xfrm_state.c 36.0 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12
/*
 * xfrm_state.c
 *
 * Changes:
 *	Mitsuru KANDA @USAGI
 * 	Kazunori MIYAZAWA @USAGI
 * 	Kunihiro Ishiguro <kunihiro@ipinfusion.com>
 * 		IPv6 support
 * 	YOSHIFUJI Hideaki @USAGI
 * 		Split up af-specific functions
 *	Derek Atkins <derek@ihtfp.com>
 *		Add UDP Encapsulation
13
 *
L
Linus Torvalds 已提交
14 15 16 17 18 19 20
 */

#include <linux/workqueue.h>
#include <net/xfrm.h>
#include <linux/pfkeyv2.h>
#include <linux/ipsec.h>
#include <linux/module.h>
21
#include <linux/cache.h>
L
Linus Torvalds 已提交
22 23
#include <asm/uaccess.h>

24 25
#include "xfrm_hash.h"

26 27 28
struct sock *xfrm_nl;
EXPORT_SYMBOL(xfrm_nl);

29
u32 sysctl_xfrm_aevent_etime = XFRM_AE_ETIME;
30 31
EXPORT_SYMBOL(sysctl_xfrm_aevent_etime);

32
u32 sysctl_xfrm_aevent_rseqth = XFRM_AE_SEQT_SIZE;
33 34
EXPORT_SYMBOL(sysctl_xfrm_aevent_rseqth);

L
Linus Torvalds 已提交
35 36 37
/* Each xfrm_state may be linked to two tables:

   1. Hash table by (spi,daddr,ah/esp) to find SA by SPI. (input,ctl)
38
   2. Hash table by (daddr,family,reqid) to find what SAs exist for given
L
Linus Torvalds 已提交
39 40 41 42 43 44 45 46 47 48 49
      destination/tunnel endpoint. (output)
 */

static DEFINE_SPINLOCK(xfrm_state_lock);

/* Hash table to find appropriate SA towards given target (endpoint
 * of tunnel or destination of transport mode) allowed by selector.
 *
 * Main use is finding SA after policy selected tunnel or transport mode.
 * Also, it can be used by ah/esp icmp error handler to find offending SA.
 */
50 51 52 53 54 55
static struct hlist_head *xfrm_state_bydst __read_mostly;
static struct hlist_head *xfrm_state_bysrc __read_mostly;
static struct hlist_head *xfrm_state_byspi __read_mostly;
static unsigned int xfrm_state_hmask __read_mostly;
static unsigned int xfrm_state_hashmax __read_mostly = 1 * 1024 * 1024;
static unsigned int xfrm_state_num;
56
static unsigned int xfrm_state_genid;
57

58 59 60
static inline unsigned int xfrm_dst_hash(xfrm_address_t *daddr,
					 xfrm_address_t *saddr,
					 u32 reqid,
61
					 unsigned short family)
62
{
63
	return __xfrm_dst_hash(daddr, saddr, reqid, family, xfrm_state_hmask);
64 65
}

66 67
static inline unsigned int xfrm_src_hash(xfrm_address_t *daddr,
					 xfrm_address_t *saddr,
68
					 unsigned short family)
69
{
70
	return __xfrm_src_hash(daddr, saddr, family, xfrm_state_hmask);
71 72 73
}

static inline unsigned int
A
Al Viro 已提交
74
xfrm_spi_hash(xfrm_address_t *daddr, __be32 spi, u8 proto, unsigned short family)
75
{
76
	return __xfrm_spi_hash(daddr, spi, proto, family, xfrm_state_hmask);
77 78 79 80 81 82 83 84 85 86 87 88 89 90
}

static void xfrm_hash_transfer(struct hlist_head *list,
			       struct hlist_head *ndsttable,
			       struct hlist_head *nsrctable,
			       struct hlist_head *nspitable,
			       unsigned int nhashmask)
{
	struct hlist_node *entry, *tmp;
	struct xfrm_state *x;

	hlist_for_each_entry_safe(x, entry, tmp, list, bydst) {
		unsigned int h;

91 92 93
		h = __xfrm_dst_hash(&x->id.daddr, &x->props.saddr,
				    x->props.reqid, x->props.family,
				    nhashmask);
94 95
		hlist_add_head(&x->bydst, ndsttable+h);

96 97
		h = __xfrm_src_hash(&x->id.daddr, &x->props.saddr,
				    x->props.family,
98 99 100
				    nhashmask);
		hlist_add_head(&x->bysrc, nsrctable+h);

101 102 103 104 105 106
		if (x->id.spi) {
			h = __xfrm_spi_hash(&x->id.daddr, x->id.spi,
					    x->id.proto, x->props.family,
					    nhashmask);
			hlist_add_head(&x->byspi, nspitable+h);
		}
107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127
	}
}

static unsigned long xfrm_hash_new_size(void)
{
	return ((xfrm_state_hmask + 1) << 1) *
		sizeof(struct hlist_head);
}

static DEFINE_MUTEX(hash_resize_mutex);

static void xfrm_hash_resize(void *__unused)
{
	struct hlist_head *ndst, *nsrc, *nspi, *odst, *osrc, *ospi;
	unsigned long nsize, osize;
	unsigned int nhashmask, ohashmask;
	int i;

	mutex_lock(&hash_resize_mutex);

	nsize = xfrm_hash_new_size();
128
	ndst = xfrm_hash_alloc(nsize);
129 130
	if (!ndst)
		goto out_unlock;
131
	nsrc = xfrm_hash_alloc(nsize);
132
	if (!nsrc) {
133
		xfrm_hash_free(ndst, nsize);
134 135
		goto out_unlock;
	}
136
	nspi = xfrm_hash_alloc(nsize);
137
	if (!nspi) {
138 139
		xfrm_hash_free(ndst, nsize);
		xfrm_hash_free(nsrc, nsize);
140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162
		goto out_unlock;
	}

	spin_lock_bh(&xfrm_state_lock);

	nhashmask = (nsize / sizeof(struct hlist_head)) - 1U;
	for (i = xfrm_state_hmask; i >= 0; i--)
		xfrm_hash_transfer(xfrm_state_bydst+i, ndst, nsrc, nspi,
				   nhashmask);

	odst = xfrm_state_bydst;
	osrc = xfrm_state_bysrc;
	ospi = xfrm_state_byspi;
	ohashmask = xfrm_state_hmask;

	xfrm_state_bydst = ndst;
	xfrm_state_bysrc = nsrc;
	xfrm_state_byspi = nspi;
	xfrm_state_hmask = nhashmask;

	spin_unlock_bh(&xfrm_state_lock);

	osize = (ohashmask + 1) * sizeof(struct hlist_head);
163 164 165
	xfrm_hash_free(odst, osize);
	xfrm_hash_free(osrc, osize);
	xfrm_hash_free(ospi, osize);
166 167 168 169 170 171 172

out_unlock:
	mutex_unlock(&hash_resize_mutex);
}

static DECLARE_WORK(xfrm_hash_work, xfrm_hash_resize, NULL);

L
Linus Torvalds 已提交
173 174 175 176 177 178 179
DECLARE_WAIT_QUEUE_HEAD(km_waitq);
EXPORT_SYMBOL(km_waitq);

static DEFINE_RWLOCK(xfrm_state_afinfo_lock);
static struct xfrm_state_afinfo *xfrm_state_afinfo[NPROTO];

static struct work_struct xfrm_state_gc_work;
180
static HLIST_HEAD(xfrm_state_gc_list);
L
Linus Torvalds 已提交
181 182
static DEFINE_SPINLOCK(xfrm_state_gc_lock);

183
int __xfrm_state_delete(struct xfrm_state *x);
L
Linus Torvalds 已提交
184 185 186 187

static struct xfrm_state_afinfo *xfrm_state_get_afinfo(unsigned short family);
static void xfrm_state_put_afinfo(struct xfrm_state_afinfo *afinfo);

188
int km_query(struct xfrm_state *x, struct xfrm_tmpl *t, struct xfrm_policy *pol);
189
void km_state_expired(struct xfrm_state *x, int hard, u32 pid);
L
Linus Torvalds 已提交
190 191 192

static void xfrm_state_gc_destroy(struct xfrm_state *x)
{
193 194
	del_timer_sync(&x->timer);
	del_timer_sync(&x->rtimer);
J
Jesper Juhl 已提交
195 196 197 198
	kfree(x->aalg);
	kfree(x->ealg);
	kfree(x->calg);
	kfree(x->encap);
199
	kfree(x->coaddr);
200 201
	if (x->mode)
		xfrm_put_mode(x->mode);
L
Linus Torvalds 已提交
202 203 204 205
	if (x->type) {
		x->type->destructor(x);
		xfrm_put_type(x->type);
	}
206
	security_xfrm_state_free(x);
L
Linus Torvalds 已提交
207 208 209 210 211 212
	kfree(x);
}

static void xfrm_state_gc_task(void *data)
{
	struct xfrm_state *x;
213 214
	struct hlist_node *entry, *tmp;
	struct hlist_head gc_list;
L
Linus Torvalds 已提交
215 216

	spin_lock_bh(&xfrm_state_gc_lock);
217 218
	gc_list.first = xfrm_state_gc_list.first;
	INIT_HLIST_HEAD(&xfrm_state_gc_list);
L
Linus Torvalds 已提交
219 220
	spin_unlock_bh(&xfrm_state_gc_lock);

221
	hlist_for_each_entry_safe(x, entry, tmp, &gc_list, bydst)
L
Linus Torvalds 已提交
222
		xfrm_state_gc_destroy(x);
223

L
Linus Torvalds 已提交
224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281
	wake_up(&km_waitq);
}

static inline unsigned long make_jiffies(long secs)
{
	if (secs >= (MAX_SCHEDULE_TIMEOUT-1)/HZ)
		return MAX_SCHEDULE_TIMEOUT-1;
	else
	        return secs*HZ;
}

static void xfrm_timer_handler(unsigned long data)
{
	struct xfrm_state *x = (struct xfrm_state*)data;
	unsigned long now = (unsigned long)xtime.tv_sec;
	long next = LONG_MAX;
	int warn = 0;

	spin_lock(&x->lock);
	if (x->km.state == XFRM_STATE_DEAD)
		goto out;
	if (x->km.state == XFRM_STATE_EXPIRED)
		goto expired;
	if (x->lft.hard_add_expires_seconds) {
		long tmo = x->lft.hard_add_expires_seconds +
			x->curlft.add_time - now;
		if (tmo <= 0)
			goto expired;
		if (tmo < next)
			next = tmo;
	}
	if (x->lft.hard_use_expires_seconds) {
		long tmo = x->lft.hard_use_expires_seconds +
			(x->curlft.use_time ? : now) - now;
		if (tmo <= 0)
			goto expired;
		if (tmo < next)
			next = tmo;
	}
	if (x->km.dying)
		goto resched;
	if (x->lft.soft_add_expires_seconds) {
		long tmo = x->lft.soft_add_expires_seconds +
			x->curlft.add_time - now;
		if (tmo <= 0)
			warn = 1;
		else if (tmo < next)
			next = tmo;
	}
	if (x->lft.soft_use_expires_seconds) {
		long tmo = x->lft.soft_use_expires_seconds +
			(x->curlft.use_time ? : now) - now;
		if (tmo <= 0)
			warn = 1;
		else if (tmo < next)
			next = tmo;
	}

282
	x->km.dying = warn;
L
Linus Torvalds 已提交
283
	if (warn)
284
		km_state_expired(x, 0, 0);
L
Linus Torvalds 已提交
285
resched:
286 287 288
	if (next != LONG_MAX)
		mod_timer(&x->timer, jiffies + make_jiffies(next));

L
Linus Torvalds 已提交
289 290 291 292 293 294 295 296 297
	goto out;

expired:
	if (x->km.state == XFRM_STATE_ACQ && x->id.spi == 0) {
		x->km.state = XFRM_STATE_EXPIRED;
		wake_up(&km_waitq);
		next = 2;
		goto resched;
	}
298
	if (!__xfrm_state_delete(x) && x->id.spi)
299
		km_state_expired(x, 1, 0);
L
Linus Torvalds 已提交
300 301 302 303 304

out:
	spin_unlock(&x->lock);
}

305 306
static void xfrm_replay_timer_handler(unsigned long data);

L
Linus Torvalds 已提交
307 308 309 310
struct xfrm_state *xfrm_state_alloc(void)
{
	struct xfrm_state *x;

311
	x = kzalloc(sizeof(struct xfrm_state), GFP_ATOMIC);
L
Linus Torvalds 已提交
312 313 314 315

	if (x) {
		atomic_set(&x->refcnt, 1);
		atomic_set(&x->tunnel_users, 0);
316 317 318
		INIT_HLIST_NODE(&x->bydst);
		INIT_HLIST_NODE(&x->bysrc);
		INIT_HLIST_NODE(&x->byspi);
L
Linus Torvalds 已提交
319 320 321
		init_timer(&x->timer);
		x->timer.function = xfrm_timer_handler;
		x->timer.data	  = (unsigned long)x;
322 323 324
		init_timer(&x->rtimer);
		x->rtimer.function = xfrm_replay_timer_handler;
		x->rtimer.data     = (unsigned long)x;
L
Linus Torvalds 已提交
325 326 327 328 329
		x->curlft.add_time = (unsigned long)xtime.tv_sec;
		x->lft.soft_byte_limit = XFRM_INF;
		x->lft.soft_packet_limit = XFRM_INF;
		x->lft.hard_byte_limit = XFRM_INF;
		x->lft.hard_packet_limit = XFRM_INF;
330 331
		x->replay_maxage = 0;
		x->replay_maxdiff = 0;
L
Linus Torvalds 已提交
332 333 334 335 336 337 338 339 340 341 342
		spin_lock_init(&x->lock);
	}
	return x;
}
EXPORT_SYMBOL(xfrm_state_alloc);

void __xfrm_state_destroy(struct xfrm_state *x)
{
	BUG_TRAP(x->km.state == XFRM_STATE_DEAD);

	spin_lock_bh(&xfrm_state_gc_lock);
343
	hlist_add_head(&x->bydst, &xfrm_state_gc_list);
L
Linus Torvalds 已提交
344 345 346 347 348
	spin_unlock_bh(&xfrm_state_gc_lock);
	schedule_work(&xfrm_state_gc_work);
}
EXPORT_SYMBOL(__xfrm_state_destroy);

349
int __xfrm_state_delete(struct xfrm_state *x)
L
Linus Torvalds 已提交
350
{
351 352
	int err = -ESRCH;

L
Linus Torvalds 已提交
353 354 355
	if (x->km.state != XFRM_STATE_DEAD) {
		x->km.state = XFRM_STATE_DEAD;
		spin_lock(&xfrm_state_lock);
356 357
		hlist_del(&x->bydst);
		hlist_del(&x->bysrc);
358
		if (x->id.spi)
359
			hlist_del(&x->byspi);
360
		xfrm_state_num--;
L
Linus Torvalds 已提交
361 362 363 364 365 366
		spin_unlock(&xfrm_state_lock);

		/* All xfrm_state objects are created by xfrm_state_alloc.
		 * The xfrm_state_alloc call gives a reference, and that
		 * is what we are dropping here.
		 */
367
		__xfrm_state_put(x);
368
		err = 0;
L
Linus Torvalds 已提交
369
	}
370 371

	return err;
L
Linus Torvalds 已提交
372
}
373
EXPORT_SYMBOL(__xfrm_state_delete);
L
Linus Torvalds 已提交
374

375
int xfrm_state_delete(struct xfrm_state *x)
L
Linus Torvalds 已提交
376
{
377 378
	int err;

L
Linus Torvalds 已提交
379
	spin_lock_bh(&x->lock);
380
	err = __xfrm_state_delete(x);
L
Linus Torvalds 已提交
381
	spin_unlock_bh(&x->lock);
382 383

	return err;
L
Linus Torvalds 已提交
384 385 386 387 388 389 390 391
}
EXPORT_SYMBOL(xfrm_state_delete);

void xfrm_state_flush(u8 proto)
{
	int i;

	spin_lock_bh(&xfrm_state_lock);
392
	for (i = 0; i <= xfrm_state_hmask; i++) {
393 394
		struct hlist_node *entry;
		struct xfrm_state *x;
L
Linus Torvalds 已提交
395
restart:
396
		hlist_for_each_entry(x, entry, xfrm_state_bydst+i, bydst) {
L
Linus Torvalds 已提交
397
			if (!xfrm_state_kern(x) &&
398
			    xfrm_id_proto_match(x->id.proto, proto)) {
L
Linus Torvalds 已提交
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
				xfrm_state_hold(x);
				spin_unlock_bh(&xfrm_state_lock);

				xfrm_state_delete(x);
				xfrm_state_put(x);

				spin_lock_bh(&xfrm_state_lock);
				goto restart;
			}
		}
	}
	spin_unlock_bh(&xfrm_state_lock);
	wake_up(&km_waitq);
}
EXPORT_SYMBOL(xfrm_state_flush);

static int
xfrm_init_tempsel(struct xfrm_state *x, struct flowi *fl,
		  struct xfrm_tmpl *tmpl,
		  xfrm_address_t *daddr, xfrm_address_t *saddr,
		  unsigned short family)
{
	struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family);
	if (!afinfo)
		return -1;
	afinfo->init_tempsel(x, fl, tmpl, daddr, saddr);
	xfrm_state_put_afinfo(afinfo);
	return 0;
}

429
static struct xfrm_state *__xfrm_state_lookup(xfrm_address_t *daddr, __be32 spi, u8 proto, unsigned short family)
430 431 432
{
	unsigned int h = xfrm_spi_hash(daddr, spi, proto, family);
	struct xfrm_state *x;
433
	struct hlist_node *entry;
434

435
	hlist_for_each_entry(x, entry, xfrm_state_byspi+h, byspi) {
436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462
		if (x->props.family != family ||
		    x->id.spi       != spi ||
		    x->id.proto     != proto)
			continue;

		switch (family) {
		case AF_INET:
			if (x->id.daddr.a4 != daddr->a4)
				continue;
			break;
		case AF_INET6:
			if (!ipv6_addr_equal((struct in6_addr *)daddr,
					     (struct in6_addr *)
					     x->id.daddr.a6))
				continue;
			break;
		};

		xfrm_state_hold(x);
		return x;
	}

	return NULL;
}

static struct xfrm_state *__xfrm_state_lookup_byaddr(xfrm_address_t *daddr, xfrm_address_t *saddr, u8 proto, unsigned short family)
{
463
	unsigned int h = xfrm_src_hash(daddr, saddr, family);
464
	struct xfrm_state *x;
465
	struct hlist_node *entry;
466

467
	hlist_for_each_entry(x, entry, xfrm_state_bysrc+h, bysrc) {
468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507
		if (x->props.family != family ||
		    x->id.proto     != proto)
			continue;

		switch (family) {
		case AF_INET:
			if (x->id.daddr.a4 != daddr->a4 ||
			    x->props.saddr.a4 != saddr->a4)
				continue;
			break;
		case AF_INET6:
			if (!ipv6_addr_equal((struct in6_addr *)daddr,
					     (struct in6_addr *)
					     x->id.daddr.a6) ||
			    !ipv6_addr_equal((struct in6_addr *)saddr,
					     (struct in6_addr *)
					     x->props.saddr.a6))
				continue;
			break;
		};

		xfrm_state_hold(x);
		return x;
	}

	return NULL;
}

static inline struct xfrm_state *
__xfrm_state_locate(struct xfrm_state *x, int use_spi, int family)
{
	if (use_spi)
		return __xfrm_state_lookup(&x->id.daddr, x->id.spi,
					   x->id.proto, family);
	else
		return __xfrm_state_lookup_byaddr(&x->id.daddr,
						  &x->props.saddr,
						  x->id.proto, family);
}

L
Linus Torvalds 已提交
508 509 510 511 512 513
struct xfrm_state *
xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr, 
		struct flowi *fl, struct xfrm_tmpl *tmpl,
		struct xfrm_policy *pol, int *err,
		unsigned short family)
{
514
	unsigned int h = xfrm_dst_hash(daddr, saddr, tmpl->reqid, family);
515
	struct hlist_node *entry;
L
Linus Torvalds 已提交
516 517 518 519 520 521
	struct xfrm_state *x, *x0;
	int acquire_in_progress = 0;
	int error = 0;
	struct xfrm_state *best = NULL;
	
	spin_lock_bh(&xfrm_state_lock);
522
	hlist_for_each_entry(x, entry, xfrm_state_bydst+h, bydst) {
L
Linus Torvalds 已提交
523 524
		if (x->props.family == family &&
		    x->props.reqid == tmpl->reqid &&
525
		    !(x->props.flags & XFRM_STATE_WILDRECV) &&
L
Linus Torvalds 已提交
526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543
		    xfrm_state_addr_check(x, daddr, saddr, family) &&
		    tmpl->mode == x->props.mode &&
		    tmpl->id.proto == x->id.proto &&
		    (tmpl->id.spi == x->id.spi || !tmpl->id.spi)) {
			/* Resolution logic:
			   1. There is a valid state with matching selector.
			      Done.
			   2. Valid state with inappropriate selector. Skip.

			   Entering area of "sysdeps".

			   3. If state is not valid, selector is temporary,
			      it selects only session which triggered
			      previous resolution. Key manager will do
			      something to install a state with proper
			      selector.
			 */
			if (x->km.state == XFRM_STATE_VALID) {
544
				if (!xfrm_selector_match(&x->sel, fl, family) ||
545
				    !security_xfrm_state_pol_flow_match(x, pol, fl))
L
Linus Torvalds 已提交
546 547 548 549 550 551 552 553 554 555
					continue;
				if (!best ||
				    best->km.dying > x->km.dying ||
				    (best->km.dying == x->km.dying &&
				     best->curlft.add_time < x->curlft.add_time))
					best = x;
			} else if (x->km.state == XFRM_STATE_ACQ) {
				acquire_in_progress = 1;
			} else if (x->km.state == XFRM_STATE_ERROR ||
				   x->km.state == XFRM_STATE_EXPIRED) {
556
 				if (xfrm_selector_match(&x->sel, fl, family) &&
557
				    security_xfrm_state_pol_flow_match(x, pol, fl))
L
Linus Torvalds 已提交
558 559 560 561 562 563 564
					error = -ESRCH;
			}
		}
	}

	x = best;
	if (!x && !error && !acquire_in_progress) {
565
		if (tmpl->id.spi &&
566 567
		    (x0 = __xfrm_state_lookup(daddr, tmpl->id.spi,
					      tmpl->id.proto, family)) != NULL) {
L
Linus Torvalds 已提交
568 569 570 571 572 573 574 575 576 577 578 579 580
			xfrm_state_put(x0);
			error = -EEXIST;
			goto out;
		}
		x = xfrm_state_alloc();
		if (x == NULL) {
			error = -ENOMEM;
			goto out;
		}
		/* Initialize temporary selector matching only
		 * to current session. */
		xfrm_init_tempsel(x, fl, tmpl, daddr, saddr, family);

581 582 583 584 585 586 587 588
		error = security_xfrm_state_alloc_acquire(x, pol->security, fl->secid);
		if (error) {
			x->km.state = XFRM_STATE_DEAD;
			xfrm_state_put(x);
			x = NULL;
			goto out;
		}

L
Linus Torvalds 已提交
589 590
		if (km_query(x, tmpl, pol) == 0) {
			x->km.state = XFRM_STATE_ACQ;
591
			hlist_add_head(&x->bydst, xfrm_state_bydst+h);
592
			h = xfrm_src_hash(daddr, saddr, family);
593
			hlist_add_head(&x->bysrc, xfrm_state_bysrc+h);
L
Linus Torvalds 已提交
594 595
			if (x->id.spi) {
				h = xfrm_spi_hash(&x->id.daddr, x->id.spi, x->id.proto, family);
596
				hlist_add_head(&x->byspi, xfrm_state_byspi+h);
L
Linus Torvalds 已提交
597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618
			}
			x->lft.hard_add_expires_seconds = XFRM_ACQ_EXPIRES;
			x->timer.expires = jiffies + XFRM_ACQ_EXPIRES*HZ;
			add_timer(&x->timer);
		} else {
			x->km.state = XFRM_STATE_DEAD;
			xfrm_state_put(x);
			x = NULL;
			error = -ESRCH;
		}
	}
out:
	if (x)
		xfrm_state_hold(x);
	else
		*err = acquire_in_progress ? -EAGAIN : error;
	spin_unlock_bh(&xfrm_state_lock);
	return x;
}

static void __xfrm_state_insert(struct xfrm_state *x)
{
619
	unsigned int h;
L
Linus Torvalds 已提交
620

621 622
	x->genid = ++xfrm_state_genid;

623 624
	h = xfrm_dst_hash(&x->id.daddr, &x->props.saddr,
			  x->props.reqid, x->props.family);
625
	hlist_add_head(&x->bydst, xfrm_state_bydst+h);
L
Linus Torvalds 已提交
626

627
	h = xfrm_src_hash(&x->id.daddr, &x->props.saddr, x->props.family);
628
	hlist_add_head(&x->bysrc, xfrm_state_bysrc+h);
L
Linus Torvalds 已提交
629

630
	if (x->id.spi) {
631 632 633
		h = xfrm_spi_hash(&x->id.daddr, x->id.spi, x->id.proto,
				  x->props.family);

634
		hlist_add_head(&x->byspi, xfrm_state_byspi+h);
635 636
	}

637 638 639
	mod_timer(&x->timer, jiffies + HZ);
	if (x->replay_maxage)
		mod_timer(&x->rtimer, jiffies + x->replay_maxage);
640

L
Linus Torvalds 已提交
641
	wake_up(&km_waitq);
642 643 644 645 646 647 648

	xfrm_state_num++;

	if (x->bydst.next != NULL &&
	    (xfrm_state_hmask + 1) < xfrm_state_hashmax &&
	    xfrm_state_num > xfrm_state_hmask)
		schedule_work(&xfrm_hash_work);
L
Linus Torvalds 已提交
649 650
}

651 652 653 654 655 656 657 658 659
/* xfrm_state_lock is held */
static void __xfrm_state_bump_genids(struct xfrm_state *xnew)
{
	unsigned short family = xnew->props.family;
	u32 reqid = xnew->props.reqid;
	struct xfrm_state *x;
	struct hlist_node *entry;
	unsigned int h;

660
	h = xfrm_dst_hash(&xnew->id.daddr, &xnew->props.saddr, reqid, family);
661 662 663
	hlist_for_each_entry(x, entry, xfrm_state_bydst+h, bydst) {
		if (x->props.family	== family &&
		    x->props.reqid	== reqid &&
664 665
		    !xfrm_addr_cmp(&x->id.daddr, &xnew->id.daddr, family) &&
		    !xfrm_addr_cmp(&x->props.saddr, &xnew->props.saddr, family))
666 667 668 669
			x->genid = xfrm_state_genid;
	}
}

L
Linus Torvalds 已提交
670 671 672
void xfrm_state_insert(struct xfrm_state *x)
{
	spin_lock_bh(&xfrm_state_lock);
673
	__xfrm_state_bump_genids(x);
L
Linus Torvalds 已提交
674 675 676 677 678
	__xfrm_state_insert(x);
	spin_unlock_bh(&xfrm_state_lock);
}
EXPORT_SYMBOL(xfrm_state_insert);

679 680 681
/* xfrm_state_lock is held */
static struct xfrm_state *__find_acq_core(unsigned short family, u8 mode, u32 reqid, u8 proto, xfrm_address_t *daddr, xfrm_address_t *saddr, int create)
{
682
	unsigned int h = xfrm_dst_hash(daddr, saddr, reqid, family);
683
	struct hlist_node *entry;
684 685
	struct xfrm_state *x;

686
	hlist_for_each_entry(x, entry, xfrm_state_bydst+h, bydst) {
687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751
		if (x->props.reqid  != reqid ||
		    x->props.mode   != mode ||
		    x->props.family != family ||
		    x->km.state     != XFRM_STATE_ACQ ||
		    x->id.spi       != 0)
			continue;

		switch (family) {
		case AF_INET:
			if (x->id.daddr.a4    != daddr->a4 ||
			    x->props.saddr.a4 != saddr->a4)
				continue;
			break;
		case AF_INET6:
			if (!ipv6_addr_equal((struct in6_addr *)x->id.daddr.a6,
					     (struct in6_addr *)daddr) ||
			    !ipv6_addr_equal((struct in6_addr *)
					     x->props.saddr.a6,
					     (struct in6_addr *)saddr))
				continue;
			break;
		};

		xfrm_state_hold(x);
		return x;
	}

	if (!create)
		return NULL;

	x = xfrm_state_alloc();
	if (likely(x)) {
		switch (family) {
		case AF_INET:
			x->sel.daddr.a4 = daddr->a4;
			x->sel.saddr.a4 = saddr->a4;
			x->sel.prefixlen_d = 32;
			x->sel.prefixlen_s = 32;
			x->props.saddr.a4 = saddr->a4;
			x->id.daddr.a4 = daddr->a4;
			break;

		case AF_INET6:
			ipv6_addr_copy((struct in6_addr *)x->sel.daddr.a6,
				       (struct in6_addr *)daddr);
			ipv6_addr_copy((struct in6_addr *)x->sel.saddr.a6,
				       (struct in6_addr *)saddr);
			x->sel.prefixlen_d = 128;
			x->sel.prefixlen_s = 128;
			ipv6_addr_copy((struct in6_addr *)x->props.saddr.a6,
				       (struct in6_addr *)saddr);
			ipv6_addr_copy((struct in6_addr *)x->id.daddr.a6,
				       (struct in6_addr *)daddr);
			break;
		};

		x->km.state = XFRM_STATE_ACQ;
		x->id.proto = proto;
		x->props.family = family;
		x->props.mode = mode;
		x->props.reqid = reqid;
		x->lft.hard_add_expires_seconds = XFRM_ACQ_EXPIRES;
		xfrm_state_hold(x);
		x->timer.expires = jiffies + XFRM_ACQ_EXPIRES*HZ;
		add_timer(&x->timer);
752
		hlist_add_head(&x->bydst, xfrm_state_bydst+h);
753
		h = xfrm_src_hash(daddr, saddr, family);
754
		hlist_add_head(&x->bysrc, xfrm_state_bysrc+h);
755 756 757 758 759 760
		wake_up(&km_waitq);
	}

	return x;
}

L
Linus Torvalds 已提交
761 762 763 764 765 766 767
static struct xfrm_state *__xfrm_find_acq_byseq(u32 seq);

int xfrm_state_add(struct xfrm_state *x)
{
	struct xfrm_state *x1;
	int family;
	int err;
768
	int use_spi = xfrm_id_proto_match(x->id.proto, IPSEC_PROTO_ANY);
L
Linus Torvalds 已提交
769 770 771 772 773

	family = x->props.family;

	spin_lock_bh(&xfrm_state_lock);

774
	x1 = __xfrm_state_locate(x, use_spi, family);
L
Linus Torvalds 已提交
775 776 777 778 779 780 781
	if (x1) {
		xfrm_state_put(x1);
		x1 = NULL;
		err = -EEXIST;
		goto out;
	}

782
	if (use_spi && x->km.seq) {
L
Linus Torvalds 已提交
783 784 785 786 787 788 789
		x1 = __xfrm_find_acq_byseq(x->km.seq);
		if (x1 && xfrm_addr_cmp(&x1->id.daddr, &x->id.daddr, family)) {
			xfrm_state_put(x1);
			x1 = NULL;
		}
	}

790
	if (use_spi && !x1)
791 792 793
		x1 = __find_acq_core(family, x->props.mode, x->props.reqid,
				     x->id.proto,
				     &x->id.daddr, &x->props.saddr, 0);
L
Linus Torvalds 已提交
794

795
	__xfrm_state_bump_genids(x);
L
Linus Torvalds 已提交
796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814
	__xfrm_state_insert(x);
	err = 0;

out:
	spin_unlock_bh(&xfrm_state_lock);

	if (x1) {
		xfrm_state_delete(x1);
		xfrm_state_put(x1);
	}

	return err;
}
EXPORT_SYMBOL(xfrm_state_add);

int xfrm_state_update(struct xfrm_state *x)
{
	struct xfrm_state *x1;
	int err;
815
	int use_spi = xfrm_id_proto_match(x->id.proto, IPSEC_PROTO_ANY);
L
Linus Torvalds 已提交
816 817

	spin_lock_bh(&xfrm_state_lock);
818
	x1 = __xfrm_state_locate(x, use_spi, x->props.family);
L
Linus Torvalds 已提交
819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852

	err = -ESRCH;
	if (!x1)
		goto out;

	if (xfrm_state_kern(x1)) {
		xfrm_state_put(x1);
		err = -EEXIST;
		goto out;
	}

	if (x1->km.state == XFRM_STATE_ACQ) {
		__xfrm_state_insert(x);
		x = NULL;
	}
	err = 0;

out:
	spin_unlock_bh(&xfrm_state_lock);

	if (err)
		return err;

	if (!x) {
		xfrm_state_delete(x1);
		xfrm_state_put(x1);
		return 0;
	}

	err = -EINVAL;
	spin_lock_bh(&x1->lock);
	if (likely(x1->km.state == XFRM_STATE_VALID)) {
		if (x->encap && x1->encap)
			memcpy(x1->encap, x->encap, sizeof(*x1->encap));
853 854 855 856 857
		if (x->coaddr && x1->coaddr) {
			memcpy(x1->coaddr, x->coaddr, sizeof(*x1->coaddr));
		}
		if (!use_spi && memcmp(&x1->sel, &x->sel, sizeof(x1->sel)))
			memcpy(&x1->sel, &x->sel, sizeof(x1->sel));
L
Linus Torvalds 已提交
858 859 860
		memcpy(&x1->lft, &x->lft, sizeof(x1->lft));
		x1->km.dying = 0;

861
		mod_timer(&x1->timer, jiffies + HZ);
L
Linus Torvalds 已提交
862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884
		if (x1->curlft.use_time)
			xfrm_state_check_expire(x1);

		err = 0;
	}
	spin_unlock_bh(&x1->lock);

	xfrm_state_put(x1);

	return err;
}
EXPORT_SYMBOL(xfrm_state_update);

int xfrm_state_check_expire(struct xfrm_state *x)
{
	if (!x->curlft.use_time)
		x->curlft.use_time = (unsigned long)xtime.tv_sec;

	if (x->km.state != XFRM_STATE_VALID)
		return -EINVAL;

	if (x->curlft.bytes >= x->lft.hard_byte_limit ||
	    x->curlft.packets >= x->lft.hard_packet_limit) {
885
		x->km.state = XFRM_STATE_EXPIRED;
886
		mod_timer(&x->timer, jiffies);
L
Linus Torvalds 已提交
887 888 889 890 891
		return -EINVAL;
	}

	if (!x->km.dying &&
	    (x->curlft.bytes >= x->lft.soft_byte_limit ||
892 893
	     x->curlft.packets >= x->lft.soft_packet_limit)) {
		x->km.dying = 1;
894
		km_state_expired(x, 0, 0);
895
	}
L
Linus Torvalds 已提交
896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923
	return 0;
}
EXPORT_SYMBOL(xfrm_state_check_expire);

static int xfrm_state_check_space(struct xfrm_state *x, struct sk_buff *skb)
{
	int nhead = x->props.header_len + LL_RESERVED_SPACE(skb->dst->dev)
		- skb_headroom(skb);

	if (nhead > 0)
		return pskb_expand_head(skb, nhead, 0, GFP_ATOMIC);

	/* Check tail too... */
	return 0;
}

int xfrm_state_check(struct xfrm_state *x, struct sk_buff *skb)
{
	int err = xfrm_state_check_expire(x);
	if (err < 0)
		goto err;
	err = xfrm_state_check_space(x, skb);
err:
	return err;
}
EXPORT_SYMBOL(xfrm_state_check);

struct xfrm_state *
924
xfrm_state_lookup(xfrm_address_t *daddr, __be32 spi, u8 proto,
L
Linus Torvalds 已提交
925 926 927 928 929
		  unsigned short family)
{
	struct xfrm_state *x;

	spin_lock_bh(&xfrm_state_lock);
930
	x = __xfrm_state_lookup(daddr, spi, proto, family);
L
Linus Torvalds 已提交
931 932 933 934 935 936
	spin_unlock_bh(&xfrm_state_lock);
	return x;
}
EXPORT_SYMBOL(xfrm_state_lookup);

struct xfrm_state *
937 938 939 940 941 942
xfrm_state_lookup_byaddr(xfrm_address_t *daddr, xfrm_address_t *saddr,
			 u8 proto, unsigned short family)
{
	struct xfrm_state *x;

	spin_lock_bh(&xfrm_state_lock);
943
	x = __xfrm_state_lookup_byaddr(daddr, saddr, proto, family);
944 945 946 947 948 949
	spin_unlock_bh(&xfrm_state_lock);
	return x;
}
EXPORT_SYMBOL(xfrm_state_lookup_byaddr);

struct xfrm_state *
L
Linus Torvalds 已提交
950 951 952 953 954 955 956
xfrm_find_acq(u8 mode, u32 reqid, u8 proto, 
	      xfrm_address_t *daddr, xfrm_address_t *saddr, 
	      int create, unsigned short family)
{
	struct xfrm_state *x;

	spin_lock_bh(&xfrm_state_lock);
957
	x = __find_acq_core(family, mode, reqid, proto, daddr, saddr, create);
L
Linus Torvalds 已提交
958
	spin_unlock_bh(&xfrm_state_lock);
959

L
Linus Torvalds 已提交
960 961 962 963
	return x;
}
EXPORT_SYMBOL(xfrm_find_acq);

964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001
#ifdef CONFIG_XFRM_SUB_POLICY
int
xfrm_tmpl_sort(struct xfrm_tmpl **dst, struct xfrm_tmpl **src, int n,
	       unsigned short family)
{
	int err = 0;
	struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family);
	if (!afinfo)
		return -EAFNOSUPPORT;

	spin_lock_bh(&xfrm_state_lock);
	if (afinfo->tmpl_sort)
		err = afinfo->tmpl_sort(dst, src, n);
	spin_unlock_bh(&xfrm_state_lock);
	xfrm_state_put_afinfo(afinfo);
	return err;
}
EXPORT_SYMBOL(xfrm_tmpl_sort);

int
xfrm_state_sort(struct xfrm_state **dst, struct xfrm_state **src, int n,
		unsigned short family)
{
	int err = 0;
	struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family);
	if (!afinfo)
		return -EAFNOSUPPORT;

	spin_lock_bh(&xfrm_state_lock);
	if (afinfo->state_sort)
		err = afinfo->state_sort(dst, src, n);
	spin_unlock_bh(&xfrm_state_lock);
	xfrm_state_put_afinfo(afinfo);
	return err;
}
EXPORT_SYMBOL(xfrm_state_sort);
#endif

L
Linus Torvalds 已提交
1002 1003 1004 1005 1006 1007
/* Silly enough, but I'm lazy to build resolution list */

static struct xfrm_state *__xfrm_find_acq_byseq(u32 seq)
{
	int i;

1008
	for (i = 0; i <= xfrm_state_hmask; i++) {
1009 1010 1011 1012 1013 1014
		struct hlist_node *entry;
		struct xfrm_state *x;

		hlist_for_each_entry(x, entry, xfrm_state_bydst+i, bydst) {
			if (x->km.seq == seq &&
			    x->km.state == XFRM_STATE_ACQ) {
L
Linus Torvalds 已提交
1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047
				xfrm_state_hold(x);
				return x;
			}
		}
	}
	return NULL;
}

struct xfrm_state *xfrm_find_acq_byseq(u32 seq)
{
	struct xfrm_state *x;

	spin_lock_bh(&xfrm_state_lock);
	x = __xfrm_find_acq_byseq(seq);
	spin_unlock_bh(&xfrm_state_lock);
	return x;
}
EXPORT_SYMBOL(xfrm_find_acq_byseq);

u32 xfrm_get_acqseq(void)
{
	u32 res;
	static u32 acqseq;
	static DEFINE_SPINLOCK(acqseq_lock);

	spin_lock_bh(&acqseq_lock);
	res = (++acqseq ? : ++acqseq);
	spin_unlock_bh(&acqseq_lock);
	return res;
}
EXPORT_SYMBOL(xfrm_get_acqseq);

void
A
Al Viro 已提交
1048
xfrm_alloc_spi(struct xfrm_state *x, __be32 minspi, __be32 maxspi)
L
Linus Torvalds 已提交
1049
{
1050
	unsigned int h;
L
Linus Torvalds 已提交
1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064
	struct xfrm_state *x0;

	if (x->id.spi)
		return;

	if (minspi == maxspi) {
		x0 = xfrm_state_lookup(&x->id.daddr, minspi, x->id.proto, x->props.family);
		if (x0) {
			xfrm_state_put(x0);
			return;
		}
		x->id.spi = minspi;
	} else {
		u32 spi = 0;
A
Al Viro 已提交
1065 1066 1067 1068
		u32 low = ntohl(minspi);
		u32 high = ntohl(maxspi);
		for (h=0; h<high-low+1; h++) {
			spi = low + net_random()%(high-low+1);
L
Linus Torvalds 已提交
1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079
			x0 = xfrm_state_lookup(&x->id.daddr, htonl(spi), x->id.proto, x->props.family);
			if (x0 == NULL) {
				x->id.spi = htonl(spi);
				break;
			}
			xfrm_state_put(x0);
		}
	}
	if (x->id.spi) {
		spin_lock_bh(&xfrm_state_lock);
		h = xfrm_spi_hash(&x->id.daddr, x->id.spi, x->id.proto, x->props.family);
1080
		hlist_add_head(&x->byspi, xfrm_state_byspi+h);
L
Linus Torvalds 已提交
1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091
		spin_unlock_bh(&xfrm_state_lock);
		wake_up(&km_waitq);
	}
}
EXPORT_SYMBOL(xfrm_alloc_spi);

int xfrm_state_walk(u8 proto, int (*func)(struct xfrm_state *, int, void*),
		    void *data)
{
	int i;
	struct xfrm_state *x;
1092
	struct hlist_node *entry;
L
Linus Torvalds 已提交
1093 1094 1095 1096
	int count = 0;
	int err = 0;

	spin_lock_bh(&xfrm_state_lock);
1097
	for (i = 0; i <= xfrm_state_hmask; i++) {
1098
		hlist_for_each_entry(x, entry, xfrm_state_bydst+i, bydst) {
1099
			if (xfrm_id_proto_match(x->id.proto, proto))
L
Linus Torvalds 已提交
1100 1101 1102 1103 1104 1105 1106 1107
				count++;
		}
	}
	if (count == 0) {
		err = -ENOENT;
		goto out;
	}

1108
	for (i = 0; i <= xfrm_state_hmask; i++) {
1109
		hlist_for_each_entry(x, entry, xfrm_state_bydst+i, bydst) {
1110
			if (!xfrm_id_proto_match(x->id.proto, proto))
L
Linus Torvalds 已提交
1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122
				continue;
			err = func(x, --count, data);
			if (err)
				goto out;
		}
	}
out:
	spin_unlock_bh(&xfrm_state_lock);
	return err;
}
EXPORT_SYMBOL(xfrm_state_walk);

1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140

void xfrm_replay_notify(struct xfrm_state *x, int event)
{
	struct km_event c;
	/* we send notify messages in case
	 *  1. we updated on of the sequence numbers, and the seqno difference
	 *     is at least x->replay_maxdiff, in this case we also update the
	 *     timeout of our timer function
	 *  2. if x->replay_maxage has elapsed since last update,
	 *     and there were changes
	 *
	 *  The state structure must be locked!
	 */

	switch (event) {
	case XFRM_REPLAY_UPDATE:
		if (x->replay_maxdiff &&
		    (x->replay.seq - x->preplay.seq < x->replay_maxdiff) &&
J
Jamal Hadi Salim 已提交
1141 1142 1143 1144 1145 1146
		    (x->replay.oseq - x->preplay.oseq < x->replay_maxdiff)) {
			if (x->xflags & XFRM_TIME_DEFER)
				event = XFRM_REPLAY_TIMEOUT;
			else
				return;
		}
1147 1148 1149 1150 1151 1152

		break;

	case XFRM_REPLAY_TIMEOUT:
		if ((x->replay.seq == x->preplay.seq) &&
		    (x->replay.bitmap == x->preplay.bitmap) &&
J
Jamal Hadi Salim 已提交
1153 1154
		    (x->replay.oseq == x->preplay.oseq)) {
			x->xflags |= XFRM_TIME_DEFER;
1155
			return;
J
Jamal Hadi Salim 已提交
1156
		}
1157 1158 1159 1160 1161 1162 1163 1164 1165 1166

		break;
	}

	memcpy(&x->preplay, &x->replay, sizeof(struct xfrm_replay_state));
	c.event = XFRM_MSG_NEWAE;
	c.data.aevent = event;
	km_state_notify(x, &c);

	if (x->replay_maxage &&
1167
	    !mod_timer(&x->rtimer, jiffies + x->replay_maxage))
J
Jamal Hadi Salim 已提交
1168
		x->xflags &= ~XFRM_TIME_DEFER;
1169
}
1170
EXPORT_SYMBOL(xfrm_replay_notify);
1171 1172 1173 1174 1175 1176 1177

static void xfrm_replay_timer_handler(unsigned long data)
{
	struct xfrm_state *x = (struct xfrm_state*)data;

	spin_lock(&x->lock);

J
Jamal Hadi Salim 已提交
1178 1179 1180 1181 1182 1183
	if (x->km.state == XFRM_STATE_VALID) {
		if (xfrm_aevent_is_on())
			xfrm_replay_notify(x, XFRM_REPLAY_TIMEOUT);
		else
			x->xflags |= XFRM_TIME_DEFER;
	}
1184 1185 1186 1187

	spin_unlock(&x->lock);
}

1188
int xfrm_replay_check(struct xfrm_state *x, __be32 net_seq)
L
Linus Torvalds 已提交
1189 1190
{
	u32 diff;
1191
	u32 seq = ntohl(net_seq);
L
Linus Torvalds 已提交
1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212

	if (unlikely(seq == 0))
		return -EINVAL;

	if (likely(seq > x->replay.seq))
		return 0;

	diff = x->replay.seq - seq;
	if (diff >= x->props.replay_window) {
		x->stats.replay_window++;
		return -EINVAL;
	}

	if (x->replay.bitmap & (1U << diff)) {
		x->stats.replay++;
		return -EINVAL;
	}
	return 0;
}
EXPORT_SYMBOL(xfrm_replay_check);

1213
void xfrm_replay_advance(struct xfrm_state *x, __be32 net_seq)
L
Linus Torvalds 已提交
1214 1215
{
	u32 diff;
1216
	u32 seq = ntohl(net_seq);
L
Linus Torvalds 已提交
1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228

	if (seq > x->replay.seq) {
		diff = seq - x->replay.seq;
		if (diff < x->props.replay_window)
			x->replay.bitmap = ((x->replay.bitmap) << diff) | 1;
		else
			x->replay.bitmap = 1;
		x->replay.seq = seq;
	} else {
		diff = x->replay.seq - seq;
		x->replay.bitmap |= (1U << diff);
	}
1229 1230 1231

	if (xfrm_aevent_is_on())
		xfrm_replay_notify(x, XFRM_REPLAY_UPDATE);
L
Linus Torvalds 已提交
1232 1233 1234 1235 1236 1237
}
EXPORT_SYMBOL(xfrm_replay_advance);

static struct list_head xfrm_km_list = LIST_HEAD_INIT(xfrm_km_list);
static DEFINE_RWLOCK(xfrm_km_lock);

1238
void km_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c)
L
Linus Torvalds 已提交
1239 1240 1241
{
	struct xfrm_mgr *km;

1242 1243 1244 1245 1246 1247
	read_lock(&xfrm_km_lock);
	list_for_each_entry(km, &xfrm_km_list, list)
		if (km->notify_policy)
			km->notify_policy(xp, dir, c);
	read_unlock(&xfrm_km_lock);
}
L
Linus Torvalds 已提交
1248

1249 1250 1251
void km_state_notify(struct xfrm_state *x, struct km_event *c)
{
	struct xfrm_mgr *km;
L
Linus Torvalds 已提交
1252 1253
	read_lock(&xfrm_km_lock);
	list_for_each_entry(km, &xfrm_km_list, list)
1254 1255
		if (km->notify)
			km->notify(x, c);
L
Linus Torvalds 已提交
1256
	read_unlock(&xfrm_km_lock);
1257 1258 1259 1260 1261
}

EXPORT_SYMBOL(km_policy_notify);
EXPORT_SYMBOL(km_state_notify);

1262
void km_state_expired(struct xfrm_state *x, int hard, u32 pid)
1263 1264 1265
{
	struct km_event c;

1266
	c.data.hard = hard;
1267
	c.pid = pid;
1268
	c.event = XFRM_MSG_EXPIRE;
1269
	km_state_notify(x, &c);
L
Linus Torvalds 已提交
1270 1271 1272 1273 1274

	if (hard)
		wake_up(&km_waitq);
}

1275
EXPORT_SYMBOL(km_state_expired);
1276 1277 1278 1279
/*
 * We send to all registered managers regardless of failure
 * We are happy with one success
*/
1280
int km_query(struct xfrm_state *x, struct xfrm_tmpl *t, struct xfrm_policy *pol)
L
Linus Torvalds 已提交
1281
{
1282
	int err = -EINVAL, acqret;
L
Linus Torvalds 已提交
1283 1284 1285 1286
	struct xfrm_mgr *km;

	read_lock(&xfrm_km_lock);
	list_for_each_entry(km, &xfrm_km_list, list) {
1287 1288 1289
		acqret = km->acquire(x, t, pol, XFRM_POLICY_OUT);
		if (!acqret)
			err = acqret;
L
Linus Torvalds 已提交
1290 1291 1292 1293
	}
	read_unlock(&xfrm_km_lock);
	return err;
}
1294
EXPORT_SYMBOL(km_query);
L
Linus Torvalds 已提交
1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312

int km_new_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr, u16 sport)
{
	int err = -EINVAL;
	struct xfrm_mgr *km;

	read_lock(&xfrm_km_lock);
	list_for_each_entry(km, &xfrm_km_list, list) {
		if (km->new_mapping)
			err = km->new_mapping(x, ipaddr, sport);
		if (!err)
			break;
	}
	read_unlock(&xfrm_km_lock);
	return err;
}
EXPORT_SYMBOL(km_new_mapping);

1313
void km_policy_expired(struct xfrm_policy *pol, int dir, int hard, u32 pid)
L
Linus Torvalds 已提交
1314
{
1315
	struct km_event c;
L
Linus Torvalds 已提交
1316

1317
	c.data.hard = hard;
1318
	c.pid = pid;
1319
	c.event = XFRM_MSG_POLEXPIRE;
1320
	km_policy_notify(pol, dir, &c);
L
Linus Torvalds 已提交
1321 1322 1323 1324

	if (hard)
		wake_up(&km_waitq);
}
1325
EXPORT_SYMBOL(km_policy_expired);
L
Linus Torvalds 已提交
1326

1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345
int km_report(u8 proto, struct xfrm_selector *sel, xfrm_address_t *addr)
{
	int err = -EINVAL;
	int ret;
	struct xfrm_mgr *km;

	read_lock(&xfrm_km_lock);
	list_for_each_entry(km, &xfrm_km_list, list) {
		if (km->report) {
			ret = km->report(proto, sel, addr);
			if (!ret)
				err = ret;
		}
	}
	read_unlock(&xfrm_km_lock);
	return err;
}
EXPORT_SYMBOL(km_report);

L
Linus Torvalds 已提交
1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366
int xfrm_user_policy(struct sock *sk, int optname, u8 __user *optval, int optlen)
{
	int err;
	u8 *data;
	struct xfrm_mgr *km;
	struct xfrm_policy *pol = NULL;

	if (optlen <= 0 || optlen > PAGE_SIZE)
		return -EMSGSIZE;

	data = kmalloc(optlen, GFP_KERNEL);
	if (!data)
		return -ENOMEM;

	err = -EFAULT;
	if (copy_from_user(data, optval, optlen))
		goto out;

	err = -EINVAL;
	read_lock(&xfrm_km_lock);
	list_for_each_entry(km, &xfrm_km_list, list) {
1367
		pol = km->compile_policy(sk, optname, data,
L
Linus Torvalds 已提交
1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410
					 optlen, &err);
		if (err >= 0)
			break;
	}
	read_unlock(&xfrm_km_lock);

	if (err >= 0) {
		xfrm_sk_policy_insert(sk, err, pol);
		xfrm_pol_put(pol);
		err = 0;
	}

out:
	kfree(data);
	return err;
}
EXPORT_SYMBOL(xfrm_user_policy);

int xfrm_register_km(struct xfrm_mgr *km)
{
	write_lock_bh(&xfrm_km_lock);
	list_add_tail(&km->list, &xfrm_km_list);
	write_unlock_bh(&xfrm_km_lock);
	return 0;
}
EXPORT_SYMBOL(xfrm_register_km);

int xfrm_unregister_km(struct xfrm_mgr *km)
{
	write_lock_bh(&xfrm_km_lock);
	list_del(&km->list);
	write_unlock_bh(&xfrm_km_lock);
	return 0;
}
EXPORT_SYMBOL(xfrm_unregister_km);

int xfrm_state_register_afinfo(struct xfrm_state_afinfo *afinfo)
{
	int err = 0;
	if (unlikely(afinfo == NULL))
		return -EINVAL;
	if (unlikely(afinfo->family >= NPROTO))
		return -EAFNOSUPPORT;
1411
	write_lock_bh(&xfrm_state_afinfo_lock);
L
Linus Torvalds 已提交
1412 1413
	if (unlikely(xfrm_state_afinfo[afinfo->family] != NULL))
		err = -ENOBUFS;
1414
	else
L
Linus Torvalds 已提交
1415
		xfrm_state_afinfo[afinfo->family] = afinfo;
1416
	write_unlock_bh(&xfrm_state_afinfo_lock);
L
Linus Torvalds 已提交
1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427
	return err;
}
EXPORT_SYMBOL(xfrm_state_register_afinfo);

int xfrm_state_unregister_afinfo(struct xfrm_state_afinfo *afinfo)
{
	int err = 0;
	if (unlikely(afinfo == NULL))
		return -EINVAL;
	if (unlikely(afinfo->family >= NPROTO))
		return -EAFNOSUPPORT;
1428
	write_lock_bh(&xfrm_state_afinfo_lock);
L
Linus Torvalds 已提交
1429 1430 1431
	if (likely(xfrm_state_afinfo[afinfo->family] != NULL)) {
		if (unlikely(xfrm_state_afinfo[afinfo->family] != afinfo))
			err = -EINVAL;
1432
		else
L
Linus Torvalds 已提交
1433 1434
			xfrm_state_afinfo[afinfo->family] = NULL;
	}
1435
	write_unlock_bh(&xfrm_state_afinfo_lock);
L
Linus Torvalds 已提交
1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446
	return err;
}
EXPORT_SYMBOL(xfrm_state_unregister_afinfo);

static struct xfrm_state_afinfo *xfrm_state_get_afinfo(unsigned short family)
{
	struct xfrm_state_afinfo *afinfo;
	if (unlikely(family >= NPROTO))
		return NULL;
	read_lock(&xfrm_state_afinfo_lock);
	afinfo = xfrm_state_afinfo[family];
1447 1448
	if (unlikely(!afinfo))
		read_unlock(&xfrm_state_afinfo_lock);
L
Linus Torvalds 已提交
1449 1450 1451 1452 1453
	return afinfo;
}

static void xfrm_state_put_afinfo(struct xfrm_state_afinfo *afinfo)
{
1454
	read_unlock(&xfrm_state_afinfo_lock);
L
Linus Torvalds 已提交
1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471
}

/* Temporarily located here until net/xfrm/xfrm_tunnel.c is created */
void xfrm_state_delete_tunnel(struct xfrm_state *x)
{
	if (x->tunnel) {
		struct xfrm_state *t = x->tunnel;

		if (atomic_read(&t->tunnel_users) == 2)
			xfrm_state_delete(t);
		atomic_dec(&t->tunnel_users);
		xfrm_state_put(t);
		x->tunnel = NULL;
	}
}
EXPORT_SYMBOL(xfrm_state_delete_tunnel);

1472 1473 1474 1475 1476 1477
/*
 * This function is NOT optimal.  For example, with ESP it will give an
 * MTU that's usually two bytes short of being optimal.  However, it will
 * usually give an answer that's a multiple of 4 provided the input is
 * also a multiple of 4.
 */
L
Linus Torvalds 已提交
1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505
int xfrm_state_mtu(struct xfrm_state *x, int mtu)
{
	int res = mtu;

	res -= x->props.header_len;

	for (;;) {
		int m = res;

		if (m < 68)
			return 68;

		spin_lock_bh(&x->lock);
		if (x->km.state == XFRM_STATE_VALID &&
		    x->type && x->type->get_max_size)
			m = x->type->get_max_size(x, m);
		else
			m += x->props.header_len;
		spin_unlock_bh(&x->lock);

		if (m <= mtu)
			break;
		res -= (m - mtu);
	}

	return res;
}

H
Herbert Xu 已提交
1506 1507
int xfrm_init_state(struct xfrm_state *x)
{
1508 1509
	struct xfrm_state_afinfo *afinfo;
	int family = x->props.family;
H
Herbert Xu 已提交
1510 1511
	int err;

1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527
	err = -EAFNOSUPPORT;
	afinfo = xfrm_state_get_afinfo(family);
	if (!afinfo)
		goto error;

	err = 0;
	if (afinfo->init_flags)
		err = afinfo->init_flags(x);

	xfrm_state_put_afinfo(afinfo);

	if (err)
		goto error;

	err = -EPROTONOSUPPORT;
	x->type = xfrm_get_type(x->id.proto, family);
H
Herbert Xu 已提交
1528 1529 1530 1531 1532 1533 1534
	if (x->type == NULL)
		goto error;

	err = x->type->init_state(x);
	if (err)
		goto error;

1535 1536 1537 1538
	x->mode = xfrm_get_mode(x->props.mode, family);
	if (x->mode == NULL)
		goto error;

H
Herbert Xu 已提交
1539 1540 1541 1542 1543 1544 1545
	x->km.state = XFRM_STATE_VALID;

error:
	return err;
}

EXPORT_SYMBOL(xfrm_init_state);
L
Linus Torvalds 已提交
1546 1547 1548
 
void __init xfrm_state_init(void)
{
1549 1550 1551 1552
	unsigned int sz;

	sz = sizeof(struct hlist_head) * 8;

1553 1554 1555
	xfrm_state_bydst = xfrm_hash_alloc(sz);
	xfrm_state_bysrc = xfrm_hash_alloc(sz);
	xfrm_state_byspi = xfrm_hash_alloc(sz);
1556 1557 1558
	if (!xfrm_state_bydst || !xfrm_state_bysrc || !xfrm_state_byspi)
		panic("XFRM: Cannot allocate bydst/bysrc/byspi hashes.");
	xfrm_state_hmask = ((sz / sizeof(struct hlist_head)) - 1);
L
Linus Torvalds 已提交
1559 1560 1561 1562

	INIT_WORK(&xfrm_state_gc_work, xfrm_state_gc_task, NULL);
}