ip6_flowlabel.c 17.2 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11
/*
 *	ip6_flowlabel.c		IPv6 flowlabel manager.
 *
 *	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.
 *
 *	Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
 */

12
#include <linux/capability.h>
L
Linus Torvalds 已提交
13 14 15 16 17 18 19 20 21 22 23
#include <linux/errno.h>
#include <linux/types.h>
#include <linux/socket.h>
#include <linux/net.h>
#include <linux/netdevice.h>
#include <linux/if_arp.h>
#include <linux/in6.h>
#include <linux/route.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>

24
#include <net/net_namespace.h>
L
Linus Torvalds 已提交
25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53
#include <net/sock.h>

#include <net/ipv6.h>
#include <net/ndisc.h>
#include <net/protocol.h>
#include <net/ip6_route.h>
#include <net/addrconf.h>
#include <net/rawv6.h>
#include <net/icmp.h>
#include <net/transp_v6.h>

#include <asm/uaccess.h>

#define FL_MIN_LINGER	6	/* Minimal linger. It is set to 6sec specified
				   in old IPv6 RFC. Well, it was reasonable value.
				 */
#define FL_MAX_LINGER	60	/* Maximal linger timeout */

/* FL hash table */

#define FL_MAX_PER_SOCK	32
#define FL_MAX_SIZE	4096
#define FL_HASH_MASK	255
#define FL_HASH(l)	(ntohl(l)&FL_HASH_MASK)

static atomic_t fl_size = ATOMIC_INIT(0);
static struct ip6_flowlabel *fl_ht[FL_HASH_MASK+1];

static void ip6_fl_gc(unsigned long dummy);
54
static DEFINE_TIMER(ip6_fl_gc_timer, ip6_fl_gc, 0, 0);
L
Linus Torvalds 已提交
55 56 57 58 59 60 61 62 63 64

/* FL hash table lock: it protects only of GC */

static DEFINE_RWLOCK(ip6_fl_lock);

/* Big socket sock */

static DEFINE_RWLOCK(ip6_sk_fl_lock);


65
static inline struct ip6_flowlabel *__fl_lookup(struct net *net, __be32 label)
L
Linus Torvalds 已提交
66 67 68 69
{
	struct ip6_flowlabel *fl;

	for (fl=fl_ht[FL_HASH(label)]; fl; fl = fl->next) {
70
		if (fl->label == label && fl->fl_net == net)
L
Linus Torvalds 已提交
71 72 73 74 75
			return fl;
	}
	return NULL;
}

76
static struct ip6_flowlabel *fl_lookup(struct net *net, __be32 label)
L
Linus Torvalds 已提交
77 78 79 80
{
	struct ip6_flowlabel *fl;

	read_lock_bh(&ip6_fl_lock);
81
	fl = __fl_lookup(net, label);
L
Linus Torvalds 已提交
82 83 84 85 86 87 88 89 90
	if (fl)
		atomic_inc(&fl->users);
	read_unlock_bh(&ip6_fl_lock);
	return fl;
}


static void fl_free(struct ip6_flowlabel *fl)
{
91 92
	if (fl) {
		release_net(fl->fl_net);
L
Linus Torvalds 已提交
93
		kfree(fl->opt);
94
	}
L
Linus Torvalds 已提交
95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151
	kfree(fl);
}

static void fl_release(struct ip6_flowlabel *fl)
{
	write_lock_bh(&ip6_fl_lock);

	fl->lastuse = jiffies;
	if (atomic_dec_and_test(&fl->users)) {
		unsigned long ttd = fl->lastuse + fl->linger;
		if (time_after(ttd, fl->expires))
			fl->expires = ttd;
		ttd = fl->expires;
		if (fl->opt && fl->share == IPV6_FL_S_EXCL) {
			struct ipv6_txoptions *opt = fl->opt;
			fl->opt = NULL;
			kfree(opt);
		}
		if (!timer_pending(&ip6_fl_gc_timer) ||
		    time_after(ip6_fl_gc_timer.expires, ttd))
			mod_timer(&ip6_fl_gc_timer, ttd);
	}
	write_unlock_bh(&ip6_fl_lock);
}

static void ip6_fl_gc(unsigned long dummy)
{
	int i;
	unsigned long now = jiffies;
	unsigned long sched = 0;

	write_lock(&ip6_fl_lock);

	for (i=0; i<=FL_HASH_MASK; i++) {
		struct ip6_flowlabel *fl, **flp;
		flp = &fl_ht[i];
		while ((fl=*flp) != NULL) {
			if (atomic_read(&fl->users) == 0) {
				unsigned long ttd = fl->lastuse + fl->linger;
				if (time_after(ttd, fl->expires))
					fl->expires = ttd;
				ttd = fl->expires;
				if (time_after_eq(now, ttd)) {
					*flp = fl->next;
					fl_free(fl);
					atomic_dec(&fl_size);
					continue;
				}
				if (!sched || time_before(ttd, sched))
					sched = ttd;
			}
			flp = &fl->next;
		}
	}
	if (!sched && atomic_read(&fl_size))
		sched = now + FL_MAX_LINGER;
	if (sched) {
152
		mod_timer(&ip6_fl_gc_timer, sched);
L
Linus Torvalds 已提交
153 154 155 156
	}
	write_unlock(&ip6_fl_lock);
}

157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179
static void ip6_fl_purge(struct net *net)
{
	int i;

	write_lock(&ip6_fl_lock);
	for (i = 0; i <= FL_HASH_MASK; i++) {
		struct ip6_flowlabel *fl, **flp;
		flp = &fl_ht[i];
		while ((fl = *flp) != NULL) {
			if (fl->fl_net == net && atomic_read(&fl->users) == 0) {
				*flp = fl->next;
				fl_free(fl);
				atomic_dec(&fl_size);
				continue;
			}
			flp = &fl->next;
		}
	}
	write_unlock(&ip6_fl_lock);
}

static struct ip6_flowlabel *fl_intern(struct net *net,
				       struct ip6_flowlabel *fl, __be32 label)
L
Linus Torvalds 已提交
180
{
181 182
	struct ip6_flowlabel *lfl;

L
Linus Torvalds 已提交
183 184 185 186 187 188 189
	fl->label = label & IPV6_FLOWLABEL_MASK;

	write_lock_bh(&ip6_fl_lock);
	if (label == 0) {
		for (;;) {
			fl->label = htonl(net_random())&IPV6_FLOWLABEL_MASK;
			if (fl->label) {
190
				lfl = __fl_lookup(net, fl->label);
L
Linus Torvalds 已提交
191 192 193 194
				if (lfl == NULL)
					break;
			}
		}
195 196 197 198 199 200 201 202 203
	} else {
		/*
		 * we dropper the ip6_fl_lock, so this entry could reappear
		 * and we need to recheck with it.
		 *
		 * OTOH no need to search the active socket first, like it is
		 * done in ipv6_flowlabel_opt - sock is locked, so new entry
		 * with the same label can only appear on another sock
		 */
204
		lfl = __fl_lookup(net, fl->label);
205 206 207 208 209
		if (lfl != NULL) {
			atomic_inc(&lfl->users);
			write_unlock_bh(&ip6_fl_lock);
			return lfl;
		}
L
Linus Torvalds 已提交
210 211 212 213 214 215 216
	}

	fl->lastuse = jiffies;
	fl->next = fl_ht[FL_HASH(fl->label)];
	fl_ht[FL_HASH(fl->label)] = fl;
	atomic_inc(&fl_size);
	write_unlock_bh(&ip6_fl_lock);
217
	return NULL;
L
Linus Torvalds 已提交
218 219 220 221 222 223
}



/* Socket flowlabel lists */

A
Al Viro 已提交
224
struct ip6_flowlabel * fl6_sock_lookup(struct sock *sk, __be32 label)
L
Linus Torvalds 已提交
225 226 227 228 229 230
{
	struct ipv6_fl_socklist *sfl;
	struct ipv6_pinfo *np = inet6_sk(sk);

	label &= IPV6_FLOWLABEL_MASK;

231
	read_lock_bh(&ip6_sk_fl_lock);
L
Linus Torvalds 已提交
232 233 234 235 236
	for (sfl=np->ipv6_fl_list; sfl; sfl = sfl->next) {
		struct ip6_flowlabel *fl = sfl->fl;
		if (fl->label == label) {
			fl->lastuse = jiffies;
			atomic_inc(&fl->users);
237
			read_unlock_bh(&ip6_sk_fl_lock);
L
Linus Torvalds 已提交
238 239 240
			return fl;
		}
	}
241
	read_unlock_bh(&ip6_sk_fl_lock);
L
Linus Torvalds 已提交
242 243 244
	return NULL;
}

245 246
EXPORT_SYMBOL_GPL(fl6_sock_lookup);

L
Linus Torvalds 已提交
247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271
void fl6_free_socklist(struct sock *sk)
{
	struct ipv6_pinfo *np = inet6_sk(sk);
	struct ipv6_fl_socklist *sfl;

	while ((sfl = np->ipv6_fl_list) != NULL) {
		np->ipv6_fl_list = sfl->next;
		fl_release(sfl->fl);
		kfree(sfl);
	}
}

/* Service routines */


/*
   It is the only difficult place. flowlabel enforces equal headers
   before and including routing header, however user may supply options
   following rthdr.
 */

struct ipv6_txoptions *fl6_merge_options(struct ipv6_txoptions * opt_space,
					 struct ip6_flowlabel * fl,
					 struct ipv6_txoptions * fopt)
{
272
	struct ipv6_txoptions * fl_opt = fl->opt;
273

274 275
	if (fopt == NULL || fopt->opt_flen == 0)
		return fl_opt;
276

L
Linus Torvalds 已提交
277 278
	if (fl_opt != NULL) {
		opt_space->hopopt = fl_opt->hopopt;
279
		opt_space->dst0opt = fl_opt->dst0opt;
L
Linus Torvalds 已提交
280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322
		opt_space->srcrt = fl_opt->srcrt;
		opt_space->opt_nflen = fl_opt->opt_nflen;
	} else {
		if (fopt->opt_nflen == 0)
			return fopt;
		opt_space->hopopt = NULL;
		opt_space->dst0opt = NULL;
		opt_space->srcrt = NULL;
		opt_space->opt_nflen = 0;
	}
	opt_space->dst1opt = fopt->dst1opt;
	opt_space->opt_flen = fopt->opt_flen;
	return opt_space;
}

static unsigned long check_linger(unsigned long ttl)
{
	if (ttl < FL_MIN_LINGER)
		return FL_MIN_LINGER*HZ;
	if (ttl > FL_MAX_LINGER && !capable(CAP_NET_ADMIN))
		return 0;
	return ttl*HZ;
}

static int fl6_renew(struct ip6_flowlabel *fl, unsigned long linger, unsigned long expires)
{
	linger = check_linger(linger);
	if (!linger)
		return -EPERM;
	expires = check_linger(expires);
	if (!expires)
		return -EPERM;
	fl->lastuse = jiffies;
	if (time_before(fl->linger, linger))
		fl->linger = linger;
	if (time_before(expires, fl->linger))
		expires = fl->linger;
	if (time_before(fl->expires, fl->lastuse + expires))
		fl->expires = fl->lastuse + expires;
	return 0;
}

static struct ip6_flowlabel *
323 324
fl_create(struct net *net, struct in6_flowlabel_req *freq, char __user *optval,
	  int optlen, int *err_p)
L
Linus Torvalds 已提交
325 326 327 328 329 330 331
{
	struct ip6_flowlabel *fl;
	int olen;
	int addr_type;
	int err;

	err = -ENOMEM;
332
	fl = kzalloc(sizeof(*fl), GFP_KERNEL);
L
Linus Torvalds 已提交
333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356
	if (fl == NULL)
		goto done;

	olen = optlen - CMSG_ALIGN(sizeof(*freq));
	if (olen > 0) {
		struct msghdr msg;
		struct flowi flowi;
		int junk;

		err = -ENOMEM;
		fl->opt = kmalloc(sizeof(*fl->opt) + olen, GFP_KERNEL);
		if (fl->opt == NULL)
			goto done;

		memset(fl->opt, 0, sizeof(*fl->opt));
		fl->opt->tot_len = sizeof(*fl->opt) + olen;
		err = -EFAULT;
		if (copy_from_user(fl->opt+1, optval+CMSG_ALIGN(sizeof(*freq)), olen))
			goto done;

		msg.msg_controllen = olen;
		msg.msg_control = (void*)(fl->opt+1);
		flowi.oif = 0;

357
		err = datagram_send_ctl(&msg, &flowi, fl->opt, &junk, &junk);
L
Linus Torvalds 已提交
358 359 360 361 362 363 364 365 366 367 368
		if (err)
			goto done;
		err = -EINVAL;
		if (fl->opt->opt_flen)
			goto done;
		if (fl->opt->opt_nflen == 0) {
			kfree(fl->opt);
			fl->opt = NULL;
		}
	}

369
	fl->fl_net = hold_net(net);
L
Linus Torvalds 已提交
370 371 372 373 374 375 376
	fl->expires = jiffies;
	err = fl6_renew(fl, freq->flr_linger, freq->flr_expires);
	if (err)
		goto done;
	fl->share = freq->flr_share;
	addr_type = ipv6_addr_type(&freq->flr_dst);
	if ((addr_type&IPV6_ADDR_MAPPED)
377 378
	    || addr_type == IPV6_ADDR_ANY) {
		err = -EINVAL;
L
Linus Torvalds 已提交
379
		goto done;
380
	}
L
Linus Torvalds 已提交
381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454
	ipv6_addr_copy(&fl->dst, &freq->flr_dst);
	atomic_set(&fl->users, 1);
	switch (fl->share) {
	case IPV6_FL_S_EXCL:
	case IPV6_FL_S_ANY:
		break;
	case IPV6_FL_S_PROCESS:
		fl->owner = current->pid;
		break;
	case IPV6_FL_S_USER:
		fl->owner = current->euid;
		break;
	default:
		err = -EINVAL;
		goto done;
	}
	return fl;

done:
	fl_free(fl);
	*err_p = err;
	return NULL;
}

static int mem_check(struct sock *sk)
{
	struct ipv6_pinfo *np = inet6_sk(sk);
	struct ipv6_fl_socklist *sfl;
	int room = FL_MAX_SIZE - atomic_read(&fl_size);
	int count = 0;

	if (room > FL_MAX_SIZE - FL_MAX_PER_SOCK)
		return 0;

	for (sfl = np->ipv6_fl_list; sfl; sfl = sfl->next)
		count++;

	if (room <= 0 ||
	    ((count >= FL_MAX_PER_SOCK ||
	     (count > 0 && room < FL_MAX_SIZE/2) || room < FL_MAX_SIZE/4)
	     && !capable(CAP_NET_ADMIN)))
		return -ENOBUFS;

	return 0;
}

static int ipv6_hdr_cmp(struct ipv6_opt_hdr *h1, struct ipv6_opt_hdr *h2)
{
	if (h1 == h2)
		return 0;
	if (h1 == NULL || h2 == NULL)
		return 1;
	if (h1->hdrlen != h2->hdrlen)
		return 1;
	return memcmp(h1+1, h2+1, ((h1->hdrlen+1)<<3) - sizeof(*h1));
}

static int ipv6_opt_cmp(struct ipv6_txoptions *o1, struct ipv6_txoptions *o2)
{
	if (o1 == o2)
		return 0;
	if (o1 == NULL || o2 == NULL)
		return 1;
	if (o1->opt_nflen != o2->opt_nflen)
		return 1;
	if (ipv6_hdr_cmp(o1->hopopt, o2->hopopt))
		return 1;
	if (ipv6_hdr_cmp(o1->dst0opt, o2->dst0opt))
		return 1;
	if (ipv6_hdr_cmp((struct ipv6_opt_hdr *)o1->srcrt, (struct ipv6_opt_hdr *)o2->srcrt))
		return 1;
	return 0;
}

455 456 457 458 459 460 461 462 463 464
static inline void fl_link(struct ipv6_pinfo *np, struct ipv6_fl_socklist *sfl,
		struct ip6_flowlabel *fl)
{
	write_lock_bh(&ip6_sk_fl_lock);
	sfl->fl = fl;
	sfl->next = np->ipv6_fl_list;
	np->ipv6_fl_list = sfl;
	write_unlock_bh(&ip6_sk_fl_lock);
}

L
Linus Torvalds 已提交
465 466 467
int ipv6_flowlabel_opt(struct sock *sk, char __user *optval, int optlen)
{
	int err;
468
	struct net *net = sock_net(sk);
L
Linus Torvalds 已提交
469 470 471 472
	struct ipv6_pinfo *np = inet6_sk(sk);
	struct in6_flowlabel_req freq;
	struct ipv6_fl_socklist *sfl1=NULL;
	struct ipv6_fl_socklist *sfl, **sflp;
473 474
	struct ip6_flowlabel *fl, *fl1 = NULL;

L
Linus Torvalds 已提交
475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510

	if (optlen < sizeof(freq))
		return -EINVAL;

	if (copy_from_user(&freq, optval, sizeof(freq)))
		return -EFAULT;

	switch (freq.flr_action) {
	case IPV6_FL_A_PUT:
		write_lock_bh(&ip6_sk_fl_lock);
		for (sflp = &np->ipv6_fl_list; (sfl=*sflp)!=NULL; sflp = &sfl->next) {
			if (sfl->fl->label == freq.flr_label) {
				if (freq.flr_label == (np->flow_label&IPV6_FLOWLABEL_MASK))
					np->flow_label &= ~IPV6_FLOWLABEL_MASK;
				*sflp = sfl->next;
				write_unlock_bh(&ip6_sk_fl_lock);
				fl_release(sfl->fl);
				kfree(sfl);
				return 0;
			}
		}
		write_unlock_bh(&ip6_sk_fl_lock);
		return -ESRCH;

	case IPV6_FL_A_RENEW:
		read_lock_bh(&ip6_sk_fl_lock);
		for (sfl = np->ipv6_fl_list; sfl; sfl = sfl->next) {
			if (sfl->fl->label == freq.flr_label) {
				err = fl6_renew(sfl->fl, freq.flr_linger, freq.flr_expires);
				read_unlock_bh(&ip6_sk_fl_lock);
				return err;
			}
		}
		read_unlock_bh(&ip6_sk_fl_lock);

		if (freq.flr_share == IPV6_FL_S_NONE && capable(CAP_NET_ADMIN)) {
511
			fl = fl_lookup(net, freq.flr_label);
L
Linus Torvalds 已提交
512 513 514 515 516 517 518 519 520 521 522 523
			if (fl) {
				err = fl6_renew(fl, freq.flr_linger, freq.flr_expires);
				fl_release(fl);
				return err;
			}
		}
		return -ESRCH;

	case IPV6_FL_A_GET:
		if (freq.flr_label & ~IPV6_FLOWLABEL_MASK)
			return -EINVAL;

524
		fl = fl_create(net, &freq, optval, optlen, &err);
L
Linus Torvalds 已提交
525 526 527 528 529 530 531 532 533 534 535 536 537 538
		if (fl == NULL)
			return err;
		sfl1 = kmalloc(sizeof(*sfl1), GFP_KERNEL);

		if (freq.flr_label) {
			err = -EEXIST;
			read_lock_bh(&ip6_sk_fl_lock);
			for (sfl = np->ipv6_fl_list; sfl; sfl = sfl->next) {
				if (sfl->fl->label == freq.flr_label) {
					if (freq.flr_flags&IPV6_FL_F_EXCL) {
						read_unlock_bh(&ip6_sk_fl_lock);
						goto done;
					}
					fl1 = sfl->fl;
539
					atomic_inc(&fl1->users);
L
Linus Torvalds 已提交
540 541 542 543 544 545
					break;
				}
			}
			read_unlock_bh(&ip6_sk_fl_lock);

			if (fl1 == NULL)
546
				fl1 = fl_lookup(net, freq.flr_label);
L
Linus Torvalds 已提交
547
			if (fl1) {
548
recheck:
L
Linus Torvalds 已提交
549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569
				err = -EEXIST;
				if (freq.flr_flags&IPV6_FL_F_EXCL)
					goto release;
				err = -EPERM;
				if (fl1->share == IPV6_FL_S_EXCL ||
				    fl1->share != fl->share ||
				    fl1->owner != fl->owner)
					goto release;

				err = -EINVAL;
				if (!ipv6_addr_equal(&fl1->dst, &fl->dst) ||
				    ipv6_opt_cmp(fl1->opt, fl->opt))
					goto release;

				err = -ENOMEM;
				if (sfl1 == NULL)
					goto release;
				if (fl->linger > fl1->linger)
					fl1->linger = fl->linger;
				if ((long)(fl->expires - fl1->expires) > 0)
					fl1->expires = fl->expires;
570
				fl_link(np, sfl1, fl1);
L
Linus Torvalds 已提交
571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586
				fl_free(fl);
				return 0;

release:
				fl_release(fl1);
				goto done;
			}
		}
		err = -ENOENT;
		if (!(freq.flr_flags&IPV6_FL_F_CREATE))
			goto done;

		err = -ENOMEM;
		if (sfl1 == NULL || (err = mem_check(sk)) != 0)
			goto done;

587
		fl1 = fl_intern(net, fl, freq.flr_label);
588 589
		if (fl1 != NULL)
			goto recheck;
L
Linus Torvalds 已提交
590

591 592 593 594 595 596
		if (!freq.flr_label) {
			if (copy_to_user(&((struct in6_flowlabel_req __user *) optval)->flr_label,
					 &fl->label, sizeof(fl->label))) {
				/* Intentionally ignore fault. */
			}
		}
L
Linus Torvalds 已提交
597

598
		fl_link(np, sfl1, fl);
L
Linus Torvalds 已提交
599 600 601 602 603 604 605 606 607 608 609 610 611 612 613
		return 0;

	default:
		return -EINVAL;
	}

done:
	fl_free(fl);
	kfree(sfl1);
	return err;
}

#ifdef CONFIG_PROC_FS

struct ip6fl_iter_state {
614
	struct seq_net_private p;
L
Linus Torvalds 已提交
615 616 617 618 619 620 621 622 623
	int bucket;
};

#define ip6fl_seq_private(seq)	((struct ip6fl_iter_state *)(seq)->private)

static struct ip6_flowlabel *ip6fl_get_first(struct seq_file *seq)
{
	struct ip6_flowlabel *fl = NULL;
	struct ip6fl_iter_state *state = ip6fl_seq_private(seq);
624
	struct net *net = seq_file_net(seq);
L
Linus Torvalds 已提交
625 626

	for (state->bucket = 0; state->bucket <= FL_HASH_MASK; ++state->bucket) {
627 628 629 630 631
		fl = fl_ht[state->bucket];

		while (fl && fl->fl_net != net)
			fl = fl->next;
		if (fl)
L
Linus Torvalds 已提交
632 633 634 635 636 637 638 639
			break;
	}
	return fl;
}

static struct ip6_flowlabel *ip6fl_get_next(struct seq_file *seq, struct ip6_flowlabel *fl)
{
	struct ip6fl_iter_state *state = ip6fl_seq_private(seq);
640
	struct net *net = seq_file_net(seq);
L
Linus Torvalds 已提交
641 642

	fl = fl->next;
643 644 645 646
try_again:
	while (fl && fl->fl_net != net)
		fl = fl->next;

L
Linus Torvalds 已提交
647
	while (!fl) {
648
		if (++state->bucket <= FL_HASH_MASK) {
L
Linus Torvalds 已提交
649
			fl = fl_ht[state->bucket];
650 651
			goto try_again;
		} else
652
			break;
L
Linus Torvalds 已提交
653 654 655 656 657 658 659 660 661 662 663 664 665 666
	}
	return fl;
}

static struct ip6_flowlabel *ip6fl_get_idx(struct seq_file *seq, loff_t pos)
{
	struct ip6_flowlabel *fl = ip6fl_get_first(seq);
	if (fl)
		while (pos && (fl = ip6fl_get_next(seq, fl)) != NULL)
			--pos;
	return pos ? NULL : fl;
}

static void *ip6fl_seq_start(struct seq_file *seq, loff_t *pos)
667
	__acquires(ip6_fl_lock)
L
Linus Torvalds 已提交
668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685
{
	read_lock_bh(&ip6_fl_lock);
	return *pos ? ip6fl_get_idx(seq, *pos - 1) : SEQ_START_TOKEN;
}

static void *ip6fl_seq_next(struct seq_file *seq, void *v, loff_t *pos)
{
	struct ip6_flowlabel *fl;

	if (v == SEQ_START_TOKEN)
		fl = ip6fl_get_first(seq);
	else
		fl = ip6fl_get_next(seq, v);
	++*pos;
	return fl;
}

static void ip6fl_seq_stop(struct seq_file *seq, void *v)
686
	__releases(ip6_fl_lock)
L
Linus Torvalds 已提交
687 688 689 690
{
	read_unlock_bh(&ip6_fl_lock);
}

691
static int ip6fl_seq_show(struct seq_file *seq, void *v)
L
Linus Torvalds 已提交
692
{
693 694 695 696 697
	if (v == SEQ_START_TOKEN)
		seq_printf(seq, "%-5s %-1s %-6s %-6s %-6s %-8s %-32s %s\n",
			   "Label", "S", "Owner", "Users", "Linger", "Expires", "Dst", "Opt");
	else {
		struct ip6_flowlabel *fl = v;
L
Linus Torvalds 已提交
698
		seq_printf(seq,
699
			   "%05X %-1d %-6d %-6d %-6ld %-8ld " NIP6_SEQFMT " %-4d\n",
L
Linus Torvalds 已提交
700 701 702 703 704 705 706 707 708 709 710 711
			   (unsigned)ntohl(fl->label),
			   fl->share,
			   (unsigned)fl->owner,
			   atomic_read(&fl->users),
			   fl->linger/HZ,
			   (long)(fl->expires - jiffies)/HZ,
			   NIP6(fl->dst),
			   fl->opt ? fl->opt->opt_nflen : 0);
	}
	return 0;
}

712
static const struct seq_operations ip6fl_seq_ops = {
L
Linus Torvalds 已提交
713 714 715 716 717 718 719 720
	.start	=	ip6fl_seq_start,
	.next	=	ip6fl_seq_next,
	.stop	=	ip6fl_seq_stop,
	.show	=	ip6fl_seq_show,
};

static int ip6fl_seq_open(struct inode *inode, struct file *file)
{
721 722
	return seq_open_net(inode, file, &ip6fl_seq_ops,
			    sizeof(struct ip6fl_iter_state));
L
Linus Torvalds 已提交
723 724
}

725
static const struct file_operations ip6fl_seq_fops = {
L
Linus Torvalds 已提交
726 727 728 729
	.owner		=	THIS_MODULE,
	.open		=	ip6fl_seq_open,
	.read		=	seq_read,
	.llseek		=	seq_lseek,
730
	.release	=	seq_release_net,
L
Linus Torvalds 已提交
731 732
};

733 734
static int ip6_flowlabel_proc_init(struct net *net)
{
735 736
	if (!proc_net_fops_create(net, "ip6_flowlabel",
				  S_IRUGO, &ip6fl_seq_fops))
737 738 739
		return -ENOMEM;
	return 0;
}
L
Linus Torvalds 已提交
740

741
static void ip6_flowlabel_proc_fini(struct net *net)
L
Linus Torvalds 已提交
742
{
743 744 745 746 747 748 749 750 751 752 753
	proc_net_remove(net, "ip6_flowlabel");
}
#else
static inline int ip6_flowlabel_proc_init(struct net *net)
{
	return 0;
}
static inline void ip6_flowlabel_proc_fini(struct net *net)
{
	return ;
}
L
Linus Torvalds 已提交
754
#endif
755

756 757 758
static inline void ip6_flowlabel_net_exit(struct net *net)
{
	ip6_fl_purge(net);
759
	ip6_flowlabel_proc_fini(net);
760 761 762
}

static struct pernet_operations ip6_flowlabel_net_ops = {
763
	.init = ip6_flowlabel_proc_init,
764 765 766
	.exit = ip6_flowlabel_net_exit,
};

767 768
int ip6_flowlabel_init(void)
{
769
	return register_pernet_subsys(&ip6_flowlabel_net_ops);
L
Linus Torvalds 已提交
770 771 772 773 774
}

void ip6_flowlabel_cleanup(void)
{
	del_timer(&ip6_fl_gc_timer);
775
	unregister_pernet_subsys(&ip6_flowlabel_net_ops);
L
Linus Torvalds 已提交
776
}