtnetlink message handler, "
		      "protocol = %d, message type = %d\n",
		      protocol, msgtype);
}
EXPORT_SYMBOL_GPL(rtnl_register);

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

	BUG_ON(protocol < 0 || protocol >= NPROTO);
	msgindex = rtm_msgindex(msgtype);

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

	rtnl_msg_handlers[protocol][msgindex].doit = NULL;
	rtnl_msg_handlers[protocol][msgindex].dumpit = NULL;

	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)
{
	BUG_ON(protocol < 0 || protocol >= NPROTO);

	kfree(rtnl_msg_handlers[protocol]);
	rtnl_msg_handlers[protocol] = NULL;
}
EXPORT_SYMBOL_GPL(rtnl_unregister_all);
L
Linus Torvalds 已提交
233

P
Patrick McHardy 已提交
234 235 236 237 238 239 240 241 242 243 244 245 246 247
static LIST_HEAD(link_ops);

/**
 * __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)
{
248
	if (!ops->dellink)
249
		ops->dellink = unregister_netdevice_queue;
250

P
Patrick McHardy 已提交
251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272
	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);

273 274 275
static void __rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops)
{
	struct net_device *dev;
276 277
	LIST_HEAD(list_kill);

278
	for_each_netdev(net, dev) {
279 280
		if (dev->rtnl_link_ops == ops)
			ops->dellink(dev, &list_kill);
281
	}
282
	unregister_netdevice_many(&list_kill);
283 284 285 286 287 288 289 290 291 292
}

void rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops)
{
	rtnl_lock();
	__rtnl_kill_links(net, ops);
	rtnl_unlock();
}
EXPORT_SYMBOL_GPL(rtnl_kill_links);

P
Patrick McHardy 已提交
293 294 295 296
/**
 * __rtnl_link_unregister - Unregister rtnl_link_ops from rtnetlink.
 * @ops: struct rtnl_link_ops * to unregister
 *
297
 * The caller must hold the rtnl_mutex.
P
Patrick McHardy 已提交
298 299 300
 */
void __rtnl_link_unregister(struct rtnl_link_ops *ops)
{
301
	struct net *net;
302

303
	for_each_net(net) {
304
		__rtnl_kill_links(net, ops);
305
	}
P
Patrick McHardy 已提交
306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
	list_del(&ops->list);
}
EXPORT_SYMBOL_GPL(__rtnl_link_unregister);

/**
 * 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)
{
	rtnl_lock();
	__rtnl_link_unregister(ops);
	rtnl_unlock();
}
EXPORT_SYMBOL_GPL(rtnl_link_unregister);

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;
}

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;

	size = nlmsg_total_size(sizeof(struct nlattr)) + /* IFLA_LINKINFO */
	       nlmsg_total_size(strlen(ops->kind) + 1);	 /* IFLA_INFO_KIND */

	if (ops->get_size)
		/* IFLA_INFO_DATA + nested data */
		size += nlmsg_total_size(sizeof(struct nlattr)) +
			ops->get_size(dev);

	if (ops->get_xstats_size)
		size += ops->get_xstats_size(dev);	/* IFLA_INFO_XSTATS */

	return size;
}

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

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

	if (nla_put_string(skb, IFLA_INFO_KIND, ops->kind) < 0)
		goto err_cancel_link;
	if (ops->fill_xstats) {
		err = ops->fill_xstats(skb, dev);
		if (err < 0)
			goto err_cancel_link;
	}
	if (ops->fill_info) {
		data = nla_nest_start(skb, IFLA_INFO_DATA);
		if (data == NULL)
			goto err_cancel_link;
		err = ops->fill_info(skb, dev);
		if (err < 0)
			goto err_cancel_data;
		nla_nest_end(skb, data);
	}

	nla_nest_end(skb, linkinfo);
	return 0;

err_cancel_data:
	nla_nest_cancel(skb, data);
err_cancel_link:
	nla_nest_cancel(skb, linkinfo);
out:
	return err;
}

393
static const int rtm_min[RTM_NR_FAMILIES] =
L
Linus Torvalds 已提交
394
{
395 396 397
	[RTM_FAM(RTM_NEWLINK)]      = NLMSG_LENGTH(sizeof(struct ifinfomsg)),
	[RTM_FAM(RTM_NEWADDR)]      = NLMSG_LENGTH(sizeof(struct ifaddrmsg)),
	[RTM_FAM(RTM_NEWROUTE)]     = NLMSG_LENGTH(sizeof(struct rtmsg)),
398
	[RTM_FAM(RTM_NEWRULE)]      = NLMSG_LENGTH(sizeof(struct fib_rule_hdr)),
399 400 401 402 403 404
	[RTM_FAM(RTM_NEWQDISC)]     = NLMSG_LENGTH(sizeof(struct tcmsg)),
	[RTM_FAM(RTM_NEWTCLASS)]    = NLMSG_LENGTH(sizeof(struct tcmsg)),
	[RTM_FAM(RTM_NEWTFILTER)]   = NLMSG_LENGTH(sizeof(struct tcmsg)),
	[RTM_FAM(RTM_NEWACTION)]    = NLMSG_LENGTH(sizeof(struct tcamsg)),
	[RTM_FAM(RTM_GETMULTICAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)),
	[RTM_FAM(RTM_GETANYCAST)]   = NLMSG_LENGTH(sizeof(struct rtgenmsg)),
L
Linus Torvalds 已提交
405 406
};

407
static const int rta_max[RTM_NR_FAMILIES] =
L
Linus Torvalds 已提交
408
{
409 410 411
	[RTM_FAM(RTM_NEWLINK)]      = IFLA_MAX,
	[RTM_FAM(RTM_NEWADDR)]      = IFA_MAX,
	[RTM_FAM(RTM_NEWROUTE)]     = RTA_MAX,
412
	[RTM_FAM(RTM_NEWRULE)]      = FRA_MAX,
413 414 415 416
	[RTM_FAM(RTM_NEWQDISC)]     = TCA_MAX,
	[RTM_FAM(RTM_NEWTCLASS)]    = TCA_MAX,
	[RTM_FAM(RTM_NEWTFILTER)]   = TCA_MAX,
	[RTM_FAM(RTM_NEWACTION)]    = TCAA_MAX,
L
Linus Torvalds 已提交
417 418 419 420 421 422 423
};

void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const void *data)
{
	struct rtattr *rta;
	int size = RTA_LENGTH(attrlen);

E
Eric Dumazet 已提交
424
	rta = (struct rtattr *)skb_put(skb, RTA_ALIGN(size));
L
Linus Torvalds 已提交
425 426 427
	rta->rta_type = attrtype;
	rta->rta_len = size;
	memcpy(RTA_DATA(rta), data, attrlen);
428
	memset(RTA_DATA(rta) + attrlen, 0, RTA_ALIGN(size) - size);
L
Linus Torvalds 已提交
429
}
E
Eric Dumazet 已提交
430
EXPORT_SYMBOL(__rta_fill);
L
Linus Torvalds 已提交
431

432
int rtnetlink_send(struct sk_buff *skb, struct net *net, u32 pid, unsigned group, int echo)
L
Linus Torvalds 已提交
433
{
434
	struct sock *rtnl = net->rtnl;
L
Linus Torvalds 已提交
435 436
	int err = 0;

437
	NETLINK_CB(skb).dst_group = group;
L
Linus Torvalds 已提交
438 439 440 441 442 443 444 445
	if (echo)
		atomic_inc(&skb->users);
	netlink_broadcast(rtnl, skb, pid, group, GFP_KERNEL);
	if (echo)
		err = netlink_unicast(rtnl, skb, pid, MSG_DONTWAIT);
	return err;
}

446
int rtnl_unicast(struct sk_buff *skb, struct net *net, u32 pid)
447
{
448 449
	struct sock *rtnl = net->rtnl;

450 451
	return nlmsg_unicast(rtnl, skb, pid);
}
E
Eric Dumazet 已提交
452
EXPORT_SYMBOL(rtnl_unicast);
453

454 455
void rtnl_notify(struct sk_buff *skb, struct net *net, u32 pid, u32 group,
		 struct nlmsghdr *nlh, gfp_t flags)
456
{
457
	struct sock *rtnl = net->rtnl;
458 459 460 461 462
	int report = 0;

	if (nlh)
		report = nlmsg_report(nlh);

463
	nlmsg_notify(rtnl, skb, pid, group, report, flags);
464
}
E
Eric Dumazet 已提交
465
EXPORT_SYMBOL(rtnl_notify);
466

467
void rtnl_set_sk_err(struct net *net, u32 group, int error)
468
{
469 470
	struct sock *rtnl = net->rtnl;

471 472
	netlink_set_err(rtnl, 0, group, error);
}
E
Eric Dumazet 已提交
473
EXPORT_SYMBOL(rtnl_set_sk_err);
474

L
Linus Torvalds 已提交
475 476
int rtnetlink_put_metrics(struct sk_buff *skb, u32 *metrics)
{
477 478 479 480 481 482 483 484 485 486 487 488
	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]) {
			valid++;
			NLA_PUT_U32(skb, i+1, metrics[i]);
		}
L
Linus Torvalds 已提交
489 490
	}

491 492 493 494
	if (!valid) {
		nla_nest_cancel(skb, mx);
		return 0;
	}
495 496 497 498

	return nla_nest_end(skb, mx);

nla_put_failure:
499 500
	nla_nest_cancel(skb, mx);
	return -EMSGSIZE;
L
Linus Torvalds 已提交
501
}
E
Eric Dumazet 已提交
502
EXPORT_SYMBOL(rtnetlink_put_metrics);
L
Linus Torvalds 已提交
503

504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522
int rtnl_put_cacheinfo(struct sk_buff *skb, struct dst_entry *dst, u32 id,
		       u32 ts, u32 tsage, long expires, u32 error)
{
	struct rta_cacheinfo ci = {
		.rta_lastuse = jiffies_to_clock_t(jiffies - dst->lastuse),
		.rta_used = dst->__use,
		.rta_clntref = atomic_read(&(dst->__refcnt)),
		.rta_error = error,
		.rta_id =  id,
		.rta_ts = ts,
		.rta_tsage = tsage,
	};

	if (expires)
		ci.rta_expires = jiffies_to_clock_t(expires);

	return nla_put(skb, RTA_CACHEINFO, sizeof(ci), &ci);
}
EXPORT_SYMBOL_GPL(rtnl_put_cacheinfo);
L
Linus Torvalds 已提交
523

524
static void set_operstate(struct net_device *dev, unsigned char transition)
S
Stefan Rompf 已提交
525 526 527
{
	unsigned char operstate = dev->operstate;

E
Eric Dumazet 已提交
528
	switch (transition) {
S
Stefan Rompf 已提交
529 530 531 532 533 534 535 536 537 538 539 540
	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;
541
	}
S
Stefan Rompf 已提交
542 543 544 545 546

	if (dev->operstate != operstate) {
		write_lock_bh(&dev_base_lock);
		dev->operstate = operstate;
		write_unlock_bh(&dev_base_lock);
547 548
		netdev_state_change(dev);
	}
S
Stefan Rompf 已提交
549 550
}

551
static void copy_rtnl_link_stats(struct rtnl_link_stats *a,
552
				 const struct net_device_stats *b)
L
Linus Torvalds 已提交
553
{
554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581
	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;
};
L
Linus Torvalds 已提交
582

P
Patrick McHardy 已提交
583
static inline size_t if_nlmsg_size(const struct net_device *dev)
584 585 586
{
	return NLMSG_ALIGN(sizeof(struct ifinfomsg))
	       + nla_total_size(IFNAMSIZ) /* IFLA_IFNAME */
587
	       + nla_total_size(IFALIASZ) /* IFLA_IFALIAS */
588 589 590 591 592 593 594 595 596 597 598
	       + nla_total_size(IFNAMSIZ) /* IFLA_QDISC */
	       + nla_total_size(sizeof(struct rtnl_link_ifmap))
	       + nla_total_size(sizeof(struct rtnl_link_stats))
	       + 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 */
	       + nla_total_size(1) /* IFLA_OPERSTATE */
P
Patrick McHardy 已提交
599 600
	       + nla_total_size(1) /* IFLA_LINKMODE */
	       + rtnl_link_get_size(dev); /* IFLA_LINKINFO */
601 602
}

603
static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
604 605
			    int type, u32 pid, u32 seq, u32 change,
			    unsigned int flags)
606 607 608
{
	struct ifinfomsg *ifm;
	struct nlmsghdr *nlh;
609
	const struct net_device_stats *stats;
610
	struct nlattr *attr;
L
Linus Torvalds 已提交
611

612 613
	nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ifm), flags);
	if (nlh == NULL)
614
		return -EMSGSIZE;
L
Linus Torvalds 已提交
615

616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635
	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;

	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);

	if (dev->ifindex != dev->iflink)
		NLA_PUT_U32(skb, IFLA_LINK, dev->iflink);

	if (dev->master)
		NLA_PUT_U32(skb, IFLA_MASTER, dev->master->ifindex);
L
Linus Torvalds 已提交
636

637 638
	if (dev->qdisc)
		NLA_PUT_STRING(skb, IFLA_QDISC, dev->qdisc->ops->id);
S
Stefan Rompf 已提交
639

640 641 642
	if (dev->ifalias)
		NLA_PUT_STRING(skb, IFLA_IFALIAS, dev->ifalias);

L
Linus Torvalds 已提交
643 644 645 646 647 648 649 650 651
	if (1) {
		struct rtnl_link_ifmap map = {
			.mem_start   = dev->mem_start,
			.mem_end     = dev->mem_end,
			.base_addr   = dev->base_addr,
			.irq         = dev->irq,
			.dma         = dev->dma,
			.port        = dev->if_port,
		};
652
		NLA_PUT(skb, IFLA_MAP, sizeof(map), &map);
L
Linus Torvalds 已提交
653 654 655
	}

	if (dev->addr_len) {
656 657
		NLA_PUT(skb, IFLA_ADDRESS, dev->addr_len, dev->dev_addr);
		NLA_PUT(skb, IFLA_BROADCAST, dev->addr_len, dev->broadcast);
L
Linus Torvalds 已提交
658 659
	}

660 661 662 663
	attr = nla_reserve(skb, IFLA_STATS,
			sizeof(struct rtnl_link_stats));
	if (attr == NULL)
		goto nla_put_failure;
664

665
	stats = dev_get_stats(dev);
666
	copy_rtnl_link_stats(nla_data(attr), stats);
L
Linus Torvalds 已提交
667

P
Patrick McHardy 已提交
668 669 670 671 672
	if (dev->rtnl_link_ops) {
		if (rtnl_link_fill(skb, dev) < 0)
			goto nla_put_failure;
	}

673 674 675
	return nlmsg_end(skb, nlh);

nla_put_failure:
676 677
	nlmsg_cancel(skb, nlh);
	return -EMSGSIZE;
L
Linus Torvalds 已提交
678 679
}

680
static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
L
Linus Torvalds 已提交
681
{
682
	struct net *net = sock_net(skb->sk);
683 684
	int h, s_h;
	int idx = 0, s_idx;
L
Linus Torvalds 已提交
685
	struct net_device *dev;
686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702
	struct hlist_head *head;
	struct hlist_node *node;

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

	for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) {
		idx = 0;
		head = &net->dev_index_head[h];
		hlist_for_each_entry(dev, node, head, index_hlist) {
			if (idx < s_idx)
				goto cont;
			if (rtnl_fill_ifinfo(skb, dev, RTM_NEWLINK,
					     NETLINK_CB(cb->skb).pid,
					     cb->nlh->nlmsg_seq, 0,
					     NLM_F_MULTI) <= 0)
				goto out;
703
cont:
704 705
			idx++;
		}
L
Linus Torvalds 已提交
706
	}
707 708 709
out:
	cb->args[1] = idx;
	cb->args[0] = h;
L
Linus Torvalds 已提交
710 711 712 713

	return skb->len;
}

714
const struct nla_policy ifla_policy[IFLA_MAX+1] = {
715
	[IFLA_IFNAME]		= { .type = NLA_STRING, .len = IFNAMSIZ-1 },
P
Patrick McHardy 已提交
716 717
	[IFLA_ADDRESS]		= { .type = NLA_BINARY, .len = MAX_ADDR_LEN },
	[IFLA_BROADCAST]	= { .type = NLA_BINARY, .len = MAX_ADDR_LEN },
718
	[IFLA_MAP]		= { .len = sizeof(struct rtnl_link_ifmap) },
719
	[IFLA_MTU]		= { .type = NLA_U32 },
720
	[IFLA_LINK]		= { .type = NLA_U32 },
721 722 723 724
	[IFLA_TXQLEN]		= { .type = NLA_U32 },
	[IFLA_WEIGHT]		= { .type = NLA_U32 },
	[IFLA_OPERSTATE]	= { .type = NLA_U8 },
	[IFLA_LINKMODE]		= { .type = NLA_U8 },
725
	[IFLA_LINKINFO]		= { .type = NLA_NESTED },
726
	[IFLA_NET_NS_PID]	= { .type = NLA_U32 },
727
	[IFLA_IFALIAS]	        = { .type = NLA_STRING, .len = IFALIASZ-1 },
728
};
E
Eric Dumazet 已提交
729
EXPORT_SYMBOL(ifla_policy);
730

P
Patrick McHardy 已提交
731 732 733 734 735
static const struct nla_policy ifla_info_policy[IFLA_INFO_MAX+1] = {
	[IFLA_INFO_KIND]	= { .type = NLA_STRING },
	[IFLA_INFO_DATA]	= { .type = NLA_NESTED },
};

736 737 738 739 740 741 742 743 744 745 746 747 748 749 750
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;
	}

	return 0;
}

751
static int do_setlink(struct net_device *dev, struct ifinfomsg *ifm,
P
Patrick McHardy 已提交
752
		      struct nlattr **tb, char *ifname, int modified)
L
Linus Torvalds 已提交
753
{
754
	const struct net_device_ops *ops = dev->netdev_ops;
P
Patrick McHardy 已提交
755
	int send_addr_notify = 0;
756
	int err;
L
Linus Torvalds 已提交
757

758 759 760 761 762 763 764 765 766 767 768 769 770 771
	if (tb[IFLA_NET_NS_PID]) {
		struct net *net;
		net = get_net_ns_by_pid(nla_get_u32(tb[IFLA_NET_NS_PID]));
		if (IS_ERR(net)) {
			err = PTR_ERR(net);
			goto errout;
		}
		err = dev_change_net_namespace(dev, net, ifname);
		put_net(net);
		if (err)
			goto errout;
		modified = 1;
	}

772
	if (tb[IFLA_MAP]) {
L
Linus Torvalds 已提交
773 774 775
		struct rtnl_link_ifmap *u_map;
		struct ifmap k_map;

776
		if (!ops->ndo_set_config) {
L
Linus Torvalds 已提交
777
			err = -EOPNOTSUPP;
778
			goto errout;
L
Linus Torvalds 已提交
779 780 781 782
		}

		if (!netif_device_present(dev)) {
			err = -ENODEV;
783
			goto errout;
L
Linus Torvalds 已提交
784 785
		}

786
		u_map = nla_data(tb[IFLA_MAP]);
L
Linus Torvalds 已提交
787 788 789 790 791 792 793
		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;

794
		err = ops->ndo_set_config(dev, &k_map);
795
		if (err < 0)
796
			goto errout;
L
Linus Torvalds 已提交
797

798
		modified = 1;
L
Linus Torvalds 已提交
799 800
	}

801
	if (tb[IFLA_ADDRESS]) {
802 803 804
		struct sockaddr *sa;
		int len;

805
		if (!ops->ndo_set_mac_address) {
L
Linus Torvalds 已提交
806
			err = -EOPNOTSUPP;
807
			goto errout;
L
Linus Torvalds 已提交
808
		}
809

L
Linus Torvalds 已提交
810 811
		if (!netif_device_present(dev)) {
			err = -ENODEV;
812
			goto errout;
L
Linus Torvalds 已提交
813 814
		}

815 816 817 818
		len = sizeof(sa_family_t) + dev->addr_len;
		sa = kmalloc(len, GFP_KERNEL);
		if (!sa) {
			err = -ENOMEM;
819
			goto errout;
820 821
		}
		sa->sa_family = dev->type;
822
		memcpy(sa->sa_data, nla_data(tb[IFLA_ADDRESS]),
823
		       dev->addr_len);
824
		err = ops->ndo_set_mac_address(dev, sa);
825
		kfree(sa);
L
Linus Torvalds 已提交
826
		if (err)
827
			goto errout;
L
Linus Torvalds 已提交
828
		send_addr_notify = 1;
829
		modified = 1;
L
Linus Torvalds 已提交
830 831
	}

832 833 834
	if (tb[IFLA_MTU]) {
		err = dev_set_mtu(dev, nla_get_u32(tb[IFLA_MTU]));
		if (err < 0)
835
			goto errout;
836
		modified = 1;
L
Linus Torvalds 已提交
837 838
	}

839 840 841 842 843
	/*
	 * Interface selected by interface index but interface
	 * name provided implies that a name change has been
	 * requested.
	 */
844
	if (ifm->ifi_index > 0 && ifname[0]) {
845 846
		err = dev_change_name(dev, ifname);
		if (err < 0)
847
			goto errout;
848
		modified = 1;
L
Linus Torvalds 已提交
849 850
	}

851 852 853 854 855 856 857 858
	if (tb[IFLA_IFALIAS]) {
		err = dev_set_alias(dev, nla_data(tb[IFLA_IFALIAS]),
				    nla_len(tb[IFLA_IFALIAS]));
		if (err < 0)
			goto errout;
		modified = 1;
	}

859 860 861
	if (tb[IFLA_BROADCAST]) {
		nla_memcpy(dev->broadcast, tb[IFLA_BROADCAST], dev->addr_len);
		send_addr_notify = 1;
L
Linus Torvalds 已提交
862 863
	}

864 865 866 867 868 869 870
	if (ifm->ifi_flags || ifm->ifi_change) {
		unsigned int flags = ifm->ifi_flags;

		/* bugwards compatibility: ifi_change == 0 is treated as ~0 */
		if (ifm->ifi_change)
			flags = (flags & ifm->ifi_change) |
				(dev->flags & ~ifm->ifi_change);
871 872 873
		err = dev_change_flags(dev, flags);
		if (err < 0)
			goto errout;
874
	}
L
Linus Torvalds 已提交
875

876 877
	if (tb[IFLA_TXQLEN])
		dev->tx_queue_len = nla_get_u32(tb[IFLA_TXQLEN]);
S
Stefan Rompf 已提交
878

879
	if (tb[IFLA_OPERSTATE])
880
		set_operstate(dev, nla_get_u8(tb[IFLA_OPERSTATE]));
S
Stefan Rompf 已提交
881

882
	if (tb[IFLA_LINKMODE]) {
883 884 885
		write_lock_bh(&dev_base_lock);
		dev->link_mode = nla_get_u8(tb[IFLA_LINKMODE]);
		write_unlock_bh(&dev_base_lock);
S
Stefan Rompf 已提交
886 887
	}

L
Linus Torvalds 已提交
888 889
	err = 0;

890
errout:
891 892 893 894 895 896
	if (err < 0 && modified && net_ratelimit())
		printk(KERN_WARNING "A link change request failed with "
		       "some changes comitted already. Interface %s may "
		       "have been left with an inconsistent configuration, "
		       "please check.\n", dev->name);

L
Linus Torvalds 已提交
897 898
	if (send_addr_notify)
		call_netdevice_notifiers(NETDEV_CHANGEADDR, dev);
899 900
	return err;
}
L
Linus Torvalds 已提交
901

902 903
static int rtnl_setlink(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
{
904
	struct net *net = sock_net(skb->sk);
905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922
	struct ifinfomsg *ifm;
	struct net_device *dev;
	int err;
	struct nlattr *tb[IFLA_MAX+1];
	char ifname[IFNAMSIZ];

	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy);
	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)
923
		dev = __dev_get_by_index(net, ifm->ifi_index);
924
	else if (tb[IFLA_IFNAME])
925
		dev = __dev_get_by_name(net, ifname);
926 927 928 929 930 931 932 933
	else
		goto errout;

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

E
Eric Dumazet 已提交
934 935
	err = validate_linkmsg(dev, tb);
	if (err < 0)
936
		goto errout;
937

P
Patrick McHardy 已提交
938
	err = do_setlink(dev, ifm, tb, ifname, 0);
939
errout:
L
Linus Torvalds 已提交
940 941 942
	return err;
}

P
Patrick McHardy 已提交
943 944
static int rtnl_dellink(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
{
945
	struct net *net = sock_net(skb->sk);
P
Patrick McHardy 已提交
946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961
	const struct rtnl_link_ops *ops;
	struct net_device *dev;
	struct ifinfomsg *ifm;
	char ifname[IFNAMSIZ];
	struct nlattr *tb[IFLA_MAX+1];
	int err;

	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy);
	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)
962
		dev = __dev_get_by_index(net, ifm->ifi_index);
P
Patrick McHardy 已提交
963
	else if (tb[IFLA_IFNAME])
964
		dev = __dev_get_by_name(net, ifname);
P
Patrick McHardy 已提交
965 966 967 968 969 970 971 972 973 974
	else
		return -EINVAL;

	if (!dev)
		return -ENODEV;

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

975
	ops->dellink(dev, NULL);
P
Patrick McHardy 已提交
976 977 978
	return 0;
}

979
struct net_device *rtnl_create_link(struct net *net, char *ifname,
980 981 982 983
		const struct rtnl_link_ops *ops, struct nlattr *tb[])
{
	int err;
	struct net_device *dev;
E
Eric Dumazet 已提交
984 985
	unsigned int num_queues = 1;
	unsigned int real_num_queues = 1;
986

E
Eric Dumazet 已提交
987
	if (ops->get_tx_queues) {
E
Eric Dumazet 已提交
988 989
		err = ops->get_tx_queues(net, tb, &num_queues,
					 &real_num_queues);
E
Eric Dumazet 已提交
990 991 992
		if (err)
			goto err;
	}
993
	err = -ENOMEM;
E
Eric Dumazet 已提交
994
	dev = alloc_netdev_mq(ops->priv_size, ifname, ops->setup, num_queues);
995 996 997
	if (!dev)
		goto err;

E
Eric Dumazet 已提交
998
	dev->real_num_tx_queues = real_num_queues;
999 1000 1001 1002 1003 1004
	if (strchr(dev->name, '%')) {
		err = dev_alloc_name(dev, dev->name);
		if (err < 0)
			goto err_free;
	}

1005
	dev_net_set(dev, net);
1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018
	dev->rtnl_link_ops = ops;

	if (tb[IFLA_MTU])
		dev->mtu = nla_get_u32(tb[IFLA_MTU]);
	if (tb[IFLA_ADDRESS])
		memcpy(dev->dev_addr, nla_data(tb[IFLA_ADDRESS]),
				nla_len(tb[IFLA_ADDRESS]));
	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])
1019
		set_operstate(dev, nla_get_u8(tb[IFLA_OPERSTATE]));
1020 1021 1022 1023 1024 1025 1026 1027 1028 1029
	if (tb[IFLA_LINKMODE])
		dev->link_mode = nla_get_u8(tb[IFLA_LINKMODE]);

	return dev;

err_free:
	free_netdev(dev);
err:
	return ERR_PTR(err);
}
E
Eric Dumazet 已提交
1030
EXPORT_SYMBOL(rtnl_create_link);
1031

P
Patrick McHardy 已提交
1032 1033
static int rtnl_newlink(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
{
1034
	struct net *net = sock_net(skb->sk);
P
Patrick McHardy 已提交
1035 1036 1037 1038 1039 1040 1041 1042 1043
	const struct rtnl_link_ops *ops;
	struct net_device *dev;
	struct ifinfomsg *ifm;
	char kind[MODULE_NAME_LEN];
	char ifname[IFNAMSIZ];
	struct nlattr *tb[IFLA_MAX+1];
	struct nlattr *linkinfo[IFLA_INFO_MAX+1];
	int err;

1044
#ifdef CONFIG_MODULES
P
Patrick McHardy 已提交
1045
replay:
1046
#endif
P
Patrick McHardy 已提交
1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy);
	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)
1058
		dev = __dev_get_by_index(net, ifm->ifi_index);
P
Patrick McHardy 已提交
1059
	else if (ifname[0])
1060
		dev = __dev_get_by_name(net, ifname);
P
Patrick McHardy 已提交
1061 1062 1063
	else
		dev = NULL;

E
Eric Dumazet 已提交
1064 1065
	err = validate_linkmsg(dev, tb);
	if (err < 0)
1066 1067
		return err;

P
Patrick McHardy 已提交
1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129
	if (tb[IFLA_LINKINFO]) {
		err = nla_parse_nested(linkinfo, IFLA_INFO_MAX,
				       tb[IFLA_LINKINFO], ifla_info_policy);
		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) {
		struct nlattr *attr[ops ? ops->maxtype + 1 : 0], **data = NULL;

		if (ops) {
			if (ops->maxtype && linkinfo[IFLA_INFO_DATA]) {
				err = nla_parse_nested(attr, ops->maxtype,
						       linkinfo[IFLA_INFO_DATA],
						       ops->policy);
				if (err < 0)
					return err;
				data = attr;
			}
			if (ops->validate) {
				err = ops->validate(tb, data);
				if (err < 0)
					return err;
			}
		}

		if (dev) {
			int modified = 0;

			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;

				err = ops->changelink(dev, tb, data);
				if (err < 0)
					return err;
				modified = 1;
			}

			return do_setlink(dev, ifm, tb, ifname, modified);
		}

		if (!(nlh->nlmsg_flags & NLM_F_CREATE))
			return -ENODEV;

		if (ifm->ifi_index || ifm->ifi_flags || ifm->ifi_change)
			return -EOPNOTSUPP;
1130
		if (tb[IFLA_MAP] || tb[IFLA_MASTER] || tb[IFLA_PROTINFO])
P
Patrick McHardy 已提交
1131 1132 1133
			return -EOPNOTSUPP;

		if (!ops) {
1134
#ifdef CONFIG_MODULES
P
Patrick McHardy 已提交
1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148
			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;
		}

		if (!ifname[0])
			snprintf(ifname, IFNAMSIZ, "%s%%d", ops->kind);
1149

1150
		dev = rtnl_create_link(net, ifname, ops, tb);
1151 1152 1153 1154

		if (IS_ERR(dev))
			err = PTR_ERR(dev);
		else if (ops->newlink)
1155 1156 1157
			err = ops->newlink(dev, tb, data);
		else
			err = register_netdevice(dev);
1158 1159

		if (err < 0 && !IS_ERR(dev))
P
Patrick McHardy 已提交
1160 1161 1162 1163 1164
			free_netdev(dev);
		return err;
	}
}

1165
static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
1166
{
1167
	struct net *net = sock_net(skb->sk);
1168
	struct ifinfomsg *ifm;
1169
	char ifname[IFNAMSIZ];
1170 1171 1172
	struct nlattr *tb[IFLA_MAX+1];
	struct net_device *dev = NULL;
	struct sk_buff *nskb;
1173
	int err;
1174

1175 1176
	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy);
	if (err < 0)
1177
		return err;
1178

1179 1180 1181
	if (tb[IFLA_IFNAME])
		nla_strlcpy(ifname, tb[IFLA_IFNAME], IFNAMSIZ);

1182
	ifm = nlmsg_data(nlh);
1183 1184 1185 1186 1187
	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
1188 1189
		return -EINVAL;

1190 1191 1192
	if (dev == NULL)
		return -ENODEV;

P
Patrick McHardy 已提交
1193
	nskb = nlmsg_new(if_nlmsg_size(dev), GFP_KERNEL);
1194 1195
	if (nskb == NULL)
		return -ENOBUFS;
1196

1197 1198
	err = rtnl_fill_ifinfo(nskb, dev, RTM_NEWLINK, NETLINK_CB(skb).pid,
			       nlh->nlmsg_seq, 0, 0);
1199 1200 1201 1202
	if (err < 0) {
		/* -EMSGSIZE implies BUG in if_nlmsg_size */
		WARN_ON(err == -EMSGSIZE);
		kfree_skb(nskb);
1203 1204
	} else
		err = rtnl_unicast(nskb, net, NETLINK_CB(skb).pid);
1205

1206
	return err;
1207 1208
}

A
Adrian Bunk 已提交
1209
static int rtnl_dump_all(struct sk_buff *skb, struct netlink_callback *cb)
L
Linus Torvalds 已提交
1210 1211 1212 1213 1214 1215
{
	int idx;
	int s_idx = cb->family;

	if (s_idx == 0)
		s_idx = 1;
E
Eric Dumazet 已提交
1216
	for (idx = 1; idx < NPROTO; idx++) {
L
Linus Torvalds 已提交
1217 1218 1219
		int type = cb->nlh->nlmsg_type-RTM_BASE;
		if (idx < s_idx || idx == PF_PACKET)
			continue;
1220 1221
		if (rtnl_msg_handlers[idx] == NULL ||
		    rtnl_msg_handlers[idx][type].dumpit == NULL)
L
Linus Torvalds 已提交
1222 1223 1224
			continue;
		if (idx > s_idx)
			memset(&cb->args[0], 0, sizeof(cb->args));
1225
		if (rtnl_msg_handlers[idx][type].dumpit(skb, cb))
L
Linus Torvalds 已提交
1226 1227 1228 1229 1230 1231 1232 1233 1234
			break;
	}
	cb->family = idx;

	return skb->len;
}

void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change)
{
1235
	struct net *net = dev_net(dev);
L
Linus Torvalds 已提交
1236
	struct sk_buff *skb;
1237
	int err = -ENOBUFS;
L
Linus Torvalds 已提交
1238

P
Patrick McHardy 已提交
1239
	skb = nlmsg_new(if_nlmsg_size(dev), GFP_KERNEL);
1240 1241
	if (skb == NULL)
		goto errout;
L
Linus Torvalds 已提交
1242

1243
	err = rtnl_fill_ifinfo(skb, dev, type, 0, 0, change, 0);
1244 1245 1246 1247 1248 1249
	if (err < 0) {
		/* -EMSGSIZE implies BUG in if_nlmsg_size() */
		WARN_ON(err == -EMSGSIZE);
		kfree_skb(skb);
		goto errout;
	}
1250 1251
	rtnl_notify(skb, net, 0, RTNLGRP_LINK, NULL, GFP_KERNEL);
	return;
1252 1253
errout:
	if (err < 0)
1254
		rtnl_set_sk_err(net, RTNLGRP_LINK, err);
L
Linus Torvalds 已提交
1255 1256 1257 1258 1259 1260 1261 1262
}

/* Protected by RTNL sempahore.  */
static struct rtattr **rta_buf;
static int rtattr_max;

/* Process one rtnetlink message. */

1263
static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
L
Linus Torvalds 已提交
1264
{
1265
	struct net *net = sock_net(skb->sk);
1266
	rtnl_doit_func doit;
L
Linus Torvalds 已提交
1267 1268 1269 1270
	int sz_idx, kind;
	int min_len;
	int family;
	int type;
1271
	int err;
L
Linus Torvalds 已提交
1272 1273 1274

	type = nlh->nlmsg_type;
	if (type > RTM_MAX)
1275
		return -EOPNOTSUPP;
L
Linus Torvalds 已提交
1276 1277 1278 1279 1280 1281 1282

	type -= RTM_BASE;

	/* All the messages must have at least 1 byte length */
	if (nlh->nlmsg_len < NLMSG_LENGTH(sizeof(struct rtgenmsg)))
		return 0;

E
Eric Dumazet 已提交
1283
	family = ((struct rtgenmsg *)NLMSG_DATA(nlh))->rtgen_family;
1284 1285
	if (family >= NPROTO)
		return -EAFNOSUPPORT;
L
Linus Torvalds 已提交
1286 1287 1288 1289

	sz_idx = type>>2;
	kind = type&3;

1290 1291
	if (kind != 2 && security_netlink_recv(skb, CAP_NET_ADMIN))
		return -EPERM;
L
Linus Torvalds 已提交
1292 1293

	if (kind == 2 && nlh->nlmsg_flags&NLM_F_DUMP) {
1294
		struct sock *rtnl;
1295
		rtnl_dumpit_func dumpit;
L
Linus Torvalds 已提交
1296

1297 1298
		dumpit = rtnl_get_dumpit(family, type);
		if (dumpit == NULL)
1299
			return -EOPNOTSUPP;
1300

1301
		__rtnl_unlock();
1302
		rtnl = net->rtnl;
1303 1304 1305
		err = netlink_dump_start(rtnl, skb, nlh, dumpit, NULL);
		rtnl_lock();
		return err;
L
Linus Torvalds 已提交
1306 1307 1308 1309 1310 1311
	}

	memset(rta_buf, 0, (rtattr_max * sizeof(struct rtattr *)));

	min_len = rtm_min[sz_idx];
	if (nlh->nlmsg_len < min_len)
1312
		return -EINVAL;
L
Linus Torvalds 已提交
1313 1314 1315

	if (nlh->nlmsg_len > min_len) {
		int attrlen = nlh->nlmsg_len - NLMSG_ALIGN(min_len);
E
Eric Dumazet 已提交
1316
		struct rtattr *attr = (void *)nlh + NLMSG_ALIGN(min_len);
L
Linus Torvalds 已提交
1317 1318 1319 1320 1321

		while (RTA_OK(attr, attrlen)) {
			unsigned flavor = attr->rta_type;
			if (flavor) {
				if (flavor > rta_max[sz_idx])
1322
					return -EINVAL;
L
Linus Torvalds 已提交
1323 1324 1325 1326 1327 1328
				rta_buf[flavor-1] = attr;
			}
			attr = RTA_NEXT(attr, attrlen);
		}
	}

1329 1330
	doit = rtnl_get_doit(family, type);
	if (doit == NULL)
1331
		return -EOPNOTSUPP;
L
Linus Torvalds 已提交
1332

1333
	return doit(skb, nlh, (void *)&rta_buf[0]);
L
Linus Torvalds 已提交
1334 1335
}

1336
static void rtnetlink_rcv(struct sk_buff *skb)
L
Linus Torvalds 已提交
1337
{
1338 1339 1340
	rtnl_lock();
	netlink_rcv_skb(skb, &rtnetlink_rcv_msg);
	rtnl_unlock();
L
Linus Torvalds 已提交
1341 1342 1343 1344 1345
}

static int rtnetlink_event(struct notifier_block *this, unsigned long event, void *ptr)
{
	struct net_device *dev = ptr;
1346

L
Linus Torvalds 已提交
1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371
	switch (event) {
	case NETDEV_UNREGISTER:
		rtmsg_ifinfo(RTM_DELLINK, dev, ~0U);
		break;
	case NETDEV_REGISTER:
		rtmsg_ifinfo(RTM_NEWLINK, dev, ~0U);
		break;
	case NETDEV_UP:
	case NETDEV_DOWN:
		rtmsg_ifinfo(RTM_NEWLINK, dev, IFF_UP|IFF_RUNNING);
		break;
	case NETDEV_CHANGE:
	case NETDEV_GOING_DOWN:
		break;
	default:
		rtmsg_ifinfo(RTM_NEWLINK, dev, 0);
		break;
	}
	return NOTIFY_DONE;
}

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

1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385

static int rtnetlink_net_init(struct net *net)
{
	struct sock *sk;
	sk = netlink_kernel_create(net, NETLINK_ROUTE, RTNLGRP_MAX,
				   rtnetlink_rcv, &rtnl_mutex, THIS_MODULE);
	if (!sk)
		return -ENOMEM;
	net->rtnl = sk;
	return 0;
}

static void rtnetlink_net_exit(struct net *net)
{
1386 1387
	netlink_kernel_release(net->rtnl);
	net->rtnl = NULL;
1388 1389 1390 1391 1392 1393 1394
}

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

L
Linus Torvalds 已提交
1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406
void __init rtnetlink_init(void)
{
	int i;

	rtattr_max = 0;
	for (i = 0; i < ARRAY_SIZE(rta_max); i++)
		if (rta_max[i] > rtattr_max)
			rtattr_max = rta_max[i];
	rta_buf = kmalloc(rtattr_max * sizeof(struct rtattr *), GFP_KERNEL);
	if (!rta_buf)
		panic("rtnetlink_init: cannot allocate rta_buf\n");

1407
	if (register_pernet_subsys(&rtnetlink_net_ops))
L
Linus Torvalds 已提交
1408
		panic("rtnetlink_init: cannot initialize rtnetlink\n");
1409

L
Linus Torvalds 已提交
1410 1411
	netlink_set_nonroot(NETLINK_ROUTE, NL_NONROOT_RECV);
	register_netdevice_notifier(&rtnetlink_dev_notifier);
1412 1413 1414

	rtnl_register(PF_UNSPEC, RTM_GETLINK, rtnl_getlink, rtnl_dump_ifinfo);
	rtnl_register(PF_UNSPEC, RTM_SETLINK, rtnl_setlink, NULL);
P
Patrick McHardy 已提交
1415 1416
	rtnl_register(PF_UNSPEC, RTM_NEWLINK, rtnl_newlink, NULL);
	rtnl_register(PF_UNSPEC, RTM_DELLINK, rtnl_dellink, NULL);
1417 1418 1419

	rtnl_register(PF_UNSPEC, RTM_GETADDR, NULL, rtnl_dump_all);
	rtnl_register(PF_UNSPEC, RTM_GETROUTE, NULL, rtnl_dump_all);
L
Linus Torvalds 已提交
1420 1421
}