ip6_fib.c 29.8 KB
Newer Older
L
Linus Torvalds 已提交
1
/*
2
 *	Linux INET6 implementation
L
Linus Torvalds 已提交
3 4 5
 *	Forwarding Information Database
 *
 *	Authors:
6
 *	Pedro Roque		<roque@di.fc.ul.pt>
L
Linus Torvalds 已提交
7 8 9 10 11 12 13 14 15 16 17 18 19 20
 *
 *	$Id: ip6_fib.c,v 1.25 2001/10/31 21:55:55 davem Exp $
 *
 *	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.
 */

/*
 * 	Changes:
 * 	Yuji SEKIYA @USAGI:	Support default route on router node;
 * 				remove ip6_null_entry from the top of
 * 				routing table.
21
 * 	Ville Nuorvala:		Fixed routing subtrees.
L
Linus Torvalds 已提交
22 23 24 25 26 27 28 29
 */
#include <linux/errno.h>
#include <linux/types.h>
#include <linux/net.h>
#include <linux/route.h>
#include <linux/netdevice.h>
#include <linux/in6.h>
#include <linux/init.h>
T
Thomas Graf 已提交
30
#include <linux/list.h>
L
Linus Torvalds 已提交
31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

#ifdef 	CONFIG_PROC_FS
#include <linux/proc_fs.h>
#endif

#include <net/ipv6.h>
#include <net/ndisc.h>
#include <net/addrconf.h>

#include <net/ip6_fib.h>
#include <net/ip6_route.h>

#define RT6_DEBUG 2

#if RT6_DEBUG >= 3
#define RT6_TRACE(x...) printk(KERN_DEBUG x)
#else
#define RT6_TRACE(x...) do { ; } while (0)
#endif

struct rt6_statistics	rt6_stats;

53
static struct kmem_cache * fib6_node_kmem __read_mostly;
L
Linus Torvalds 已提交
54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72

enum fib_walk_state_t
{
#ifdef CONFIG_IPV6_SUBTREES
	FWS_S,
#endif
	FWS_L,
	FWS_R,
	FWS_C,
	FWS_U
};

struct fib6_cleaner_t
{
	struct fib6_walker_t w;
	int (*func)(struct rt6_info *, void *arg);
	void *arg;
};

A
Adrian Bunk 已提交
73
static DEFINE_RWLOCK(fib6_walker_lock);
L
Linus Torvalds 已提交
74 75 76 77 78 79 80 81

#ifdef CONFIG_IPV6_SUBTREES
#define FWS_INIT FWS_S
#else
#define FWS_INIT FWS_L
#endif

static void fib6_prune_clones(struct fib6_node *fn, struct rt6_info *rt);
82
static struct rt6_info * fib6_find_prefix(struct fib6_node *fn);
L
Linus Torvalds 已提交
83
static struct fib6_node * fib6_repair_tree(struct fib6_node *fn);
A
Adrian Bunk 已提交
84 85
static int fib6_walk(struct fib6_walker_t *w);
static int fib6_walk_continue(struct fib6_walker_t *w);
L
Linus Torvalds 已提交
86 87 88 89 90 91 92 93 94 95

/*
 *	A routing update causes an increase of the serial number on the
 *	affected subtree. This allows for cached routes to be asynchronously
 *	tested when modifications are made to the destination cache as a
 *	result of redirects, path MTU changes, etc.
 */

static __u32 rt_sernum;

96
static DEFINE_TIMER(ip6_fib_timer, fib6_run_gc, 0, 0);
L
Linus Torvalds 已提交
97

A
Adrian Bunk 已提交
98
static struct fib6_walker_t fib6_walker_list = {
L
Linus Torvalds 已提交
99
	.prev	= &fib6_walker_list,
100
	.next	= &fib6_walker_list,
L
Linus Torvalds 已提交
101 102 103 104
};

#define FOR_WALKERS(w) for ((w)=fib6_walker_list.next; (w) != &fib6_walker_list; (w)=(w)->next)

A
Adrian Bunk 已提交
105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122
static inline void fib6_walker_link(struct fib6_walker_t *w)
{
	write_lock_bh(&fib6_walker_lock);
	w->next = fib6_walker_list.next;
	w->prev = &fib6_walker_list;
	w->next->prev = w;
	w->prev->next = w;
	write_unlock_bh(&fib6_walker_lock);
}

static inline void fib6_walker_unlink(struct fib6_walker_t *w)
{
	write_lock_bh(&fib6_walker_lock);
	w->next->prev = w->prev;
	w->prev->next = w->next;
	w->prev = w->next = w;
	write_unlock_bh(&fib6_walker_lock);
}
L
Linus Torvalds 已提交
123 124 125 126 127 128 129 130 131 132 133
static __inline__ u32 fib6_new_sernum(void)
{
	u32 n = ++rt_sernum;
	if ((__s32)n <= 0)
		rt_sernum = n = 1;
	return n;
}

/*
 *	Auxiliary address test functions for the radix tree.
 *
134
 *	These assume a 32bit processor (although it will work on
L
Linus Torvalds 已提交
135 136 137 138 139 140 141
 *	64bit processors)
 */

/*
 *	test bit
 */

A
Al Viro 已提交
142
static __inline__ __be32 addr_bit_set(void *token, int fn_bit)
L
Linus Torvalds 已提交
143
{
A
Al Viro 已提交
144
	__be32 *addr = token;
L
Linus Torvalds 已提交
145 146 147 148 149 150 151 152

	return htonl(1 << ((~fn_bit)&0x1F)) & addr[fn_bit>>5];
}

static __inline__ struct fib6_node * node_alloc(void)
{
	struct fib6_node *fn;

153
	fn = kmem_cache_zalloc(fib6_node_kmem, GFP_ATOMIC);
L
Linus Torvalds 已提交
154 155 156 157 158 159 160 161 162 163 164 165 166 167 168

	return fn;
}

static __inline__ void node_free(struct fib6_node * fn)
{
	kmem_cache_free(fib6_node_kmem, fn);
}

static __inline__ void rt6_release(struct rt6_info *rt)
{
	if (atomic_dec_and_test(&rt->rt6i_ref))
		dst_free(&rt->u.dst);
}

T
Thomas Graf 已提交
169 170 171 172 173 174 175 176 177
static struct fib6_table fib6_main_tbl = {
	.tb6_id		= RT6_TABLE_MAIN,
	.tb6_root	= {
		.leaf		= &ip6_null_entry,
		.fn_flags	= RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO,
	},
};

#ifdef CONFIG_IPV6_MULTIPLE_TABLES
178 179 180 181 182 183 184 185 186 187
#define FIB_TABLE_HASHSZ 256
#else
#define FIB_TABLE_HASHSZ 1
#endif
static struct hlist_head fib_table_hash[FIB_TABLE_HASHSZ];

static void fib6_link_table(struct fib6_table *tb)
{
	unsigned int h;

188 189 190 191 192 193
	/*
	 * Initialize table lock at a single place to give lockdep a key,
	 * tables aren't visible prior to being linked to the list.
	 */
	rwlock_init(&tb->tb6_lock);

194 195 196 197 198 199 200 201
	h = tb->tb6_id & (FIB_TABLE_HASHSZ - 1);

	/*
	 * No protection necessary, this is the only list mutatation
	 * operation, tables never disappear once they exist.
	 */
	hlist_add_head_rcu(&tb->tb6_hlist, &fib_table_hash[h]);
}
T
Thomas Graf 已提交
202

203
#ifdef CONFIG_IPV6_MULTIPLE_TABLES
T
Thomas Graf 已提交
204 205 206 207 208 209 210 211
static struct fib6_table fib6_local_tbl = {
	.tb6_id		= RT6_TABLE_LOCAL,
	.tb6_root 	= {
		.leaf		= &ip6_null_entry,
		.fn_flags	= RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO,
	},
};

T
Thomas Graf 已提交
212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266
static struct fib6_table *fib6_alloc_table(u32 id)
{
	struct fib6_table *table;

	table = kzalloc(sizeof(*table), GFP_ATOMIC);
	if (table != NULL) {
		table->tb6_id = id;
		table->tb6_root.leaf = &ip6_null_entry;
		table->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO;
	}

	return table;
}

struct fib6_table *fib6_new_table(u32 id)
{
	struct fib6_table *tb;

	if (id == 0)
		id = RT6_TABLE_MAIN;
	tb = fib6_get_table(id);
	if (tb)
		return tb;

	tb = fib6_alloc_table(id);
	if (tb != NULL)
		fib6_link_table(tb);

	return tb;
}

struct fib6_table *fib6_get_table(u32 id)
{
	struct fib6_table *tb;
	struct hlist_node *node;
	unsigned int h;

	if (id == 0)
		id = RT6_TABLE_MAIN;
	h = id & (FIB_TABLE_HASHSZ - 1);
	rcu_read_lock();
	hlist_for_each_entry_rcu(tb, node, &fib_table_hash[h], tb6_hlist) {
		if (tb->tb6_id == id) {
			rcu_read_unlock();
			return tb;
		}
	}
	rcu_read_unlock();

	return NULL;
}

static void __init fib6_tables_init(void)
{
	fib6_link_table(&fib6_main_tbl);
T
Thomas Graf 已提交
267
	fib6_link_table(&fib6_local_tbl);
T
Thomas Graf 已提交
268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289
}

#else

struct fib6_table *fib6_new_table(u32 id)
{
	return fib6_get_table(id);
}

struct fib6_table *fib6_get_table(u32 id)
{
	return &fib6_main_tbl;
}

struct dst_entry *fib6_rule_lookup(struct flowi *fl, int flags,
				   pol_lookup_t lookup)
{
	return (struct dst_entry *) lookup(&fib6_main_tbl, fl, flags);
}

static void __init fib6_tables_init(void)
{
290
	fib6_link_table(&fib6_main_tbl);
T
Thomas Graf 已提交
291 292 293 294
}

#endif

295 296 297 298 299
static int fib6_dump_node(struct fib6_walker_t *w)
{
	int res;
	struct rt6_info *rt;

300
	for (rt = w->leaf; rt; rt = rt->u.dst.rt6_next) {
301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361
		res = rt6_dump_route(rt, w->args);
		if (res < 0) {
			/* Frame is full, suspend walking */
			w->leaf = rt;
			return 1;
		}
		BUG_TRAP(res!=0);
	}
	w->leaf = NULL;
	return 0;
}

static void fib6_dump_end(struct netlink_callback *cb)
{
	struct fib6_walker_t *w = (void*)cb->args[2];

	if (w) {
		cb->args[2] = 0;
		kfree(w);
	}
	cb->done = (void*)cb->args[3];
	cb->args[1] = 3;
}

static int fib6_dump_done(struct netlink_callback *cb)
{
	fib6_dump_end(cb);
	return cb->done ? cb->done(cb) : 0;
}

static int fib6_dump_table(struct fib6_table *table, struct sk_buff *skb,
			   struct netlink_callback *cb)
{
	struct fib6_walker_t *w;
	int res;

	w = (void *)cb->args[2];
	w->root = &table->tb6_root;

	if (cb->args[4] == 0) {
		read_lock_bh(&table->tb6_lock);
		res = fib6_walk(w);
		read_unlock_bh(&table->tb6_lock);
		if (res > 0)
			cb->args[4] = 1;
	} else {
		read_lock_bh(&table->tb6_lock);
		res = fib6_walk_continue(w);
		read_unlock_bh(&table->tb6_lock);
		if (res != 0) {
			if (res < 0)
				fib6_walker_unlink(w);
			goto end;
		}
		fib6_walker_unlink(w);
		cb->args[4] = 0;
	}
end:
	return res;
}

362
static int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
363
{
364
	struct net *net = skb->sk->sk_net;
365 366 367 368 369 370 371 372
	unsigned int h, s_h;
	unsigned int e = 0, s_e;
	struct rt6_rtnl_dump_arg arg;
	struct fib6_walker_t *w;
	struct fib6_table *tb;
	struct hlist_node *node;
	int res = 0;

373 374 375
	if (net != &init_net)
		return 0;

376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422
	s_h = cb->args[0];
	s_e = cb->args[1];

	w = (void *)cb->args[2];
	if (w == NULL) {
		/* New dump:
		 *
		 * 1. hook callback destructor.
		 */
		cb->args[3] = (long)cb->done;
		cb->done = fib6_dump_done;

		/*
		 * 2. allocate and initialize walker.
		 */
		w = kzalloc(sizeof(*w), GFP_ATOMIC);
		if (w == NULL)
			return -ENOMEM;
		w->func = fib6_dump_node;
		cb->args[2] = (long)w;
	}

	arg.skb = skb;
	arg.cb = cb;
	w->args = &arg;

	for (h = s_h; h < FIB_TABLE_HASHSZ; h++, s_e = 0) {
		e = 0;
		hlist_for_each_entry(tb, node, &fib_table_hash[h], tb6_hlist) {
			if (e < s_e)
				goto next;
			res = fib6_dump_table(tb, skb, cb);
			if (res != 0)
				goto out;
next:
			e++;
		}
	}
out:
	cb->args[1] = e;
	cb->args[0] = h;

	res = res < 0 ? res : skb->len;
	if (res <= 0)
		fib6_dump_end(cb);
	return res;
}
L
Linus Torvalds 已提交
423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439

/*
 *	Routing Table
 *
 *	return the appropriate node for a routing tree "add" operation
 *	by either creating and inserting or by returning an existing
 *	node.
 */

static struct fib6_node * fib6_add_1(struct fib6_node *root, void *addr,
				     int addrlen, int plen,
				     int offset)
{
	struct fib6_node *fn, *in, *ln;
	struct fib6_node *pn = NULL;
	struct rt6key *key;
	int	bit;
440
	__be32	dir = 0;
L
Linus Torvalds 已提交
441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457
	__u32	sernum = fib6_new_sernum();

	RT6_TRACE("fib6_add_1\n");

	/* insert node in tree */

	fn = root;

	do {
		key = (struct rt6key *)((u8 *)fn->leaf + offset);

		/*
		 *	Prefix match
		 */
		if (plen < fn->fn_bit ||
		    !ipv6_prefix_equal(&key->addr, addr, fn->fn_bit))
			goto insert_above;
458

L
Linus Torvalds 已提交
459 460 461
		/*
		 *	Exact match ?
		 */
462

L
Linus Torvalds 已提交
463 464 465 466 467 468
		if (plen == fn->fn_bit) {
			/* clean up an intermediate node */
			if ((fn->fn_flags & RTN_RTINFO) == 0) {
				rt6_release(fn->leaf);
				fn->leaf = NULL;
			}
469

L
Linus Torvalds 已提交
470
			fn->fn_sernum = sernum;
471

L
Linus Torvalds 已提交
472 473 474 475 476 477
			return fn;
		}

		/*
		 *	We have more bits to go
		 */
478

L
Linus Torvalds 已提交
479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495
		/* Try to walk down on tree. */
		fn->fn_sernum = sernum;
		dir = addr_bit_set(addr, fn->fn_bit);
		pn = fn;
		fn = dir ? fn->right: fn->left;
	} while (fn);

	/*
	 *	We walked to the bottom of tree.
	 *	Create new leaf node without children.
	 */

	ln = node_alloc();

	if (ln == NULL)
		return NULL;
	ln->fn_bit = plen;
496

L
Linus Torvalds 已提交
497 498 499 500 501 502 503 504 505 506 507 508 509
	ln->parent = pn;
	ln->fn_sernum = sernum;

	if (dir)
		pn->right = ln;
	else
		pn->left  = ln;

	return ln;


insert_above:
	/*
510
	 * split since we don't have a common prefix anymore or
L
Linus Torvalds 已提交
511 512 513 514 515 516 517 518 519 520
	 * we have a less significant route.
	 * we've to insert an intermediate node on the list
	 * this new node will point to the one we need to create
	 * and the current
	 */

	pn = fn->parent;

	/* find 1st bit in difference between the 2 addrs.

521
	   See comment in __ipv6_addr_diff: bit may be an invalid value,
L
Linus Torvalds 已提交
522 523
	   but if it is >= plen, the value is ignored in any case.
	 */
524

525
	bit = __ipv6_addr_diff(addr, &key->addr, addrlen);
L
Linus Torvalds 已提交
526

527 528
	/*
	 *		(intermediate)[in]
L
Linus Torvalds 已提交
529 530 531 532 533 534
	 *	          /	   \
	 *	(new leaf node)[ln] (old node)[fn]
	 */
	if (plen > bit) {
		in = node_alloc();
		ln = node_alloc();
535

L
Linus Torvalds 已提交
536 537 538 539 540 541 542 543
		if (in == NULL || ln == NULL) {
			if (in)
				node_free(in);
			if (ln)
				node_free(ln);
			return NULL;
		}

544 545
		/*
		 * new intermediate node.
L
Linus Torvalds 已提交
546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581
		 * RTN_RTINFO will
		 * be off since that an address that chooses one of
		 * the branches would not match less specific routes
		 * in the other branch
		 */

		in->fn_bit = bit;

		in->parent = pn;
		in->leaf = fn->leaf;
		atomic_inc(&in->leaf->rt6i_ref);

		in->fn_sernum = sernum;

		/* update parent pointer */
		if (dir)
			pn->right = in;
		else
			pn->left  = in;

		ln->fn_bit = plen;

		ln->parent = in;
		fn->parent = in;

		ln->fn_sernum = sernum;

		if (addr_bit_set(addr, bit)) {
			in->right = ln;
			in->left  = fn;
		} else {
			in->left  = ln;
			in->right = fn;
		}
	} else { /* plen <= bit */

582
		/*
L
Linus Torvalds 已提交
583 584 585 586 587 588 589 590 591 592 593 594 595 596 597
		 *		(new leaf node)[ln]
		 *	          /	   \
		 *	     (old node)[fn] NULL
		 */

		ln = node_alloc();

		if (ln == NULL)
			return NULL;

		ln->fn_bit = plen;

		ln->parent = pn;

		ln->fn_sernum = sernum;
598

L
Linus Torvalds 已提交
599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618
		if (dir)
			pn->right = ln;
		else
			pn->left  = ln;

		if (addr_bit_set(&key->addr, plen))
			ln->right = fn;
		else
			ln->left  = fn;

		fn->parent = ln;
	}
	return ln;
}

/*
 *	Insert routing information in a node.
 */

static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt,
619
			    struct nl_info *info)
L
Linus Torvalds 已提交
620 621 622 623 624 625
{
	struct rt6_info *iter = NULL;
	struct rt6_info **ins;

	ins = &fn->leaf;

626
	for (iter = fn->leaf; iter; iter=iter->u.dst.rt6_next) {
L
Linus Torvalds 已提交
627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653
		/*
		 *	Search for duplicates
		 */

		if (iter->rt6i_metric == rt->rt6i_metric) {
			/*
			 *	Same priority level
			 */

			if (iter->rt6i_dev == rt->rt6i_dev &&
			    iter->rt6i_idev == rt->rt6i_idev &&
			    ipv6_addr_equal(&iter->rt6i_gateway,
					    &rt->rt6i_gateway)) {
				if (!(iter->rt6i_flags&RTF_EXPIRES))
					return -EEXIST;
				iter->rt6i_expires = rt->rt6i_expires;
				if (!(rt->rt6i_flags&RTF_EXPIRES)) {
					iter->rt6i_flags &= ~RTF_EXPIRES;
					iter->rt6i_expires = 0;
				}
				return -EEXIST;
			}
		}

		if (iter->rt6i_metric > rt->rt6i_metric)
			break;

654
		ins = &iter->u.dst.rt6_next;
L
Linus Torvalds 已提交
655 656
	}

657 658 659 660
	/* Reset round-robin state, if necessary */
	if (ins == &fn->leaf)
		fn->rr_ptr = NULL;

L
Linus Torvalds 已提交
661 662 663 664
	/*
	 *	insert node
	 */

665
	rt->u.dst.rt6_next = iter;
L
Linus Torvalds 已提交
666 667 668
	*ins = rt;
	rt->rt6i_node = fn;
	atomic_inc(&rt->rt6i_ref);
669
	inet6_rt_notify(RTM_NEWROUTE, rt, info);
L
Linus Torvalds 已提交
670 671 672 673 674 675 676 677 678 679 680 681 682 683
	rt6_stats.fib_rt_entries++;

	if ((fn->fn_flags & RTN_RTINFO) == 0) {
		rt6_stats.fib_route_nodes++;
		fn->fn_flags |= RTN_RTINFO;
	}

	return 0;
}

static __inline__ void fib6_start_gc(struct rt6_info *rt)
{
	if (ip6_fib_timer.expires == 0 &&
	    (rt->rt6i_flags & (RTF_EXPIRES|RTF_CACHE)))
684 685
		mod_timer(&ip6_fib_timer, jiffies +
			  init_net.ipv6.sysctl.ip6_rt_gc_interval);
L
Linus Torvalds 已提交
686 687 688 689 690
}

void fib6_force_start_gc(void)
{
	if (ip6_fib_timer.expires == 0)
691 692
		mod_timer(&ip6_fib_timer, jiffies +
			  init_net.ipv6.sysctl.ip6_rt_gc_interval);
L
Linus Torvalds 已提交
693 694 695 696 697 698 699 700
}

/*
 *	Add routing information to the routing tree.
 *	<destination addr>/<source addr>
 *	with source addr info in sub-trees
 */

701
int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nl_info *info)
L
Linus Torvalds 已提交
702
{
703
	struct fib6_node *fn, *pn = NULL;
L
Linus Torvalds 已提交
704 705 706 707 708 709 710 711
	int err = -ENOMEM;

	fn = fib6_add_1(root, &rt->rt6i_dst.addr, sizeof(struct in6_addr),
			rt->rt6i_dst.plen, offsetof(struct rt6_info, rt6i_dst));

	if (fn == NULL)
		goto out;

712 713
	pn = fn;

L
Linus Torvalds 已提交
714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767
#ifdef CONFIG_IPV6_SUBTREES
	if (rt->rt6i_src.plen) {
		struct fib6_node *sn;

		if (fn->subtree == NULL) {
			struct fib6_node *sfn;

			/*
			 * Create subtree.
			 *
			 *		fn[main tree]
			 *		|
			 *		sfn[subtree root]
			 *		   \
			 *		    sn[new leaf node]
			 */

			/* Create subtree root node */
			sfn = node_alloc();
			if (sfn == NULL)
				goto st_failure;

			sfn->leaf = &ip6_null_entry;
			atomic_inc(&ip6_null_entry.rt6i_ref);
			sfn->fn_flags = RTN_ROOT;
			sfn->fn_sernum = fib6_new_sernum();

			/* Now add the first leaf node to new subtree */

			sn = fib6_add_1(sfn, &rt->rt6i_src.addr,
					sizeof(struct in6_addr), rt->rt6i_src.plen,
					offsetof(struct rt6_info, rt6i_src));

			if (sn == NULL) {
				/* If it is failed, discard just allocated
				   root, and then (in st_failure) stale node
				   in main tree.
				 */
				node_free(sfn);
				goto st_failure;
			}

			/* Now link new subtree to main tree */
			sfn->parent = fn;
			fn->subtree = sfn;
		} else {
			sn = fib6_add_1(fn->subtree, &rt->rt6i_src.addr,
					sizeof(struct in6_addr), rt->rt6i_src.plen,
					offsetof(struct rt6_info, rt6i_src));

			if (sn == NULL)
				goto st_failure;
		}

768 769 770 771
		if (fn->leaf == NULL) {
			fn->leaf = rt;
			atomic_inc(&rt->rt6i_ref);
		}
L
Linus Torvalds 已提交
772 773 774 775
		fn = sn;
	}
#endif

776
	err = fib6_add_rt2node(fn, rt, info);
L
Linus Torvalds 已提交
777 778 779 780

	if (err == 0) {
		fib6_start_gc(rt);
		if (!(rt->rt6i_flags&RTF_CACHE))
781
			fib6_prune_clones(pn, rt);
L
Linus Torvalds 已提交
782 783 784
	}

out:
785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801
	if (err) {
#ifdef CONFIG_IPV6_SUBTREES
		/*
		 * If fib6_add_1 has cleared the old leaf pointer in the
		 * super-tree leaf node we have to find a new one for it.
		 */
		if (pn != fn && !pn->leaf && !(pn->fn_flags & RTN_RTINFO)) {
			pn->leaf = fib6_find_prefix(pn);
#if RT6_DEBUG >= 2
			if (!pn->leaf) {
				BUG_TRAP(pn->leaf != NULL);
				pn->leaf = &ip6_null_entry;
			}
#endif
			atomic_inc(&pn->leaf->rt6i_ref);
		}
#endif
L
Linus Torvalds 已提交
802
		dst_free(&rt->u.dst);
803
	}
L
Linus Torvalds 已提交
804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831
	return err;

#ifdef CONFIG_IPV6_SUBTREES
	/* Subtree creation failed, probably main tree node
	   is orphan. If it is, shoot it.
	 */
st_failure:
	if (fn && !(fn->fn_flags & (RTN_RTINFO|RTN_ROOT)))
		fib6_repair_tree(fn);
	dst_free(&rt->u.dst);
	return err;
#endif
}

/*
 *	Routing tree lookup
 *
 */

struct lookup_args {
	int		offset;		/* key offset on rt6_info	*/
	struct in6_addr	*addr;		/* search key			*/
};

static struct fib6_node * fib6_lookup_1(struct fib6_node *root,
					struct lookup_args *args)
{
	struct fib6_node *fn;
A
Al Viro 已提交
832
	__be32 dir;
L
Linus Torvalds 已提交
833

834 835 836
	if (unlikely(args->offset == 0))
		return NULL;

L
Linus Torvalds 已提交
837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857
	/*
	 *	Descend on a tree
	 */

	fn = root;

	for (;;) {
		struct fib6_node *next;

		dir = addr_bit_set(args->addr, fn->fn_bit);

		next = dir ? fn->right : fn->left;

		if (next) {
			fn = next;
			continue;
		}

		break;
	}

858
	while(fn) {
859
		if (FIB6_SUBTREE(fn) || fn->fn_flags & RTN_RTINFO) {
L
Linus Torvalds 已提交
860 861 862 863 864
			struct rt6key *key;

			key = (struct rt6key *) ((u8 *) fn->leaf +
						 args->offset);

865 866 867 868 869 870 871 872
			if (ipv6_prefix_equal(&key->addr, args->addr, key->plen)) {
#ifdef CONFIG_IPV6_SUBTREES
				if (fn->subtree)
					fn = fib6_lookup_1(fn->subtree, args + 1);
#endif
				if (!fn || fn->fn_flags & RTN_RTINFO)
					return fn;
			}
L
Linus Torvalds 已提交
873 874
		}

875 876 877
		if (fn->fn_flags & RTN_ROOT)
			break;

L
Linus Torvalds 已提交
878 879 880 881 882 883 884 885 886 887
		fn = fn->parent;
	}

	return NULL;
}

struct fib6_node * fib6_lookup(struct fib6_node *root, struct in6_addr *daddr,
			       struct in6_addr *saddr)
{
	struct fib6_node *fn;
888 889 890 891 892
	struct lookup_args args[] = {
		{
			.offset = offsetof(struct rt6_info, rt6i_dst),
			.addr = daddr,
		},
L
Linus Torvalds 已提交
893
#ifdef CONFIG_IPV6_SUBTREES
894 895 896 897
		{
			.offset = offsetof(struct rt6_info, rt6i_src),
			.addr = saddr,
		},
L
Linus Torvalds 已提交
898
#endif
899 900 901 902
		{
			.offset = 0,	/* sentinel */
		}
	};
L
Linus Torvalds 已提交
903

904
	fn = fib6_lookup_1(root, daddr ? args : args + 1);
L
Linus Torvalds 已提交
905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959

	if (fn == NULL || fn->fn_flags & RTN_TL_ROOT)
		fn = root;

	return fn;
}

/*
 *	Get node with specified destination prefix (and source prefix,
 *	if subtrees are used)
 */


static struct fib6_node * fib6_locate_1(struct fib6_node *root,
					struct in6_addr *addr,
					int plen, int offset)
{
	struct fib6_node *fn;

	for (fn = root; fn ; ) {
		struct rt6key *key = (struct rt6key *)((u8 *)fn->leaf + offset);

		/*
		 *	Prefix match
		 */
		if (plen < fn->fn_bit ||
		    !ipv6_prefix_equal(&key->addr, addr, fn->fn_bit))
			return NULL;

		if (plen == fn->fn_bit)
			return fn;

		/*
		 *	We have more bits to go
		 */
		if (addr_bit_set(addr, fn->fn_bit))
			fn = fn->right;
		else
			fn = fn->left;
	}
	return NULL;
}

struct fib6_node * fib6_locate(struct fib6_node *root,
			       struct in6_addr *daddr, int dst_len,
			       struct in6_addr *saddr, int src_len)
{
	struct fib6_node *fn;

	fn = fib6_locate_1(root, daddr, dst_len,
			   offsetof(struct rt6_info, rt6i_dst));

#ifdef CONFIG_IPV6_SUBTREES
	if (src_len) {
		BUG_TRAP(saddr!=NULL);
960 961
		if (fn && fn->subtree)
			fn = fib6_locate_1(fn->subtree, saddr, src_len,
L
Linus Torvalds 已提交
962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989
					   offsetof(struct rt6_info, rt6i_src));
	}
#endif

	if (fn && fn->fn_flags&RTN_RTINFO)
		return fn;

	return NULL;
}


/*
 *	Deletion
 *
 */

static struct rt6_info * fib6_find_prefix(struct fib6_node *fn)
{
	if (fn->fn_flags&RTN_ROOT)
		return &ip6_null_entry;

	while(fn) {
		if(fn->left)
			return fn->left->leaf;

		if(fn->right)
			return fn->right->leaf;

990
		fn = FIB6_SUBTREE(fn);
L
Linus Torvalds 已提交
991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020
	}
	return NULL;
}

/*
 *	Called to trim the tree of intermediate nodes when possible. "fn"
 *	is the node we want to try and remove.
 */

static struct fib6_node * fib6_repair_tree(struct fib6_node *fn)
{
	int children;
	int nstate;
	struct fib6_node *child, *pn;
	struct fib6_walker_t *w;
	int iter = 0;

	for (;;) {
		RT6_TRACE("fixing tree: plen=%d iter=%d\n", fn->fn_bit, iter);
		iter++;

		BUG_TRAP(!(fn->fn_flags&RTN_RTINFO));
		BUG_TRAP(!(fn->fn_flags&RTN_TL_ROOT));
		BUG_TRAP(fn->leaf==NULL);

		children = 0;
		child = NULL;
		if (fn->right) child = fn->right, children |= 1;
		if (fn->left) child = fn->left, children |= 2;

1021
		if (children == 3 || FIB6_SUBTREE(fn)
L
Linus Torvalds 已提交
1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039
#ifdef CONFIG_IPV6_SUBTREES
		    /* Subtree root (i.e. fn) may have one child */
		    || (children && fn->fn_flags&RTN_ROOT)
#endif
		    ) {
			fn->leaf = fib6_find_prefix(fn);
#if RT6_DEBUG >= 2
			if (fn->leaf==NULL) {
				BUG_TRAP(fn->leaf);
				fn->leaf = &ip6_null_entry;
			}
#endif
			atomic_inc(&fn->leaf->rt6i_ref);
			return fn->parent;
		}

		pn = fn->parent;
#ifdef CONFIG_IPV6_SUBTREES
1040
		if (FIB6_SUBTREE(pn) == fn) {
L
Linus Torvalds 已提交
1041
			BUG_TRAP(fn->fn_flags&RTN_ROOT);
1042
			FIB6_SUBTREE(pn) = NULL;
L
Linus Torvalds 已提交
1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089
			nstate = FWS_L;
		} else {
			BUG_TRAP(!(fn->fn_flags&RTN_ROOT));
#endif
			if (pn->right == fn) pn->right = child;
			else if (pn->left == fn) pn->left = child;
#if RT6_DEBUG >= 2
			else BUG_TRAP(0);
#endif
			if (child)
				child->parent = pn;
			nstate = FWS_R;
#ifdef CONFIG_IPV6_SUBTREES
		}
#endif

		read_lock(&fib6_walker_lock);
		FOR_WALKERS(w) {
			if (child == NULL) {
				if (w->root == fn) {
					w->root = w->node = NULL;
					RT6_TRACE("W %p adjusted by delroot 1\n", w);
				} else if (w->node == fn) {
					RT6_TRACE("W %p adjusted by delnode 1, s=%d/%d\n", w, w->state, nstate);
					w->node = pn;
					w->state = nstate;
				}
			} else {
				if (w->root == fn) {
					w->root = child;
					RT6_TRACE("W %p adjusted by delroot 2\n", w);
				}
				if (w->node == fn) {
					w->node = child;
					if (children&2) {
						RT6_TRACE("W %p adjusted by delnode 2, s=%d\n", w, w->state);
						w->state = w->state>=FWS_R ? FWS_U : FWS_INIT;
					} else {
						RT6_TRACE("W %p adjusted by delnode 2, s=%d\n", w, w->state);
						w->state = w->state>=FWS_C ? FWS_U : FWS_INIT;
					}
				}
			}
		}
		read_unlock(&fib6_walker_lock);

		node_free(fn);
1090
		if (pn->fn_flags&RTN_RTINFO || FIB6_SUBTREE(pn))
L
Linus Torvalds 已提交
1091 1092 1093 1094 1095 1096 1097 1098 1099
			return pn;

		rt6_release(pn->leaf);
		pn->leaf = NULL;
		fn = pn;
	}
}

static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
1100
			   struct nl_info *info)
L
Linus Torvalds 已提交
1101 1102 1103 1104 1105 1106 1107
{
	struct fib6_walker_t *w;
	struct rt6_info *rt = *rtp;

	RT6_TRACE("fib6_del_route\n");

	/* Unlink it */
1108
	*rtp = rt->u.dst.rt6_next;
L
Linus Torvalds 已提交
1109 1110 1111 1112
	rt->rt6i_node = NULL;
	rt6_stats.fib_rt_entries--;
	rt6_stats.fib_discarded_routes++;

1113 1114 1115 1116
	/* Reset round-robin state, if necessary */
	if (fn->rr_ptr == rt)
		fn->rr_ptr = NULL;

L
Linus Torvalds 已提交
1117 1118 1119 1120 1121
	/* Adjust walkers */
	read_lock(&fib6_walker_lock);
	FOR_WALKERS(w) {
		if (w->state == FWS_C && w->leaf == rt) {
			RT6_TRACE("walker %p adjusted by delroute\n", w);
1122
			w->leaf = rt->u.dst.rt6_next;
L
Linus Torvalds 已提交
1123 1124 1125 1126 1127 1128
			if (w->leaf == NULL)
				w->state = FWS_U;
		}
	}
	read_unlock(&fib6_walker_lock);

1129
	rt->u.dst.rt6_next = NULL;
L
Linus Torvalds 已提交
1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156

	/* If it was last route, expunge its radix tree node */
	if (fn->leaf == NULL) {
		fn->fn_flags &= ~RTN_RTINFO;
		rt6_stats.fib_route_nodes--;
		fn = fib6_repair_tree(fn);
	}

	if (atomic_read(&rt->rt6i_ref) != 1) {
		/* This route is used as dummy address holder in some split
		 * nodes. It is not leaked, but it still holds other resources,
		 * which must be released in time. So, scan ascendant nodes
		 * and replace dummy references to this route with references
		 * to still alive ones.
		 */
		while (fn) {
			if (!(fn->fn_flags&RTN_RTINFO) && fn->leaf == rt) {
				fn->leaf = fib6_find_prefix(fn);
				atomic_inc(&fn->leaf->rt6i_ref);
				rt6_release(rt);
			}
			fn = fn->parent;
		}
		/* No more references are possible at this point. */
		if (atomic_read(&rt->rt6i_ref) != 1) BUG();
	}

1157
	inet6_rt_notify(RTM_DELROUTE, rt, info);
L
Linus Torvalds 已提交
1158 1159 1160
	rt6_release(rt);
}

1161
int fib6_del(struct rt6_info *rt, struct nl_info *info)
L
Linus Torvalds 已提交
1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176
{
	struct fib6_node *fn = rt->rt6i_node;
	struct rt6_info **rtp;

#if RT6_DEBUG >= 2
	if (rt->u.dst.obsolete>0) {
		BUG_TRAP(fn==NULL);
		return -ENOENT;
	}
#endif
	if (fn == NULL || rt == &ip6_null_entry)
		return -ENOENT;

	BUG_TRAP(fn->fn_flags&RTN_RTINFO);

1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188
	if (!(rt->rt6i_flags&RTF_CACHE)) {
		struct fib6_node *pn = fn;
#ifdef CONFIG_IPV6_SUBTREES
		/* clones of this route might be in another subtree */
		if (rt->rt6i_src.plen) {
			while (!(pn->fn_flags&RTN_ROOT))
				pn = pn->parent;
			pn = pn->parent;
		}
#endif
		fib6_prune_clones(pn, rt);
	}
L
Linus Torvalds 已提交
1189 1190 1191 1192 1193

	/*
	 *	Walk the leaf entries looking for ourself
	 */

1194
	for (rtp = &fn->leaf; *rtp; rtp = &(*rtp)->u.dst.rt6_next) {
L
Linus Torvalds 已提交
1195
		if (*rtp == rt) {
1196
			fib6_del_route(fn, rtp, info);
L
Linus Torvalds 已提交
1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209
			return 0;
		}
	}
	return -ENOENT;
}

/*
 *	Tree traversal function.
 *
 *	Certainly, it is not interrupt safe.
 *	However, it is internally reenterable wrt itself and fib6_add/fib6_del.
 *	It means, that we can modify tree during walking
 *	and use this function for garbage collection, clone pruning,
1210
 *	cleaning tree when a device goes down etc. etc.
L
Linus Torvalds 已提交
1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226
 *
 *	It guarantees that every node will be traversed,
 *	and that it will be traversed only once.
 *
 *	Callback function w->func may return:
 *	0 -> continue walking.
 *	positive value -> walking is suspended (used by tree dumps,
 *	and probably by gc, if it will be split to several slices)
 *	negative value -> terminate walking.
 *
 *	The function itself returns:
 *	0   -> walk is complete.
 *	>0  -> walk is incomplete (i.e. suspended)
 *	<0  -> walk is terminated by an error.
 */

A
Adrian Bunk 已提交
1227
static int fib6_walk_continue(struct fib6_walker_t *w)
L
Linus Torvalds 已提交
1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243
{
	struct fib6_node *fn, *pn;

	for (;;) {
		fn = w->node;
		if (fn == NULL)
			return 0;

		if (w->prune && fn != w->root &&
		    fn->fn_flags&RTN_RTINFO && w->state < FWS_C) {
			w->state = FWS_C;
			w->leaf = fn->leaf;
		}
		switch (w->state) {
#ifdef CONFIG_IPV6_SUBTREES
		case FWS_S:
1244 1245
			if (FIB6_SUBTREE(fn)) {
				w->node = FIB6_SUBTREE(fn);
L
Linus Torvalds 已提交
1246 1247 1248
				continue;
			}
			w->state = FWS_L;
1249
#endif
L
Linus Torvalds 已提交
1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278
		case FWS_L:
			if (fn->left) {
				w->node = fn->left;
				w->state = FWS_INIT;
				continue;
			}
			w->state = FWS_R;
		case FWS_R:
			if (fn->right) {
				w->node = fn->right;
				w->state = FWS_INIT;
				continue;
			}
			w->state = FWS_C;
			w->leaf = fn->leaf;
		case FWS_C:
			if (w->leaf && fn->fn_flags&RTN_RTINFO) {
				int err = w->func(w);
				if (err)
					return err;
				continue;
			}
			w->state = FWS_U;
		case FWS_U:
			if (fn == w->root)
				return 0;
			pn = fn->parent;
			w->node = pn;
#ifdef CONFIG_IPV6_SUBTREES
1279
			if (FIB6_SUBTREE(pn) == fn) {
L
Linus Torvalds 已提交
1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300
				BUG_TRAP(fn->fn_flags&RTN_ROOT);
				w->state = FWS_L;
				continue;
			}
#endif
			if (pn->left == fn) {
				w->state = FWS_R;
				continue;
			}
			if (pn->right == fn) {
				w->state = FWS_C;
				w->leaf = w->node->leaf;
				continue;
			}
#if RT6_DEBUG >= 2
			BUG_TRAP(0);
#endif
		}
	}
}

A
Adrian Bunk 已提交
1301
static int fib6_walk(struct fib6_walker_t *w)
L
Linus Torvalds 已提交
1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316
{
	int res;

	w->state = FWS_INIT;
	w->node = w->root;

	fib6_walker_link(w);
	res = fib6_walk_continue(w);
	if (res <= 0)
		fib6_walker_unlink(w);
	return res;
}

static int fib6_clean_node(struct fib6_walker_t *w)
{
1317 1318 1319
	struct nl_info info = {
		.nl_net = &init_net,
	};
L
Linus Torvalds 已提交
1320 1321
	int res;
	struct rt6_info *rt;
1322
	struct fib6_cleaner_t *c = container_of(w, struct fib6_cleaner_t, w);
L
Linus Torvalds 已提交
1323

1324
	for (rt = w->leaf; rt; rt = rt->u.dst.rt6_next) {
L
Linus Torvalds 已提交
1325 1326 1327
		res = c->func(rt, c->arg);
		if (res < 0) {
			w->leaf = rt;
1328
			res = fib6_del(rt, &info);
L
Linus Torvalds 已提交
1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344
			if (res) {
#if RT6_DEBUG >= 2
				printk(KERN_DEBUG "fib6_clean_node: del failed: rt=%p@%p err=%d\n", rt, rt->rt6i_node, res);
#endif
				continue;
			}
			return 0;
		}
		BUG_TRAP(res==0);
	}
	w->leaf = rt;
	return 0;
}

/*
 *	Convenient frontend to tree walker.
1345
 *
L
Linus Torvalds 已提交
1346 1347 1348 1349 1350 1351 1352 1353
 *	func is called on each route.
 *		It may return -1 -> delete this route.
 *		              0  -> continue walking
 *
 *	prune==1 -> only immediate children of node (certainly,
 *	ignoring pure split nodes) will be scanned.
 */

A
Adrian Bunk 已提交
1354 1355 1356
static void fib6_clean_tree(struct fib6_node *root,
			    int (*func)(struct rt6_info *, void *arg),
			    int prune, void *arg)
L
Linus Torvalds 已提交
1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368
{
	struct fib6_cleaner_t c;

	c.w.root = root;
	c.w.func = fib6_clean_node;
	c.w.prune = prune;
	c.func = func;
	c.arg = arg;

	fib6_walk(&c.w);
}

T
Thomas Graf 已提交
1369 1370 1371 1372
void fib6_clean_all(int (*func)(struct rt6_info *, void *arg),
		    int prune, void *arg)
{
	struct fib6_table *table;
1373 1374
	struct hlist_node *node;
	unsigned int h;
T
Thomas Graf 已提交
1375

1376 1377 1378 1379
	rcu_read_lock();
	for (h = 0; h < FIB_TABLE_HASHSZ; h++) {
		hlist_for_each_entry_rcu(table, node, &fib_table_hash[h],
					 tb6_hlist) {
T
Thomas Graf 已提交
1380 1381 1382 1383 1384
			write_lock_bh(&table->tb6_lock);
			fib6_clean_tree(&table->tb6_root, func, prune, arg);
			write_unlock_bh(&table->tb6_lock);
		}
	}
1385
	rcu_read_unlock();
T
Thomas Graf 已提交
1386 1387
}

L
Linus Torvalds 已提交
1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453
static int fib6_prune_clone(struct rt6_info *rt, void *arg)
{
	if (rt->rt6i_flags & RTF_CACHE) {
		RT6_TRACE("pruning clone %p\n", rt);
		return -1;
	}

	return 0;
}

static void fib6_prune_clones(struct fib6_node *fn, struct rt6_info *rt)
{
	fib6_clean_tree(fn, fib6_prune_clone, 1, rt);
}

/*
 *	Garbage collection
 */

static struct fib6_gc_args
{
	int			timeout;
	int			more;
} gc_args;

static int fib6_age(struct rt6_info *rt, void *arg)
{
	unsigned long now = jiffies;

	/*
	 *	check addrconf expiration here.
	 *	Routes are expired even if they are in use.
	 *
	 *	Also age clones. Note, that clones are aged out
	 *	only if they are not in use now.
	 */

	if (rt->rt6i_flags&RTF_EXPIRES && rt->rt6i_expires) {
		if (time_after(now, rt->rt6i_expires)) {
			RT6_TRACE("expiring %p\n", rt);
			return -1;
		}
		gc_args.more++;
	} else if (rt->rt6i_flags & RTF_CACHE) {
		if (atomic_read(&rt->u.dst.__refcnt) == 0 &&
		    time_after_eq(now, rt->u.dst.lastuse + gc_args.timeout)) {
			RT6_TRACE("aging clone %p\n", rt);
			return -1;
		} else if ((rt->rt6i_flags & RTF_GATEWAY) &&
			   (!(rt->rt6i_nexthop->flags & NTF_ROUTER))) {
			RT6_TRACE("purging route %p via non-router but gateway\n",
				  rt);
			return -1;
		}
		gc_args.more++;
	}

	return 0;
}

static DEFINE_SPINLOCK(fib6_gc_lock);

void fib6_run_gc(unsigned long dummy)
{
	if (dummy != ~0UL) {
		spin_lock_bh(&fib6_gc_lock);
1454 1455
		gc_args.timeout = dummy ? (int)dummy :
			init_net.ipv6.sysctl.ip6_rt_gc_interval;
L
Linus Torvalds 已提交
1456 1457 1458 1459 1460 1461 1462
	} else {
		local_bh_disable();
		if (!spin_trylock(&fib6_gc_lock)) {
			mod_timer(&ip6_fib_timer, jiffies + HZ);
			local_bh_enable();
			return;
		}
1463
		gc_args.timeout = init_net.ipv6.sysctl.ip6_rt_gc_interval;
L
Linus Torvalds 已提交
1464 1465 1466 1467
	}
	gc_args.more = 0;

	ndisc_dst_gc(&gc_args.more);
T
Thomas Graf 已提交
1468
	fib6_clean_all(fib6_age, 0, NULL);
L
Linus Torvalds 已提交
1469 1470

	if (gc_args.more)
1471 1472
		mod_timer(&ip6_fib_timer, jiffies +
			  init_net.ipv6.sysctl.ip6_rt_gc_interval);
L
Linus Torvalds 已提交
1473 1474 1475 1476 1477 1478 1479
	else {
		del_timer(&ip6_fib_timer);
		ip6_fib_timer.expires = 0;
	}
	spin_unlock_bh(&fib6_gc_lock);
}

1480
int __init fib6_init(void)
L
Linus Torvalds 已提交
1481
{
1482
	int ret;
L
Linus Torvalds 已提交
1483 1484
	fib6_node_kmem = kmem_cache_create("fib6_nodes",
					   sizeof(struct fib6_node),
1485
					   0, SLAB_HWCACHE_ALIGN,
1486
					   NULL);
1487 1488 1489
	if (!fib6_node_kmem)
		return -ENOMEM;

T
Thomas Graf 已提交
1490
	fib6_tables_init();
1491

1492 1493 1494 1495 1496 1497 1498 1499 1500
	ret = __rtnl_register(PF_INET6, RTM_GETROUTE, NULL, inet6_dump_fib);
	if (ret)
		goto out_kmem_cache_create;
out:
	return ret;

out_kmem_cache_create:
	kmem_cache_destroy(fib6_node_kmem);
	goto out;
L
Linus Torvalds 已提交
1501 1502 1503 1504 1505 1506 1507
}

void fib6_gc_cleanup(void)
{
	del_timer(&ip6_fib_timer);
	kmem_cache_destroy(fib6_node_kmem);
}