net/xfrm/xfrm_policy.c

   1 // SPDX-License-Identifier: GPL-2.0-only
   2 /*
   3  * xfrm_policy.c
   4  *
   5  * Changes:
   6  *      Mitsuru KANDA @USAGI
   7  *      Kazunori MIYAZAWA @USAGI
   8  *      Kunihiro Ishiguro <kunihiro@ipinfusion.com>
   9  *              IPv6 support
  10  *      Kazunori MIYAZAWA @USAGI
  11  *      YOSHIFUJI Hideaki
  12  *              Split up af-specific portion
  13  *      Derek Atkins <derek@ihtfp.com>          Add the post_input processor
  14  *
  15  */
  16
  17 #include <linux/err.h>
  18 #include <linux/slab.h>
  19 #include <linux/kmod.h>
  20 #include <linux/list.h>
  21 #include <linux/spinlock.h>
  22 #include <linux/workqueue.h>
  23 #include <linux/notifier.h>
  24 #include <linux/netdevice.h>
  25 #include <linux/netfilter.h>
  26 #include <linux/module.h>
  27 #include <linux/cache.h>
  28 #include <linux/cpu.h>
  29 #include <linux/audit.h>
  30 #include <linux/rhashtable.h>
  31 #include <linux/if_tunnel.h>
  32 #include <net/dst.h>
  33 #include <net/flow.h>
  34 #include <net/xfrm.h>
  35 #include <net/ip.h>
  36 #if IS_ENABLED(CONFIG_IPV6_MIP6)
  37 #include <net/mip6.h>
  38 #endif
  39 #ifdef CONFIG_XFRM_STATISTICS
  40 #include <net/snmp.h>
  41 #endif
  42
  43 #include "xfrm_hash.h"
  44
  45 #define XFRM_QUEUE_TMO_MIN ((unsigned)(HZ/10))
  46 #define XFRM_QUEUE_TMO_MAX ((unsigned)(60*HZ))
  47 #define XFRM_MAX_QUEUE_LEN      100
  48
  49 struct xfrm_flo {
  50         struct dst_entry *dst_orig;
  51         u8 flags;
  52 };
  53
  54 /* prefixes smaller than this are stored in lists, not trees. */
  55 #define INEXACT_PREFIXLEN_IPV4  16
  56 #define INEXACT_PREFIXLEN_IPV6  48
  57
  58 struct xfrm_pol_inexact_node {
  59         struct rb_node node;
  60         union {
  61                 xfrm_address_t addr;
  62                 struct rcu_head rcu;
  63         };
  64         u8 prefixlen;
  65
  66         struct rb_root root;
  67
  68         /* the policies matching this node, can be empty list */
  69         struct hlist_head hhead;
  70 };
  71
  72 /* xfrm inexact policy search tree:
  73  * xfrm_pol_inexact_bin = hash(dir,type,family,if_id);
  74  *  |
  75  * +---- root_d: sorted by daddr:prefix
  76  * |                 |
  77  * |        xfrm_pol_inexact_node
  78  * |                 |
  79  * |                 +- root: sorted by saddr/prefix
  80  * |                 |              |
  81  * |                 |         xfrm_pol_inexact_node
  82  * |                 |              |
  83  * |                 |              + root: unused
  84  * |                 |              |
  85  * |                 |              + hhead: saddr:daddr policies
  86  * |                 |
  87  * |                 +- coarse policies and all any:daddr policies
  88  * |
  89  * +---- root_s: sorted by saddr:prefix
  90  * |                 |
  91  * |        xfrm_pol_inexact_node
  92  * |                 |
  93  * |                 + root: unused
  94  * |                 |
  95  * |                 + hhead: saddr:any policies
  96  * |
  97  * +---- coarse policies and all any:any policies
  98  *
  99  * Lookups return four candidate lists:
 100  * 1. any:any list from top-level xfrm_pol_inexact_bin
 101  * 2. any:daddr list from daddr tree
 102  * 3. saddr:daddr list from 2nd level daddr tree
 103  * 4. saddr:any list from saddr tree
 104  *
 105  * This result set then needs to be searched for the policy with
 106  * the lowest priority.  If two results have same prio, youngest one wins.
 107  */
 108
 109 struct xfrm_pol_inexact_key {
 110         possible_net_t net;
 111         u32 if_id;
 112         u16 family;
 113         u8 dir, type;
 114 };
 115
 116 struct xfrm_pol_inexact_bin {
 117         struct xfrm_pol_inexact_key k;
 118         struct rhash_head head;
 119         /* list containing '*:*' policies */
 120         struct hlist_head hhead;
 121
 122         seqcount_t count;
 123         /* tree sorted by daddr/prefix */
 124         struct rb_root root_d;
 125
 126         /* tree sorted by saddr/prefix */
 127         struct rb_root root_s;
 128
 129         /* slow path below */
 130         struct list_head inexact_bins;
 131         struct rcu_head rcu;
 132 };
 133
 134 enum xfrm_pol_inexact_candidate_type {
 135         XFRM_POL_CAND_BOTH,
 136         XFRM_POL_CAND_SADDR,
 137         XFRM_POL_CAND_DADDR,
 138         XFRM_POL_CAND_ANY,
 139
 140         XFRM_POL_CAND_MAX,
 141 };
 142
 143 struct xfrm_pol_inexact_candidates {
 144         struct hlist_head *res[XFRM_POL_CAND_MAX];
 145 };
 146
 147 static DEFINE_SPINLOCK(xfrm_if_cb_lock);
 148 static struct xfrm_if_cb const __rcu *xfrm_if_cb __read_mostly;
 149
 150 static DEFINE_SPINLOCK(xfrm_policy_afinfo_lock);
 151 static struct xfrm_policy_afinfo const __rcu *xfrm_policy_afinfo[AF_INET6 + 1]
 152                                                 __read_mostly;
 153
 154 static struct kmem_cache *xfrm_dst_cache __ro_after_init;
 155 static __read_mostly seqcount_t xfrm_policy_hash_generation;
 156
 157 static struct rhashtable xfrm_policy_inexact_table;
 158 static const struct rhashtable_params xfrm_pol_inexact_params;
 159
 160 static void xfrm_init_pmtu(struct xfrm_dst **bundle, int nr);
 161 static int stale_bundle(struct dst_entry *dst);
 162 static int xfrm_bundle_ok(struct xfrm_dst *xdst);
 163 static void xfrm_policy_queue_process(struct timer_list *t);
 164
 165 static void __xfrm_policy_link(struct xfrm_policy *pol, int dir);
 166 static struct xfrm_policy *__xfrm_policy_unlink(struct xfrm_policy *pol,
 167                                                 int dir);
 168
 169 static struct xfrm_pol_inexact_bin *
 170 xfrm_policy_inexact_lookup(struct net *net, u8 type, u16 family, u8 dir,
 171                            u32 if_id);
 172
 173 static struct xfrm_pol_inexact_bin *
 174 xfrm_policy_inexact_lookup_rcu(struct net *net,
 175                                u8 type, u16 family, u8 dir, u32 if_id);
 176 static struct xfrm_policy *
 177 xfrm_policy_insert_list(struct hlist_head *chain, struct xfrm_policy *policy,
 178                         bool excl);
 179 static void xfrm_policy_insert_inexact_list(struct hlist_head *chain,
 180                                             struct xfrm_policy *policy);
 181
 182 static bool
 183 xfrm_policy_find_inexact_candidates(struct xfrm_pol_inexact_candidates *cand,
 184                                     struct xfrm_pol_inexact_bin *b,
 185                                     const xfrm_address_t *saddr,
 186                                     const xfrm_address_t *daddr);
 187
 188 static inline bool xfrm_pol_hold_rcu(struct xfrm_policy *policy)
 189 {
 190         return refcount_inc_not_zero(&policy->refcnt);
 191 }
 192
 193 static inline bool
 194 __xfrm4_selector_match(const struct xfrm_selector *sel, const struct flowi *fl)
 195 {
 196         const struct flowi4 *fl4 = &fl->u.ip4;
 197
 198         return  addr4_match(fl4->daddr, sel->daddr.a4, sel->prefixlen_d) &&
 199                 addr4_match(fl4->saddr, sel->saddr.a4, sel->prefixlen_s) &&
 200                 !((xfrm_flowi_dport(fl, &fl4->uli) ^ sel->dport) & sel->dport_mask) &&
 201                 !((xfrm_flowi_sport(fl, &fl4->uli) ^ sel->sport) & sel->sport_mask) &&
 202                 (fl4->flowi4_proto == sel->proto || !sel->proto) &&
 203                 (fl4->flowi4_oif == sel->ifindex || !sel->ifindex);
 204 }
 205
 206 static inline bool
 207 __xfrm6_selector_match(const struct xfrm_selector *sel, const struct flowi *fl)
 208 {
 209         const struct flowi6 *fl6 = &fl->u.ip6;
 210
 211         return  addr_match(&fl6->daddr, &sel->daddr, sel->prefixlen_d) &&
 212                 addr_match(&fl6->saddr, &sel->saddr, sel->prefixlen_s) &&
 213                 !((xfrm_flowi_dport(fl, &fl6->uli) ^ sel->dport) & sel->dport_mask) &&
 214                 !((xfrm_flowi_sport(fl, &fl6->uli) ^ sel->sport) & sel->sport_mask) &&
 215                 (fl6->flowi6_proto == sel->proto || !sel->proto) &&
 216                 (fl6->flowi6_oif == sel->ifindex || !sel->ifindex);
 217 }
 218
 219 bool xfrm_selector_match(const struct xfrm_selector *sel, const struct flowi *fl,
 220                          unsigned short family)
 221 {
 222         switch (family) {
 223         case AF_INET:
 224                 return __xfrm4_selector_match(sel, fl);
 225         case AF_INET6:
 226                 return __xfrm6_selector_match(sel, fl);
 227         }
 228         return false;
 229 }
 230
 231 static const struct xfrm_policy_afinfo *xfrm_policy_get_afinfo(unsigned short family)
 232 {
 233         const struct xfrm_policy_afinfo *afinfo;
 234
 235         if (unlikely(family >= ARRAY_SIZE(xfrm_policy_afinfo)))
 236                 return NULL;
 237         rcu_read_lock();
 238         afinfo = rcu_dereference(xfrm_policy_afinfo[family]);
 239         if (unlikely(!afinfo))
 240                 rcu_read_unlock();
 241         return afinfo;
 242 }
 243
 244 /* Called with rcu_read_lock(). */
 245 static const struct xfrm_if_cb *xfrm_if_get_cb(void)
 246 {
 247         return rcu_dereference(xfrm_if_cb);
 248 }
 249
 250 struct dst_entry *__xfrm_dst_lookup(struct net *net, int tos, int oif,
 251                                     const xfrm_address_t *saddr,
 252                                     const xfrm_address_t *daddr,
 253                                     int family, u32 mark)
 254 {
 255         const struct xfrm_policy_afinfo *afinfo;
 256         struct dst_entry *dst;
 257
 258         afinfo = xfrm_policy_get_afinfo(family);
 259         if (unlikely(afinfo == NULL))
 260                 return ERR_PTR(-EAFNOSUPPORT);
 261
 262         dst = afinfo->dst_lookup(net, tos, oif, saddr, daddr, mark);
 263
 264         rcu_read_unlock();
 265
 266         return dst;
 267 }
 268 EXPORT_SYMBOL(__xfrm_dst_lookup);
 269
 270 static inline struct dst_entry *xfrm_dst_lookup(struct xfrm_state *x,
 271                                                 int tos, int oif,
 272                                                 xfrm_address_t *prev_saddr,
 273                                                 xfrm_address_t *prev_daddr,
 274                                                 int family, u32 mark)
 275 {
 276         struct net *net = xs_net(x);
 277         xfrm_address_t *saddr = &x->props.saddr;
 278         xfrm_address_t *daddr = &x->id.daddr;
 279         struct dst_entry *dst;
 280
 281         if (x->type->flags & XFRM_TYPE_LOCAL_COADDR) {
 282                 saddr = x->coaddr;
 283                 daddr = prev_daddr;
 284         }
 285         if (x->type->flags & XFRM_TYPE_REMOTE_COADDR) {
 286                 saddr = prev_saddr;
 287                 daddr = x->coaddr;
 288         }
 289
 290         dst = __xfrm_dst_lookup(net, tos, oif, saddr, daddr, family, mark);
 291
 292         if (!IS_ERR(dst)) {
 293                 if (prev_saddr != saddr)
 294                         memcpy(prev_saddr, saddr,  sizeof(*prev_saddr));
 295                 if (prev_daddr != daddr)
 296                         memcpy(prev_daddr, daddr,  sizeof(*prev_daddr));
 297         }
 298
 299         return dst;
 300 }
 301
 302 static inline unsigned long make_jiffies(long secs)
 303 {
 304         if (secs >= (MAX_SCHEDULE_TIMEOUT-1)/HZ)
 305                 return MAX_SCHEDULE_TIMEOUT-1;
 306         else
 307                 return secs*HZ;
 308 }
 309
 310 static void xfrm_policy_timer(struct timer_list *t)
 311 {
 312         struct xfrm_policy *xp = from_timer(xp, t, timer);
 313         time64_t now = ktime_get_real_seconds();
 314         time64_t next = TIME64_MAX;
 315         int warn = 0;
 316         int dir;
 317
 318         read_lock(&xp->lock);
 319
 320         if (unlikely(xp->walk.dead))
 321                 goto out;
 322
 323         dir = xfrm_policy_id2dir(xp->index);
 324
 325         if (xp->lft.hard_add_expires_seconds) {
 326                 time64_t tmo = xp->lft.hard_add_expires_seconds +
 327                         xp->curlft.add_time - now;
 328                 if (tmo <= 0)
 329                         goto expired;
 330                 if (tmo < next)
 331                         next = tmo;
 332         }
 333         if (xp->lft.hard_use_expires_seconds) {
 334                 time64_t tmo = xp->lft.hard_use_expires_seconds +
 335                         (xp->curlft.use_time ? : xp->curlft.add_time) - now;
 336                 if (tmo <= 0)
 337                         goto expired;
 338                 if (tmo < next)
 339                         next = tmo;
 340         }
 341         if (xp->lft.soft_add_expires_seconds) {
 342                 time64_t tmo = xp->lft.soft_add_expires_seconds +
 343                         xp->curlft.add_time - now;
 344                 if (tmo <= 0) {
 345                         warn = 1;
 346                         tmo = XFRM_KM_TIMEOUT;
 347                 }
 348                 if (tmo < next)
 349                         next = tmo;
 350         }
 351         if (xp->lft.soft_use_expires_seconds) {
 352                 time64_t tmo = xp->lft.soft_use_expires_seconds +
 353                         (xp->curlft.use_time ? : xp->curlft.add_time) - now;
 354                 if (tmo <= 0) {
 355                         warn = 1;
 356                         tmo = XFRM_KM_TIMEOUT;
 357                 }
 358                 if (tmo < next)
 359                         next = tmo;
 360         }
 361
 362         if (warn)
 363                 km_policy_expired(xp, dir, 0, 0);
 364         if (next != TIME64_MAX &&
 365             !mod_timer(&xp->timer, jiffies + make_jiffies(next)))
 366                 xfrm_pol_hold(xp);
 367
 368 out:
 369         read_unlock(&xp->lock);
 370         xfrm_pol_put(xp);
 371         return;
 372
 373 expired:
 374         read_unlock(&xp->lock);
 375         if (!xfrm_policy_delete(xp, dir))
 376                 km_policy_expired(xp, dir, 1, 0);
 377         xfrm_pol_put(xp);
 378 }
 379
 380 /* Allocate xfrm_policy. Not used here, it is supposed to be used by pfkeyv2
 381  * SPD calls.
 382  */
 383
 384 struct xfrm_policy *xfrm_policy_alloc(struct net *net, gfp_t gfp)
 385 {
 386         struct xfrm_policy *policy;
 387
 388         policy = kzalloc(sizeof(struct xfrm_policy), gfp);
 389
 390         if (policy) {
 391                 write_pnet(&policy->xp_net, net);
 392                 INIT_LIST_HEAD(&policy->walk.all);
 393                 INIT_HLIST_NODE(&policy->bydst_inexact_list);
 394                 INIT_HLIST_NODE(&policy->bydst);
 395                 INIT_HLIST_NODE(&policy->byidx);
 396                 rwlock_init(&policy->lock);
 397                 refcount_set(&policy->refcnt, 1);
 398                 skb_queue_head_init(&policy->polq.hold_queue);
 399                 timer_setup(&policy->timer, xfrm_policy_timer, 0);
 400                 timer_setup(&policy->polq.hold_timer,
 401                             xfrm_policy_queue_process, 0);
 402         }
 403         return policy;
 404 }
 405 EXPORT_SYMBOL(xfrm_policy_alloc);
 406
 407 static void xfrm_policy_destroy_rcu(struct rcu_head *head)
 408 {
 409         struct xfrm_policy *policy = container_of(head, struct xfrm_policy, rcu);
 410
 411         security_xfrm_policy_free(policy->security);
 412         kfree(policy);
 413 }
 414
 415 /* Destroy xfrm_policy: descendant resources must be released to this moment. */
 416
 417 void xfrm_policy_destroy(struct xfrm_policy *policy)
 418 {
 419         BUG_ON(!policy->walk.dead);
 420
 421         if (del_timer(&policy->timer) || del_timer(&policy->polq.hold_timer))
 422                 BUG();
 423
 424         call_rcu(&policy->rcu, xfrm_policy_destroy_rcu);
 425 }
 426 EXPORT_SYMBOL(xfrm_policy_destroy);
 427
 428 /* Rule must be locked. Release descendant resources, announce
 429  * entry dead. The rule must be unlinked from lists to the moment.
 430  */
 431
 432 static void xfrm_policy_kill(struct xfrm_policy *policy)
 433 {
 434         policy->walk.dead = 1;
 435
 436         atomic_inc(&policy->genid);
 437
 438         if (del_timer(&policy->polq.hold_timer))
 439                 xfrm_pol_put(policy);
 440         skb_queue_purge(&policy->polq.hold_queue);
 441
 442         if (del_timer(&policy->timer))
 443                 xfrm_pol_put(policy);
 444
 445         xfrm_pol_put(policy);
 446 }
 447
 448 static unsigned int xfrm_policy_hashmax __read_mostly = 1 * 1024 * 1024;
 449
 450 static inline unsigned int idx_hash(struct net *net, u32 index)
 451 {
 452         return __idx_hash(index, net->xfrm.policy_idx_hmask);
 453 }
 454
 455 /* calculate policy hash thresholds */
 456 static void __get_hash_thresh(struct net *net,
 457                               unsigned short family, int dir,
 458                               u8 *dbits, u8 *sbits)
 459 {
 460         switch (family) {
 461         case AF_INET:
 462                 *dbits = net->xfrm.policy_bydst[dir].dbits4;
 463                 *sbits = net->xfrm.policy_bydst[dir].sbits4;
 464                 break;
 465
 466         case AF_INET6:
 467                 *dbits = net->xfrm.policy_bydst[dir].dbits6;
 468                 *sbits = net->xfrm.policy_bydst[dir].sbits6;
 469                 break;
 470
 471         default:
 472                 *dbits = 0;
 473                 *sbits = 0;
 474         }
 475 }
 476
 477 static struct hlist_head *policy_hash_bysel(struct net *net,
 478                                             const struct xfrm_selector *sel,
 479                                             unsigned short family, int dir)
 480 {
 481         unsigned int hmask = net->xfrm.policy_bydst[dir].hmask;
 482         unsigned int hash;
 483         u8 dbits;
 484         u8 sbits;
 485
 486         __get_hash_thresh(net, family, dir, &dbits, &sbits);
 487         hash = __sel_hash(sel, family, hmask, dbits, sbits);
 488
 489         if (hash == hmask + 1)
 490                 return NULL;
 491
 492         return rcu_dereference_check(net->xfrm.policy_bydst[dir].table,
 493                      lockdep_is_held(&net->xfrm.xfrm_policy_lock)) + hash;
 494 }
 495
 496 static struct hlist_head *policy_hash_direct(struct net *net,
 497                                              const xfrm_address_t *daddr,
 498                                              const xfrm_address_t *saddr,
 499                                              unsigned short family, int dir)
 500 {
 501         unsigned int hmask = net->xfrm.policy_bydst[dir].hmask;
 502         unsigned int hash;
 503         u8 dbits;
 504         u8 sbits;
 505
 506         __get_hash_thresh(net, family, dir, &dbits, &sbits);
 507         hash = __addr_hash(daddr, saddr, family, hmask, dbits, sbits);
 508
 509         return rcu_dereference_check(net->xfrm.policy_bydst[dir].table,
 510                      lockdep_is_held(&net->xfrm.xfrm_policy_lock)) + hash;
 511 }
 512
 513 static void xfrm_dst_hash_transfer(struct net *net,
 514                                    struct hlist_head *list,
 515                                    struct hlist_head *ndsttable,
 516                                    unsigned int nhashmask,
 517                                    int dir)
 518 {
 519         struct hlist_node *tmp, *entry0 = NULL;
 520         struct xfrm_policy *pol;
 521         unsigned int h0 = 0;
 522         u8 dbits;
 523         u8 sbits;
 524
 525 redo:
 526         hlist_for_each_entry_safe(pol, tmp, list, bydst) {
 527                 unsigned int h;
 528
 529                 __get_hash_thresh(net, pol->family, dir, &dbits, &sbits);
 530                 h = __addr_hash(&pol->selector.daddr, &pol->selector.saddr,
 531                                 pol->family, nhashmask, dbits, sbits);
 532                 if (!entry0) {
 533                         hlist_del_rcu(&pol->bydst);
 534                         hlist_add_head_rcu(&pol->bydst, ndsttable + h);
 535                         h0 = h;
 536                 } else {
 537                         if (h != h0)
 538                                 continue;
 539                         hlist_del_rcu(&pol->bydst);
 540                         hlist_add_behind_rcu(&pol->bydst, entry0);
 541                 }
 542                 entry0 = &pol->bydst;
 543         }
 544         if (!hlist_empty(list)) {
 545                 entry0 = NULL;
 546                 goto redo;
 547         }
 548 }
 549
 550 static void xfrm_idx_hash_transfer(struct hlist_head *list,
 551                                    struct hlist_head *nidxtable,
 552                                    unsigned int nhashmask)
 553 {
 554         struct hlist_node *tmp;
 555         struct xfrm_policy *pol;
 556
 557         hlist_for_each_entry_safe(pol, tmp, list, byidx) {
 558                 unsigned int h;
 559
 560                 h = __idx_hash(pol->index, nhashmask);
 561                 hlist_add_head(&pol->byidx, nidxtable+h);
 562         }
 563 }
 564
 565 static unsigned long xfrm_new_hash_mask(unsigned int old_hmask)
 566 {
 567         return ((old_hmask + 1) << 1) - 1;
 568 }
 569
 570 static void xfrm_bydst_resize(struct net *net, int dir)
 571 {
 572         unsigned int hmask = net->xfrm.policy_bydst[dir].hmask;
 573         unsigned int nhashmask = xfrm_new_hash_mask(hmask);
 574         unsigned int nsize = (nhashmask + 1) * sizeof(struct hlist_head);
 575         struct hlist_head *ndst = xfrm_hash_alloc(nsize);
 576         struct hlist_head *odst;
 577         int i;
 578
 579         if (!ndst)
 580                 return;
 581
 582         spin_lock_bh(&net->xfrm.xfrm_policy_lock);
 583         write_seqcount_begin(&xfrm_policy_hash_generation);
 584
 585         odst = rcu_dereference_protected(net->xfrm.policy_bydst[dir].table,
 586                                 lockdep_is_held(&net->xfrm.xfrm_policy_lock));
 587
 588         for (i = hmask; i >= 0; i--)
 589                 xfrm_dst_hash_transfer(net, odst + i, ndst, nhashmask, dir);
 590
 591         rcu_assign_pointer(net->xfrm.policy_bydst[dir].table, ndst);
 592         net->xfrm.policy_bydst[dir].hmask = nhashmask;
 593
 594         write_seqcount_end(&xfrm_policy_hash_generation);
 595         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
 596
 597         synchronize_rcu();
 598
 599         xfrm_hash_free(odst, (hmask + 1) * sizeof(struct hlist_head));
 600 }
 601
 602 static void xfrm_byidx_resize(struct net *net, int total)
 603 {
 604         unsigned int hmask = net->xfrm.policy_idx_hmask;
 605         unsigned int nhashmask = xfrm_new_hash_mask(hmask);
 606         unsigned int nsize = (nhashmask + 1) * sizeof(struct hlist_head);
 607         struct hlist_head *oidx = net->xfrm.policy_byidx;
 608         struct hlist_head *nidx = xfrm_hash_alloc(nsize);
 609         int i;
 610
 611         if (!nidx)
 612                 return;
 613
 614         spin_lock_bh(&net->xfrm.xfrm_policy_lock);
 615
 616         for (i = hmask; i >= 0; i--)
 617                 xfrm_idx_hash_transfer(oidx + i, nidx, nhashmask);
 618
 619         net->xfrm.policy_byidx = nidx;
 620         net->xfrm.policy_idx_hmask = nhashmask;
 621
 622         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
 623
 624         xfrm_hash_free(oidx, (hmask + 1) * sizeof(struct hlist_head));
 625 }
 626
 627 static inline int xfrm_bydst_should_resize(struct net *net, int dir, int *total)
 628 {
 629         unsigned int cnt = net->xfrm.policy_count[dir];
 630         unsigned int hmask = net->xfrm.policy_bydst[dir].hmask;
 631
 632         if (total)
 633                 *total += cnt;
 634
 635         if ((hmask + 1) < xfrm_policy_hashmax &&
 636             cnt > hmask)
 637                 return 1;
 638
 639         return 0;
 640 }
 641
 642 static inline int xfrm_byidx_should_resize(struct net *net, int total)
 643 {
 644         unsigned int hmask = net->xfrm.policy_idx_hmask;
 645
 646         if ((hmask + 1) < xfrm_policy_hashmax &&
 647             total > hmask)
 648                 return 1;
 649
 650         return 0;
 651 }
 652
 653 void xfrm_spd_getinfo(struct net *net, struct xfrmk_spdinfo *si)
 654 {
 655         si->incnt = net->xfrm.policy_count[XFRM_POLICY_IN];
 656         si->outcnt = net->xfrm.policy_count[XFRM_POLICY_OUT];
 657         si->fwdcnt = net->xfrm.policy_count[XFRM_POLICY_FWD];
 658         si->inscnt = net->xfrm.policy_count[XFRM_POLICY_IN+XFRM_POLICY_MAX];
 659         si->outscnt = net->xfrm.policy_count[XFRM_POLICY_OUT+XFRM_POLICY_MAX];
 660         si->fwdscnt = net->xfrm.policy_count[XFRM_POLICY_FWD+XFRM_POLICY_MAX];
 661         si->spdhcnt = net->xfrm.policy_idx_hmask;
 662         si->spdhmcnt = xfrm_policy_hashmax;
 663 }
 664 EXPORT_SYMBOL(xfrm_spd_getinfo);
 665
 666 static DEFINE_MUTEX(hash_resize_mutex);
 667 static void xfrm_hash_resize(struct work_struct *work)
 668 {
 669         struct net *net = container_of(work, struct net, xfrm.policy_hash_work);
 670         int dir, total;
 671
 672         mutex_lock(&hash_resize_mutex);
 673
 674         total = 0;
 675         for (dir = 0; dir < XFRM_POLICY_MAX; dir++) {
 676                 if (xfrm_bydst_should_resize(net, dir, &total))
 677                         xfrm_bydst_resize(net, dir);
 678         }
 679         if (xfrm_byidx_should_resize(net, total))
 680                 xfrm_byidx_resize(net, total);
 681
 682         mutex_unlock(&hash_resize_mutex);
 683 }
 684
 685 /* Make sure *pol can be inserted into fastbin.
 686  * Useful to check that later insert requests will be sucessful
 687  * (provided xfrm_policy_lock is held throughout).
 688  */
 689 static struct xfrm_pol_inexact_bin *
 690 xfrm_policy_inexact_alloc_bin(const struct xfrm_policy *pol, u8 dir)
 691 {
 692         struct xfrm_pol_inexact_bin *bin, *prev;
 693         struct xfrm_pol_inexact_key k = {
 694                 .family = pol->family,
 695                 .type = pol->type,
 696                 .dir = dir,
 697                 .if_id = pol->if_id,
 698         };
 699         struct net *net = xp_net(pol);
 700
 701         lockdep_assert_held(&net->xfrm.xfrm_policy_lock);
 702
 703         write_pnet(&k.net, net);
 704         bin = rhashtable_lookup_fast(&xfrm_policy_inexact_table, &k,
 705                                      xfrm_pol_inexact_params);
 706         if (bin)
 707                 return bin;
 708
 709         bin = kzalloc(sizeof(*bin), GFP_ATOMIC);
 710         if (!bin)
 711                 return NULL;
 712
 713         bin->k = k;
 714         INIT_HLIST_HEAD(&bin->hhead);
 715         bin->root_d = RB_ROOT;
 716         bin->root_s = RB_ROOT;
 717         seqcount_init(&bin->count);
 718
 719         prev = rhashtable_lookup_get_insert_key(&xfrm_policy_inexact_table,
 720                                                 &bin->k, &bin->head,
 721                                                 xfrm_pol_inexact_params);
 722         if (!prev) {
 723                 list_add(&bin->inexact_bins, &net->xfrm.inexact_bins);
 724                 return bin;
 725         }
 726
 727         kfree(bin);
 728
 729         return IS_ERR(prev) ? NULL : prev;
 730 }
 731
 732 static bool xfrm_pol_inexact_addr_use_any_list(const xfrm_address_t *addr,
 733                                                int family, u8 prefixlen)
 734 {
 735         if (xfrm_addr_any(addr, family))
 736                 return true;
 737
 738         if (family == AF_INET6 && prefixlen < INEXACT_PREFIXLEN_IPV6)
 739                 return true;
 740
 741         if (family == AF_INET && prefixlen < INEXACT_PREFIXLEN_IPV4)
 742                 return true;
 743
 744         return false;
 745 }
 746
 747 static bool
 748 xfrm_policy_inexact_insert_use_any_list(const struct xfrm_policy *policy)
 749 {
 750         const xfrm_address_t *addr;
 751         bool saddr_any, daddr_any;
 752         u8 prefixlen;
 753
 754         addr = &policy->selector.saddr;
 755         prefixlen = policy->selector.prefixlen_s;
 756
 757         saddr_any = xfrm_pol_inexact_addr_use_any_list(addr,
 758                                                        policy->family,
 759                                                        prefixlen);
 760         addr = &policy->selector.daddr;
 761         prefixlen = policy->selector.prefixlen_d;
 762         daddr_any = xfrm_pol_inexact_addr_use_any_list(addr,
 763                                                        policy->family,
 764                                                        prefixlen);
 765         return saddr_any && daddr_any;
 766 }
 767
 768 static void xfrm_pol_inexact_node_init(struct xfrm_pol_inexact_node *node,
 769                                        const xfrm_address_t *addr, u8 prefixlen)
 770 {
 771         node->addr = *addr;
 772         node->prefixlen = prefixlen;
 773 }
 774
 775 static struct xfrm_pol_inexact_node *
 776 xfrm_pol_inexact_node_alloc(const xfrm_address_t *addr, u8 prefixlen)
 777 {
 778         struct xfrm_pol_inexact_node *node;
 779
 780         node = kzalloc(sizeof(*node), GFP_ATOMIC);
 781         if (node)
 782                 xfrm_pol_inexact_node_init(node, addr, prefixlen);
 783
 784         return node;
 785 }
 786
 787 static int xfrm_policy_addr_delta(const xfrm_address_t *a,
 788                                   const xfrm_address_t *b,
 789                                   u8 prefixlen, u16 family)
 790 {
 791         unsigned int pdw, pbi;
 792         int delta = 0;
 793
 794         switch (family) {
 795         case AF_INET:
 796                 if (sizeof(long) == 4 && prefixlen == 0)
 797                         return ntohl(a->a4) - ntohl(b->a4);
 798                 return (ntohl(a->a4) & ((~0UL << (32 - prefixlen)))) -
 799                        (ntohl(b->a4) & ((~0UL << (32 - prefixlen))));
 800         case AF_INET6:
 801                 pdw = prefixlen >> 5;
 802                 pbi = prefixlen & 0x1f;
 803
 804                 if (pdw) {
 805                         delta = memcmp(a->a6, b->a6, pdw << 2);
 806                         if (delta)
 807                                 return delta;
 808                 }
 809                 if (pbi) {
 810                         u32 mask = ~0u << (32 - pbi);
 811
 812                         delta = (ntohl(a->a6[pdw]) & mask) -
 813                                 (ntohl(b->a6[pdw]) & mask);
 814                 }
 815                 break;
 816         default:
 817                 break;
 818         }
 819
 820         return delta;
 821 }
 822
 823 static void xfrm_policy_inexact_list_reinsert(struct net *net,
 824                                               struct xfrm_pol_inexact_node *n,
 825                                               u16 family)
 826 {
 827         unsigned int matched_s, matched_d;
 828         struct xfrm_policy *policy, *p;
 829
 830         matched_s = 0;
 831         matched_d = 0;
 832
 833         list_for_each_entry_reverse(policy, &net->xfrm.policy_all, walk.all) {
 834                 struct hlist_node *newpos = NULL;
 835                 bool matches_s, matches_d;
 836
 837                 if (!policy->bydst_reinsert)
 838                         continue;
 839
 840                 WARN_ON_ONCE(policy->family != family);
 841
 842                 policy->bydst_reinsert = false;
 843                 hlist_for_each_entry(p, &n->hhead, bydst) {
 844                         if (policy->priority > p->priority)
 845                                 newpos = &p->bydst;
 846                         else if (policy->priority == p->priority &&
 847                                  policy->pos > p->pos)
 848                                 newpos = &p->bydst;
 849                         else
 850                                 break;
 851                 }
 852
 853                 if (newpos)
 854                         hlist_add_behind_rcu(&policy->bydst, newpos);
 855                 else
 856                         hlist_add_head_rcu(&policy->bydst, &n->hhead);
 857
 858                 /* paranoia checks follow.
 859                  * Check that the reinserted policy matches at least
 860                  * saddr or daddr for current node prefix.
 861                  *
 862                  * Matching both is fine, matching saddr in one policy
 863                  * (but not daddr) and then matching only daddr in another
 864                  * is a bug.
 865                  */
 866                 matches_s = xfrm_policy_addr_delta(&policy->selector.saddr,
 867                                                    &n->addr,
 868                                                    n->prefixlen,
 869                                                    family) == 0;
 870                 matches_d = xfrm_policy_addr_delta(&policy->selector.daddr,
 871                                                    &n->addr,
 872                                                    n->prefixlen,
 873                                                    family) == 0;
 874                 if (matches_s && matches_d)
 875                         continue;
 876
 877                 WARN_ON_ONCE(!matches_s && !matches_d);
 878                 if (matches_s)
 879                         matched_s++;
 880                 if (matches_d)
 881                         matched_d++;
 882                 WARN_ON_ONCE(matched_s && matched_d);
 883         }
 884 }
 885
 886 static void xfrm_policy_inexact_node_reinsert(struct net *net,
 887                                               struct xfrm_pol_inexact_node *n,
 888                                               struct rb_root *new,
 889                                               u16 family)
 890 {
 891         struct xfrm_pol_inexact_node *node;
 892         struct rb_node **p, *parent;
 893
 894         /* we should not have another subtree here */
 895         WARN_ON_ONCE(!RB_EMPTY_ROOT(&n->root));
 896 restart:
 897         parent = NULL;
 898         p = &new->rb_node;
 899         while (*p) {
 900                 u8 prefixlen;
 901                 int delta;
 902
 903                 parent = *p;
 904                 node = rb_entry(*p, struct xfrm_pol_inexact_node, node);
 905
 906                 prefixlen = min(node->prefixlen, n->prefixlen);
 907
 908                 delta = xfrm_policy_addr_delta(&n->addr, &node->addr,
 909                                                prefixlen, family);
 910                 if (delta < 0) {
 911                         p = &parent->rb_left;
 912                 } else if (delta > 0) {
 913                         p = &parent->rb_right;
 914                 } else {
 915                         bool same_prefixlen = node->prefixlen == n->prefixlen;
 916                         struct xfrm_policy *tmp;
 917
 918                         hlist_for_each_entry(tmp, &n->hhead, bydst) {
 919                                 tmp->bydst_reinsert = true;
 920                                 hlist_del_rcu(&tmp->bydst);
 921                         }
 922
 923                         node->prefixlen = prefixlen;
 924
 925                         xfrm_policy_inexact_list_reinsert(net, node, family);
 926
 927                         if (same_prefixlen) {
 928                                 kfree_rcu(n, rcu);
 929                                 return;
 930                         }
 931
 932                         rb_erase(*p, new);
 933                         kfree_rcu(n, rcu);
 934                         n = node;
 935                         goto restart;
 936                 }
 937         }
 938
 939         rb_link_node_rcu(&n->node, parent, p);
 940         rb_insert_color(&n->node, new);
 941 }
 942
 943 /* merge nodes v and n */
 944 static void xfrm_policy_inexact_node_merge(struct net *net,
 945                                            struct xfrm_pol_inexact_node *v,
 946                                            struct xfrm_pol_inexact_node *n,
 947                                            u16 family)
 948 {
 949         struct xfrm_pol_inexact_node *node;
 950         struct xfrm_policy *tmp;
 951         struct rb_node *rnode;
 952
 953         /* To-be-merged node v has a subtree.
 954          *
 955          * Dismantle it and insert its nodes to n->root.
 956          */
 957         while ((rnode = rb_first(&v->root)) != NULL) {
 958                 node = rb_entry(rnode, struct xfrm_pol_inexact_node, node);
 959                 rb_erase(&node->node, &v->root);
 960                 xfrm_policy_inexact_node_reinsert(net, node, &n->root,
 961                                                   family);
 962         }
 963
 964         hlist_for_each_entry(tmp, &v->hhead, bydst) {
 965                 tmp->bydst_reinsert = true;
 966                 hlist_del_rcu(&tmp->bydst);
 967         }
 968
 969         xfrm_policy_inexact_list_reinsert(net, n, family);
 970 }
 971
 972 static struct xfrm_pol_inexact_node *
 973 xfrm_policy_inexact_insert_node(struct net *net,
 974                                 struct rb_root *root,
 975                                 xfrm_address_t *addr,
 976                                 u16 family, u8 prefixlen, u8 dir)
 977 {
 978         struct xfrm_pol_inexact_node *cached = NULL;
 979         struct rb_node **p, *parent = NULL;
 980         struct xfrm_pol_inexact_node *node;
 981
 982         p = &root->rb_node;
 983         while (*p) {
 984                 int delta;
 985
 986                 parent = *p;
 987                 node = rb_entry(*p, struct xfrm_pol_inexact_node, node);
 988
 989                 delta = xfrm_policy_addr_delta(addr, &node->addr,
 990                                                node->prefixlen,
 991                                                family);
 992                 if (delta == 0 && prefixlen >= node->prefixlen) {
 993                         WARN_ON_ONCE(cached); /* ipsec policies got lost */
 994                         return node;
 995                 }
 996
 997                 if (delta < 0)
 998                         p = &parent->rb_left;
 999                 else
1000                         p = &parent->rb_right;
1001
1002                 if (prefixlen < node->prefixlen) {
1003                         delta = xfrm_policy_addr_delta(addr, &node->addr,
1004                                                        prefixlen,
1005                                                        family);
1006                         if (delta)
1007                                 continue;
1008
1009                         /* This node is a subnet of the new prefix. It needs
1010                          * to be removed and re-inserted with the smaller
1011                          * prefix and all nodes that are now also covered
1012                          * by the reduced prefixlen.
1013                          */
1014                         rb_erase(&node->node, root);
1015
1016                         if (!cached) {
1017                                 xfrm_pol_inexact_node_init(node, addr,
1018                                                            prefixlen);
1019                                 cached = node;
1020                         } else {
1021                                 /* This node also falls within the new
1022                                  * prefixlen. Merge the to-be-reinserted
1023                                  * node and this one.
1024                                  */
1025                                 xfrm_policy_inexact_node_merge(net, node,
1026                                                                cached, family);
1027                                 kfree_rcu(node, rcu);
1028                         }
1029
1030                         /* restart */
1031                         p = &root->rb_node;
1032                         parent = NULL;
1033                 }
1034         }
1035
1036         node = cached;
1037         if (!node) {
1038                 node = xfrm_pol_inexact_node_alloc(addr, prefixlen);
1039                 if (!node)
1040                         return NULL;
1041         }
1042
1043         rb_link_node_rcu(&node->node, parent, p);
1044         rb_insert_color(&node->node, root);
1045
1046         return node;
1047 }
1048
1049 static void xfrm_policy_inexact_gc_tree(struct rb_root *r, bool rm)
1050 {
1051         struct xfrm_pol_inexact_node *node;
1052         struct rb_node *rn = rb_first(r);
1053
1054         while (rn) {
1055                 node = rb_entry(rn, struct xfrm_pol_inexact_node, node);
1056
1057                 xfrm_policy_inexact_gc_tree(&node->root, rm);
1058                 rn = rb_next(rn);
1059
1060                 if (!hlist_empty(&node->hhead) || !RB_EMPTY_ROOT(&node->root)) {
1061                         WARN_ON_ONCE(rm);
1062                         continue;
1063                 }
1064
1065                 rb_erase(&node->node, r);
1066                 kfree_rcu(node, rcu);
1067         }
1068 }
1069
1070 static void __xfrm_policy_inexact_prune_bin(struct xfrm_pol_inexact_bin *b, bool net_exit)
1071 {
1072         write_seqcount_begin(&b->count);
1073         xfrm_policy_inexact_gc_tree(&b->root_d, net_exit);
1074         xfrm_policy_inexact_gc_tree(&b->root_s, net_exit);
1075         write_seqcount_end(&b->count);
1076
1077         if (!RB_EMPTY_ROOT(&b->root_d) || !RB_EMPTY_ROOT(&b->root_s) ||
1078             !hlist_empty(&b->hhead)) {
1079                 WARN_ON_ONCE(net_exit);
1080                 return;
1081         }
1082
1083         if (rhashtable_remove_fast(&xfrm_policy_inexact_table, &b->head,
1084                                    xfrm_pol_inexact_params) == 0) {
1085                 list_del(&b->inexact_bins);
1086                 kfree_rcu(b, rcu);
1087         }
1088 }
1089
1090 static void xfrm_policy_inexact_prune_bin(struct xfrm_pol_inexact_bin *b)
1091 {
1092         struct net *net = read_pnet(&b->k.net);
1093
1094         spin_lock_bh(&net->xfrm.xfrm_policy_lock);
1095         __xfrm_policy_inexact_prune_bin(b, false);
1096         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
1097 }
1098
1099 static void __xfrm_policy_inexact_flush(struct net *net)
1100 {
1101         struct xfrm_pol_inexact_bin *bin, *t;
1102
1103         lockdep_assert_held(&net->xfrm.xfrm_policy_lock);
1104
1105         list_for_each_entry_safe(bin, t, &net->xfrm.inexact_bins, inexact_bins)
1106                 __xfrm_policy_inexact_prune_bin(bin, false);
1107 }
1108
1109 static struct hlist_head *
1110 xfrm_policy_inexact_alloc_chain(struct xfrm_pol_inexact_bin *bin,
1111                                 struct xfrm_policy *policy, u8 dir)
1112 {
1113         struct xfrm_pol_inexact_node *n;
1114         struct net *net;
1115
1116         net = xp_net(policy);
1117         lockdep_assert_held(&net->xfrm.xfrm_policy_lock);
1118
1119         if (xfrm_policy_inexact_insert_use_any_list(policy))
1120                 return &bin->hhead;
1121
1122         if (xfrm_pol_inexact_addr_use_any_list(&policy->selector.daddr,
1123                                                policy->family,
1124                                                policy->selector.prefixlen_d)) {
1125                 write_seqcount_begin(&bin->count);
1126                 n = xfrm_policy_inexact_insert_node(net,
1127                                                     &bin->root_s,
1128                                                     &policy->selector.saddr,
1129                                                     policy->family,
1130                                                     policy->selector.prefixlen_s,
1131                                                     dir);
1132                 write_seqcount_end(&bin->count);
1133                 if (!n)
1134                         return NULL;
1135
1136                 return &n->hhead;
1137         }
1138
1139         /* daddr is fixed */
1140         write_seqcount_begin(&bin->count);
1141         n = xfrm_policy_inexact_insert_node(net,
1142                                             &bin->root_d,
1143                                             &policy->selector.daddr,
1144                                             policy->family,
1145                                             policy->selector.prefixlen_d, dir);
1146         write_seqcount_end(&bin->count);
1147         if (!n)
1148                 return NULL;
1149
1150         /* saddr is wildcard */
1151         if (xfrm_pol_inexact_addr_use_any_list(&policy->selector.saddr,
1152                                                policy->family,
1153                                                policy->selector.prefixlen_s))
1154                 return &n->hhead;
1155
1156         write_seqcount_begin(&bin->count);
1157         n = xfrm_policy_inexact_insert_node(net,
1158                                             &n->root,
1159                                             &policy->selector.saddr,
1160                                             policy->family,
1161                                             policy->selector.prefixlen_s, dir);
1162         write_seqcount_end(&bin->count);
1163         if (!n)
1164                 return NULL;
1165
1166         return &n->hhead;
1167 }
1168
1169 static struct xfrm_policy *
1170 xfrm_policy_inexact_insert(struct xfrm_policy *policy, u8 dir, int excl)
1171 {
1172         struct xfrm_pol_inexact_bin *bin;
1173         struct xfrm_policy *delpol;
1174         struct hlist_head *chain;
1175         struct net *net;
1176
1177         bin = xfrm_policy_inexact_alloc_bin(policy, dir);
1178         if (!bin)
1179                 return ERR_PTR(-ENOMEM);
1180
1181         net = xp_net(policy);
1182         lockdep_assert_held(&net->xfrm.xfrm_policy_lock);
1183
1184         chain = xfrm_policy_inexact_alloc_chain(bin, policy, dir);
1185         if (!chain) {
1186                 __xfrm_policy_inexact_prune_bin(bin, false);
1187                 return ERR_PTR(-ENOMEM);
1188         }
1189
1190         delpol = xfrm_policy_insert_list(chain, policy, excl);
1191         if (delpol && excl) {
1192                 __xfrm_policy_inexact_prune_bin(bin, false);
1193                 return ERR_PTR(-EEXIST);
1194         }
1195
1196         chain = &net->xfrm.policy_inexact[dir];
1197         xfrm_policy_insert_inexact_list(chain, policy);
1198
1199         if (delpol)
1200                 __xfrm_policy_inexact_prune_bin(bin, false);
1201
1202         return delpol;
1203 }
1204
1205 static void xfrm_hash_rebuild(struct work_struct *work)
1206 {
1207         struct net *net = container_of(work, struct net,
1208                                        xfrm.policy_hthresh.work);
1209         unsigned int hmask;
1210         struct xfrm_policy *pol;
1211         struct xfrm_policy *policy;
1212         struct hlist_head *chain;
1213         struct hlist_head *odst;
1214         struct hlist_node *newpos;
1215         int i;
1216         int dir;
1217         unsigned seq;
1218         u8 lbits4, rbits4, lbits6, rbits6;
1219
1220         mutex_lock(&hash_resize_mutex);
1221
1222         /* read selector prefixlen thresholds */
1223         do {
1224                 seq = read_seqbegin(&net->xfrm.policy_hthresh.lock);
1225
1226                 lbits4 = net->xfrm.policy_hthresh.lbits4;
1227                 rbits4 = net->xfrm.policy_hthresh.rbits4;
1228                 lbits6 = net->xfrm.policy_hthresh.lbits6;
1229                 rbits6 = net->xfrm.policy_hthresh.rbits6;
1230         } while (read_seqretry(&net->xfrm.policy_hthresh.lock, seq));
1231
1232         spin_lock_bh(&net->xfrm.xfrm_policy_lock);
1233         write_seqcount_begin(&xfrm_policy_hash_generation);
1234
1235         /* make sure that we can insert the indirect policies again before
1236          * we start with destructive action.
1237          */
1238         list_for_each_entry(policy, &net->xfrm.policy_all, walk.all) {
1239                 struct xfrm_pol_inexact_bin *bin;
1240                 u8 dbits, sbits;
1241
1242                 dir = xfrm_policy_id2dir(policy->index);
1243                 if (policy->walk.dead || dir >= XFRM_POLICY_MAX)
1244                         continue;
1245
1246                 if ((dir & XFRM_POLICY_MASK) == XFRM_POLICY_OUT) {
1247                         if (policy->family == AF_INET) {
1248                                 dbits = rbits4;
1249                                 sbits = lbits4;
1250                         } else {
1251                                 dbits = rbits6;
1252                                 sbits = lbits6;
1253                         }
1254                 } else {
1255                         if (policy->family == AF_INET) {
1256                                 dbits = lbits4;
1257                                 sbits = rbits4;
1258                         } else {
1259                                 dbits = lbits6;
1260                                 sbits = rbits6;
1261                         }
1262                 }
1263
1264                 if (policy->selector.prefixlen_d < dbits ||
1265                     policy->selector.prefixlen_s < sbits)
1266                         continue;
1267
1268                 bin = xfrm_policy_inexact_alloc_bin(policy, dir);
1269                 if (!bin)
1270                         goto out_unlock;
1271
1272                 if (!xfrm_policy_inexact_alloc_chain(bin, policy, dir))
1273                         goto out_unlock;
1274         }
1275
1276         /* reset the bydst and inexact table in all directions */
1277         for (dir = 0; dir < XFRM_POLICY_MAX; dir++) {
1278                 struct hlist_node *n;
1279
1280                 hlist_for_each_entry_safe(policy, n,
1281                                           &net->xfrm.policy_inexact[dir],
1282                                           bydst_inexact_list) {
1283                         hlist_del_rcu(&policy->bydst);
1284                         hlist_del_init(&policy->bydst_inexact_list);
1285                 }
1286
1287                 hmask = net->xfrm.policy_bydst[dir].hmask;
1288                 odst = net->xfrm.policy_bydst[dir].table;
1289                 for (i = hmask; i >= 0; i--) {
1290                         hlist_for_each_entry_safe(policy, n, odst + i, bydst)
1291                                 hlist_del_rcu(&policy->bydst);
1292                 }
1293                 if ((dir & XFRM_POLICY_MASK) == XFRM_POLICY_OUT) {
1294                         /* dir out => dst = remote, src = local */
1295                         net->xfrm.policy_bydst[dir].dbits4 = rbits4;
1296                         net->xfrm.policy_bydst[dir].sbits4 = lbits4;
1297                         net->xfrm.policy_bydst[dir].dbits6 = rbits6;
1298                         net->xfrm.policy_bydst[dir].sbits6 = lbits6;
1299                 } else {
1300                         /* dir in/fwd => dst = local, src = remote */
1301                         net->xfrm.policy_bydst[dir].dbits4 = lbits4;
1302                         net->xfrm.policy_bydst[dir].sbits4 = rbits4;
1303                         net->xfrm.policy_bydst[dir].dbits6 = lbits6;
1304                         net->xfrm.policy_bydst[dir].sbits6 = rbits6;
1305                 }
1306         }
1307
1308         /* re-insert all policies by order of creation */
1309         list_for_each_entry_reverse(policy, &net->xfrm.policy_all, walk.all) {
1310                 if (policy->walk.dead)
1311                         continue;
1312                 dir = xfrm_policy_id2dir(policy->index);
1313                 if (dir >= XFRM_POLICY_MAX) {
1314                         /* skip socket policies */
1315                         continue;
1316                 }
1317                 newpos = NULL;
1318                 chain = policy_hash_bysel(net, &policy->selector,
1319                                           policy->family, dir);
1320
1321                 if (!chain) {
1322                         void *p = xfrm_policy_inexact_insert(policy, dir, 0);
1323
1324                         WARN_ONCE(IS_ERR(p), "reinsert: %ld\n", PTR_ERR(p));
1325                         continue;
1326                 }
1327
1328                 hlist_for_each_entry(pol, chain, bydst) {
1329                         if (policy->priority >= pol->priority)
1330                                 newpos = &pol->bydst;
1331                         else
1332                                 break;
1333                 }
1334                 if (newpos)
1335                         hlist_add_behind_rcu(&policy->bydst, newpos);
1336                 else
1337                         hlist_add_head_rcu(&policy->bydst, chain);
1338         }
1339
1340 out_unlock:
1341         __xfrm_policy_inexact_flush(net);
1342         write_seqcount_end(&xfrm_policy_hash_generation);
1343         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
1344
1345         mutex_unlock(&hash_resize_mutex);
1346 }
1347
1348 void xfrm_policy_hash_rebuild(struct net *net)
1349 {
1350         schedule_work(&net->xfrm.policy_hthresh.work);
1351 }
1352 EXPORT_SYMBOL(xfrm_policy_hash_rebuild);
1353
1354 /* Generate new index... KAME seems to generate them ordered by cost
1355  * of an absolute inpredictability of ordering of rules. This will not pass. */
1356 static u32 xfrm_gen_index(struct net *net, int dir, u32 index)
1357 {
1358         static u32 idx_generator;
1359
1360         for (;;) {
1361                 struct hlist_head *list;
1362                 struct xfrm_policy *p;
1363                 u32 idx;
1364                 int found;
1365
1366                 if (!index) {
1367                         idx = (idx_generator | dir);
1368                         idx_generator += 8;
1369                 } else {
1370                         idx = index;
1371                         index = 0;
1372                 }
1373
1374                 if (idx == 0)
1375                         idx = 8;
1376                 list = net->xfrm.policy_byidx + idx_hash(net, idx);
1377                 found = 0;
1378                 hlist_for_each_entry(p, list, byidx) {
1379                         if (p->index == idx) {
1380                                 found = 1;
1381                                 break;
1382                         }
1383                 }
1384                 if (!found)
1385                         return idx;
1386         }
1387 }
1388
1389 static inline int selector_cmp(struct xfrm_selector *s1, struct xfrm_selector *s2)
1390 {
1391         u32 *p1 = (u32 *) s1;
1392         u32 *p2 = (u32 *) s2;
1393         int len = sizeof(struct xfrm_selector) / sizeof(u32);
1394         int i;
1395
1396         for (i = 0; i < len; i++) {
1397                 if (p1[i] != p2[i])
1398                         return 1;
1399         }
1400
1401         return 0;
1402 }
1403
1404 static void xfrm_policy_requeue(struct xfrm_policy *old,
1405                                 struct xfrm_policy *new)
1406 {
1407         struct xfrm_policy_queue *pq = &old->polq;
1408         struct sk_buff_head list;
1409
1410         if (skb_queue_empty(&pq->hold_queue))
1411                 return;
1412
1413         __skb_queue_head_init(&list);
1414
1415         spin_lock_bh(&pq->hold_queue.lock);
1416         skb_queue_splice_init(&pq->hold_queue, &list);
1417         if (del_timer(&pq->hold_timer))
1418                 xfrm_pol_put(old);
1419         spin_unlock_bh(&pq->hold_queue.lock);
1420
1421         pq = &new->polq;
1422
1423         spin_lock_bh(&pq->hold_queue.lock);
1424         skb_queue_splice(&list, &pq->hold_queue);
1425         pq->timeout = XFRM_QUEUE_TMO_MIN;
1426         if (!mod_timer(&pq->hold_timer, jiffies))
1427                 xfrm_pol_hold(new);
1428         spin_unlock_bh(&pq->hold_queue.lock);
1429 }
1430
1431 static bool xfrm_policy_mark_match(struct xfrm_policy *policy,
1432                                    struct xfrm_policy *pol)
1433 {
1434         u32 mark = policy->mark.v & policy->mark.m;
1435
1436         if (policy->mark.v == pol->mark.v && policy->mark.m == pol->mark.m)
1437                 return true;
1438
1439         if ((mark & pol->mark.m) == pol->mark.v &&
1440             policy->priority == pol->priority)
1441                 return true;
1442
1443         return false;
1444 }
1445
1446 static u32 xfrm_pol_bin_key(const void *data, u32 len, u32 seed)
1447 {
1448         const struct xfrm_pol_inexact_key *k = data;
1449         u32 a = k->type << 24 | k->dir << 16 | k->family;
1450
1451         return jhash_3words(a, k->if_id, net_hash_mix(read_pnet(&k->net)),
1452                             seed);
1453 }
1454
1455 static u32 xfrm_pol_bin_obj(const void *data, u32 len, u32 seed)
1456 {
1457         const struct xfrm_pol_inexact_bin *b = data;
1458
1459         return xfrm_pol_bin_key(&b->k, 0, seed);
1460 }
1461
1462 static int xfrm_pol_bin_cmp(struct rhashtable_compare_arg *arg,
1463                             const void *ptr)
1464 {
1465         const struct xfrm_pol_inexact_key *key = arg->key;
1466         const struct xfrm_pol_inexact_bin *b = ptr;
1467         int ret;
1468
1469         if (!net_eq(read_pnet(&b->k.net), read_pnet(&key->net)))
1470                 return -1;
1471
1472         ret = b->k.dir ^ key->dir;
1473         if (ret)
1474                 return ret;
1475
1476         ret = b->k.type ^ key->type;
1477         if (ret)
1478                 return ret;
1479
1480         ret = b->k.family ^ key->family;
1481         if (ret)
1482                 return ret;
1483
1484         return b->k.if_id ^ key->if_id;
1485 }
1486
1487 static const struct rhashtable_params xfrm_pol_inexact_params = {
1488         .head_offset            = offsetof(struct xfrm_pol_inexact_bin, head),
1489         .hashfn                 = xfrm_pol_bin_key,
1490         .obj_hashfn             = xfrm_pol_bin_obj,
1491         .obj_cmpfn              = xfrm_pol_bin_cmp,
1492         .automatic_shrinking    = true,
1493 };
1494
1495 static void xfrm_policy_insert_inexact_list(struct hlist_head *chain,
1496                                             struct xfrm_policy *policy)
1497 {
1498         struct xfrm_policy *pol, *delpol = NULL;
1499         struct hlist_node *newpos = NULL;
1500         int i = 0;
1501
1502         hlist_for_each_entry(pol, chain, bydst_inexact_list) {
1503                 if (pol->type == policy->type &&
1504                     pol->if_id == policy->if_id &&
1505                     !selector_cmp(&pol->selector, &policy->selector) &&
1506                     xfrm_policy_mark_match(policy, pol) &&
1507                     xfrm_sec_ctx_match(pol->security, policy->security) &&
1508                     !WARN_ON(delpol)) {
1509                         delpol = pol;
1510                         if (policy->priority > pol->priority)
1511                                 continue;
1512                 } else if (policy->priority >= pol->priority) {
1513                         newpos = &pol->bydst_inexact_list;
1514                         continue;
1515                 }
1516                 if (delpol)
1517                         break;
1518         }
1519
1520         if (newpos)
1521                 hlist_add_behind_rcu(&policy->bydst_inexact_list, newpos);
1522         else
1523                 hlist_add_head_rcu(&policy->bydst_inexact_list, chain);
1524
1525         hlist_for_each_entry(pol, chain, bydst_inexact_list) {
1526                 pol->pos = i;
1527                 i++;
1528         }
1529 }
1530
1531 static struct xfrm_policy *xfrm_policy_insert_list(struct hlist_head *chain,
1532                                                    struct xfrm_policy *policy,
1533                                                    bool excl)
1534 {
1535         struct xfrm_policy *pol, *newpos = NULL, *delpol = NULL;
1536
1537         hlist_for_each_entry(pol, chain, bydst) {
1538                 if (pol->type == policy->type &&
1539                     pol->if_id == policy->if_id &&
1540                     !selector_cmp(&pol->selector, &policy->selector) &&
1541                     xfrm_policy_mark_match(policy, pol) &&
1542                     xfrm_sec_ctx_match(pol->security, policy->security) &&
1543                     !WARN_ON(delpol)) {
1544                         if (excl)
1545                                 return ERR_PTR(-EEXIST);
1546                         delpol = pol;
1547                         if (policy->priority > pol->priority)
1548                                 continue;
1549                 } else if (policy->priority >= pol->priority) {
1550                         newpos = pol;
1551                         continue;
1552                 }
1553                 if (delpol)
1554                         break;
1555         }
1556
1557         if (newpos)
1558                 hlist_add_behind_rcu(&policy->bydst, &newpos->bydst);
1559         else
1560                 hlist_add_head_rcu(&policy->bydst, chain);
1561
1562         return delpol;
1563 }
1564
1565 int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl)
1566 {
1567         struct net *net = xp_net(policy);
1568         struct xfrm_policy *delpol;
1569         struct hlist_head *chain;
1570
1571         spin_lock_bh(&net->xfrm.xfrm_policy_lock);
1572         chain = policy_hash_bysel(net, &policy->selector, policy->family, dir);
1573         if (chain)
1574                 delpol = xfrm_policy_insert_list(chain, policy, excl);
1575         else
1576                 delpol = xfrm_policy_inexact_insert(policy, dir, excl);
1577
1578         if (IS_ERR(delpol)) {
1579                 spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
1580                 return PTR_ERR(delpol);
1581         }
1582
1583         __xfrm_policy_link(policy, dir);
1584
1585         /* After previous checking, family can either be AF_INET or AF_INET6 */
1586         if (policy->family == AF_INET)
1587                 rt_genid_bump_ipv4(net);
1588         else
1589                 rt_genid_bump_ipv6(net);
1590
1591         if (delpol) {
1592                 xfrm_policy_requeue(delpol, policy);
1593                 __xfrm_policy_unlink(delpol, dir);
1594         }
1595         policy->index = delpol ? delpol->index : xfrm_gen_index(net, dir, policy->index);
1596         hlist_add_head(&policy->byidx, net->xfrm.policy_byidx+idx_hash(net, policy->index));
1597         policy->curlft.add_time = ktime_get_real_seconds();
1598         policy->curlft.use_time = 0;
1599         if (!mod_timer(&policy->timer, jiffies + HZ))
1600                 xfrm_pol_hold(policy);
1601         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
1602
1603         if (delpol)
1604                 xfrm_policy_kill(delpol);
1605         else if (xfrm_bydst_should_resize(net, dir, NULL))
1606                 schedule_work(&net->xfrm.policy_hash_work);
1607
1608         return 0;
1609 }
1610 EXPORT_SYMBOL(xfrm_policy_insert);
1611
1612 static struct xfrm_policy *
1613 __xfrm_policy_bysel_ctx(struct hlist_head *chain, u32 mark, u32 if_id,
1614                         u8 type, int dir,
1615                         struct xfrm_selector *sel,
1616                         struct xfrm_sec_ctx *ctx)
1617 {
1618         struct xfrm_policy *pol;
1619
1620         if (!chain)
1621                 return NULL;
1622
1623         hlist_for_each_entry(pol, chain, bydst) {
1624                 if (pol->type == type &&
1625                     pol->if_id == if_id &&
1626                     (mark & pol->mark.m) == pol->mark.v &&
1627                     !selector_cmp(sel, &pol->selector) &&
1628                     xfrm_sec_ctx_match(ctx, pol->security))
1629                         return pol;
1630         }
1631
1632         return NULL;
1633 }
1634
1635 struct xfrm_policy *xfrm_policy_bysel_ctx(struct net *net, u32 mark, u32 if_id,
1636                                           u8 type, int dir,
1637                                           struct xfrm_selector *sel,
1638                                           struct xfrm_sec_ctx *ctx, int delete,
1639                                           int *err)
1640 {
1641         struct xfrm_pol_inexact_bin *bin = NULL;
1642         struct xfrm_policy *pol, *ret = NULL;
1643         struct hlist_head *chain;
1644
1645         *err = 0;
1646         spin_lock_bh(&net->xfrm.xfrm_policy_lock);
1647         chain = policy_hash_bysel(net, sel, sel->family, dir);
1648         if (!chain) {
1649                 struct xfrm_pol_inexact_candidates cand;
1650                 int i;
1651
1652                 bin = xfrm_policy_inexact_lookup(net, type,
1653                                                  sel->family, dir, if_id);
1654                 if (!bin) {
1655                         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
1656                         return NULL;
1657                 }
1658
1659                 if (!xfrm_policy_find_inexact_candidates(&cand, bin,
1660                                                          &sel->saddr,
1661                                                          &sel->daddr)) {
1662                         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
1663                         return NULL;
1664                 }
1665
1666                 pol = NULL;
1667                 for (i = 0; i < ARRAY_SIZE(cand.res); i++) {
1668                         struct xfrm_policy *tmp;
1669
1670                         tmp = __xfrm_policy_bysel_ctx(cand.res[i], mark,
1671                                                       if_id, type, dir,
1672                                                       sel, ctx);
1673                         if (!tmp)
1674                                 continue;
1675
1676                         if (!pol || tmp->pos < pol->pos)
1677                                 pol = tmp;
1678                 }
1679         } else {
1680                 pol = __xfrm_policy_bysel_ctx(chain, mark, if_id, type, dir,
1681                                               sel, ctx);
1682         }
1683
1684         if (pol) {
1685                 xfrm_pol_hold(pol);
1686                 if (delete) {
1687                         *err = security_xfrm_policy_delete(pol->security);
1688                         if (*err) {
1689                                 spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
1690                                 return pol;
1691                         }
1692                         __xfrm_policy_unlink(pol, dir);
1693                 }
1694                 ret = pol;
1695         }
1696         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
1697
1698         if (ret && delete)
1699                 xfrm_policy_kill(ret);
1700         if (bin && delete)
1701                 xfrm_policy_inexact_prune_bin(bin);
1702         return ret;
1703 }
1704 EXPORT_SYMBOL(xfrm_policy_bysel_ctx);
1705
1706 struct xfrm_policy *xfrm_policy_byid(struct net *net, u32 mark, u32 if_id,
1707                                      u8 type, int dir, u32 id, int delete,
1708                                      int *err)
1709 {
1710         struct xfrm_policy *pol, *ret;
1711         struct hlist_head *chain;
1712
1713         *err = -ENOENT;
1714         if (xfrm_policy_id2dir(id) != dir)
1715                 return NULL;
1716
1717         *err = 0;
1718         spin_lock_bh(&net->xfrm.xfrm_policy_lock);
1719         chain = net->xfrm.policy_byidx + idx_hash(net, id);
1720         ret = NULL;
1721         hlist_for_each_entry(pol, chain, byidx) {
1722                 if (pol->type == type && pol->index == id &&
1723                     pol->if_id == if_id &&
1724                     (mark & pol->mark.m) == pol->mark.v) {
1725                         xfrm_pol_hold(pol);
1726                         if (delete) {
1727                                 *err = security_xfrm_policy_delete(
1728                                                                 pol->security);
1729                                 if (*err) {
1730                                         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
1731                                         return pol;
1732                                 }
1733                                 __xfrm_policy_unlink(pol, dir);
1734                         }
1735                         ret = pol;
1736                         break;
1737                 }
1738         }
1739         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
1740
1741         if (ret && delete)
1742                 xfrm_policy_kill(ret);
1743         return ret;
1744 }
1745 EXPORT_SYMBOL(xfrm_policy_byid);
1746
1747 #ifdef CONFIG_SECURITY_NETWORK_XFRM
1748 static inline int
1749 xfrm_policy_flush_secctx_check(struct net *net, u8 type, bool task_valid)
1750 {
1751         struct xfrm_policy *pol;
1752         int err = 0;
1753
1754         list_for_each_entry(pol, &net->xfrm.policy_all, walk.all) {
1755                 if (pol->walk.dead ||
1756                     xfrm_policy_id2dir(pol->index) >= XFRM_POLICY_MAX ||
1757                     pol->type != type)
1758                         continue;
1759
1760                 err = security_xfrm_policy_delete(pol->security);
1761                 if (err) {
1762                         xfrm_audit_policy_delete(pol, 0, task_valid);
1763                         return err;
1764                 }
1765         }
1766         return err;
1767 }
1768 #else
1769 static inline int
1770 xfrm_policy_flush_secctx_check(struct net *net, u8 type, bool task_valid)
1771 {
1772         return 0;
1773 }
1774 #endif
1775
1776 int xfrm_policy_flush(struct net *net, u8 type, bool task_valid)
1777 {
1778         int dir, err = 0, cnt = 0;
1779         struct xfrm_policy *pol;
1780
1781         spin_lock_bh(&net->xfrm.xfrm_policy_lock);
1782
1783         err = xfrm_policy_flush_secctx_check(net, type, task_valid);
1784         if (err)
1785                 goto out;
1786
1787 again:
1788         list_for_each_entry(pol, &net->xfrm.policy_all, walk.all) {
1789                 dir = xfrm_policy_id2dir(pol->index);
1790                 if (pol->walk.dead ||
1791                     dir >= XFRM_POLICY_MAX ||
1792                     pol->type != type)
1793                         continue;
1794
1795                 __xfrm_policy_unlink(pol, dir);
1796                 spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
1797                 cnt++;
1798                 xfrm_audit_policy_delete(pol, 1, task_valid);
1799                 xfrm_policy_kill(pol);
1800                 spin_lock_bh(&net->xfrm.xfrm_policy_lock);
1801                 goto again;
1802         }
1803         if (cnt)
1804                 __xfrm_policy_inexact_flush(net);
1805         else
1806                 err = -ESRCH;
1807 out:
1808         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
1809         return err;
1810 }
1811 EXPORT_SYMBOL(xfrm_policy_flush);
1812
1813 int xfrm_policy_walk(struct net *net, struct xfrm_policy_walk *walk,
1814                      int (*func)(struct xfrm_policy *, int, int, void*),
1815                      void *data)
1816 {
1817         struct xfrm_policy *pol;
1818         struct xfrm_policy_walk_entry *x;
1819         int error = 0;
1820
1821         if (walk->type >= XFRM_POLICY_TYPE_MAX &&
1822             walk->type != XFRM_POLICY_TYPE_ANY)
1823                 return -EINVAL;
1824
1825         if (list_empty(&walk->walk.all) && walk->seq != 0)
1826                 return 0;
1827
1828         spin_lock_bh(&net->xfrm.xfrm_policy_lock);
1829         if (list_empty(&walk->walk.all))
1830                 x = list_first_entry(&net->xfrm.policy_all, struct xfrm_policy_walk_entry, all);
1831         else
1832                 x = list_first_entry(&walk->walk.all,
1833                                      struct xfrm_policy_walk_entry, all);
1834
1835         list_for_each_entry_from(x, &net->xfrm.policy_all, all) {
1836                 if (x->dead)
1837                         continue;
1838                 pol = container_of(x, struct xfrm_policy, walk);
1839                 if (walk->type != XFRM_POLICY_TYPE_ANY &&
1840                     walk->type != pol->type)
1841                         continue;
1842                 error = func(pol, xfrm_policy_id2dir(pol->index),
1843                              walk->seq, data);
1844                 if (error) {
1845                         list_move_tail(&walk->walk.all, &x->all);
1846                         goto out;
1847                 }
1848                 walk->seq++;
1849         }
1850         if (walk->seq == 0) {
1851                 error = -ENOENT;
1852                 goto out;
1853         }
1854         list_del_init(&walk->walk.all);
1855 out:
1856         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
1857         return error;
1858 }
1859 EXPORT_SYMBOL(xfrm_policy_walk);
1860
1861 void xfrm_policy_walk_init(struct xfrm_policy_walk *walk, u8 type)
1862 {
1863         INIT_LIST_HEAD(&walk->walk.all);
1864         walk->walk.dead = 1;
1865         walk->type = type;
1866         walk->seq = 0;
1867 }
1868 EXPORT_SYMBOL(xfrm_policy_walk_init);
1869
1870 void xfrm_policy_walk_done(struct xfrm_policy_walk *walk, struct net *net)
1871 {
1872         if (list_empty(&walk->walk.all))
1873                 return;
1874
1875         spin_lock_bh(&net->xfrm.xfrm_policy_lock); /*FIXME where is net? */
1876         list_del(&walk->walk.all);
1877         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
1878 }
1879 EXPORT_SYMBOL(xfrm_policy_walk_done);
1880
1881 /*
1882  * Find policy to apply to this flow.
1883  *
1884  * Returns 0 if policy found, else an -errno.
1885  */
1886 static int xfrm_policy_match(const struct xfrm_policy *pol,
1887                              const struct flowi *fl,
1888                              u8 type, u16 family, int dir, u32 if_id)
1889 {
1890         const struct xfrm_selector *sel = &pol->selector;
1891         int ret = -ESRCH;
1892         bool match;
1893
1894         if (pol->family != family ||
1895             pol->if_id != if_id ||
1896             (fl->flowi_mark & pol->mark.m) != pol->mark.v ||
1897             pol->type != type)
1898                 return ret;
1899
1900         match = xfrm_selector_match(sel, fl, family);
1901         if (match)
1902                 ret = security_xfrm_policy_lookup(pol->security, fl->flowi_secid,
1903                                                   dir);
1904         return ret;
1905 }
1906
1907 static struct xfrm_pol_inexact_node *
1908 xfrm_policy_lookup_inexact_addr(const struct rb_root *r,
1909                                 seqcount_t *count,
1910                                 const xfrm_address_t *addr, u16 family)
1911 {
1912         const struct rb_node *parent;
1913         int seq;
1914
1915 again:
1916         seq = read_seqcount_begin(count);
1917
1918         parent = rcu_dereference_raw(r->rb_node);
1919         while (parent) {
1920                 struct xfrm_pol_inexact_node *node;
1921                 int delta;
1922
1923                 node = rb_entry(parent, struct xfrm_pol_inexact_node, node);
1924
1925                 delta = xfrm_policy_addr_delta(addr, &node->addr,
1926                                                node->prefixlen, family);
1927                 if (delta < 0) {
1928                         parent = rcu_dereference_raw(parent->rb_left);
1929                         continue;
1930                 } else if (delta > 0) {
1931                         parent = rcu_dereference_raw(parent->rb_right);
1932                         continue;
1933                 }
1934
1935                 return node;
1936         }
1937
1938         if (read_seqcount_retry(count, seq))
1939                 goto again;
1940
1941         return NULL;
1942 }
1943
1944 static bool
1945 xfrm_policy_find_inexact_candidates(struct xfrm_pol_inexact_candidates *cand,
1946                                     struct xfrm_pol_inexact_bin *b,
1947                                     const xfrm_address_t *saddr,
1948                                     const xfrm_address_t *daddr)
1949 {
1950         struct xfrm_pol_inexact_node *n;
1951         u16 family;
1952
1953         if (!b)
1954                 return false;
1955
1956         family = b->k.family;
1957         memset(cand, 0, sizeof(*cand));
1958         cand->res[XFRM_POL_CAND_ANY] = &b->hhead;
1959
1960         n = xfrm_policy_lookup_inexact_addr(&b->root_d, &b->count, daddr,
1961                                             family);
1962         if (n) {
1963                 cand->res[XFRM_POL_CAND_DADDR] = &n->hhead;
1964                 n = xfrm_policy_lookup_inexact_addr(&n->root, &b->count, saddr,
1965                                                     family);
1966                 if (n)
1967                         cand->res[XFRM_POL_CAND_BOTH] = &n->hhead;
1968         }
1969
1970         n = xfrm_policy_lookup_inexact_addr(&b->root_s, &b->count, saddr,
1971                                             family);
1972         if (n)
1973                 cand->res[XFRM_POL_CAND_SADDR] = &n->hhead;
1974
1975         return true;
1976 }
1977
1978 static struct xfrm_pol_inexact_bin *
1979 xfrm_policy_inexact_lookup_rcu(struct net *net, u8 type, u16 family,
1980                                u8 dir, u32 if_id)
1981 {
1982         struct xfrm_pol_inexact_key k = {
1983                 .family = family,
1984                 .type = type,
1985                 .dir = dir,
1986                 .if_id = if_id,
1987         };
1988
1989         write_pnet(&k.net, net);
1990
1991         return rhashtable_lookup(&xfrm_policy_inexact_table, &k,
1992                                  xfrm_pol_inexact_params);
1993 }
1994
1995 static struct xfrm_pol_inexact_bin *
1996 xfrm_policy_inexact_lookup(struct net *net, u8 type, u16 family,
1997                            u8 dir, u32 if_id)
1998 {
1999         struct xfrm_pol_inexact_bin *bin;
2000
2001         lockdep_assert_held(&net->xfrm.xfrm_policy_lock);
2002
2003         rcu_read_lock();
2004         bin = xfrm_policy_inexact_lookup_rcu(net, type, family, dir, if_id);
2005         rcu_read_unlock();
2006
2007         return bin;
2008 }
2009
2010 static struct xfrm_policy *
2011 __xfrm_policy_eval_candidates(struct hlist_head *chain,
2012                               struct xfrm_policy *prefer,
2013                               const struct flowi *fl,
2014                               u8 type, u16 family, int dir, u32 if_id)
2015 {
2016         u32 priority = prefer ? prefer->priority : ~0u;
2017         struct xfrm_policy *pol;
2018
2019         if (!chain)
2020                 return NULL;
2021
2022         hlist_for_each_entry_rcu(pol, chain, bydst) {
2023                 int err;
2024
2025                 if (pol->priority > priority)
2026                         break;
2027
2028                 err = xfrm_policy_match(pol, fl, type, family, dir, if_id);
2029                 if (err) {
2030                         if (err != -ESRCH)
2031                                 return ERR_PTR(err);
2032
2033                         continue;
2034                 }
2035
2036                 if (prefer) {
2037                         /* matches.  Is it older than *prefer? */
2038                         if (pol->priority == priority &&
2039                             prefer->pos < pol->pos)
2040                                 return prefer;
2041                 }
2042
2043                 return pol;
2044         }
2045
2046         return NULL;
2047 }
2048
2049 static struct xfrm_policy *
2050 xfrm_policy_eval_candidates(struct xfrm_pol_inexact_candidates *cand,
2051                             struct xfrm_policy *prefer,
2052                             const struct flowi *fl,
2053                             u8 type, u16 family, int dir, u32 if_id)
2054 {
2055         struct xfrm_policy *tmp;
2056         int i;
2057
2058         for (i = 0; i < ARRAY_SIZE(cand->res); i++) {
2059                 tmp = __xfrm_policy_eval_candidates(cand->res[i],
2060                                                     prefer,
2061                                                     fl, type, family, dir,
2062                                                     if_id);
2063                 if (!tmp)
2064                         continue;
2065
2066                 if (IS_ERR(tmp))
2067                         return tmp;
2068                 prefer = tmp;
2069         }
2070
2071         return prefer;
2072 }
2073
2074 static struct xfrm_policy *xfrm_policy_lookup_bytype(struct net *net, u8 type,
2075                                                      const struct flowi *fl,
2076                                                      u16 family, u8 dir,
2077                                                      u32 if_id)
2078 {
2079         struct xfrm_pol_inexact_candidates cand;
2080         const xfrm_address_t *daddr, *saddr;
2081         struct xfrm_pol_inexact_bin *bin;
2082         struct xfrm_policy *pol, *ret;
2083         struct hlist_head *chain;
2084         unsigned int sequence;
2085         int err;
2086
2087         daddr = xfrm_flowi_daddr(fl, family);
2088         saddr = xfrm_flowi_saddr(fl, family);
2089         if (unlikely(!daddr || !saddr))
2090                 return NULL;
2091
2092         rcu_read_lock();
2093  retry:
2094         do {
2095                 sequence = read_seqcount_begin(&xfrm_policy_hash_generation);
2096                 chain = policy_hash_direct(net, daddr, saddr, family, dir);
2097         } while (read_seqcount_retry(&xfrm_policy_hash_generation, sequence));
2098
2099         ret = NULL;
2100         hlist_for_each_entry_rcu(pol, chain, bydst) {
2101                 err = xfrm_policy_match(pol, fl, type, family, dir, if_id);
2102                 if (err) {
2103                         if (err == -ESRCH)
2104                                 continue;
2105                         else {
2106                                 ret = ERR_PTR(err);
2107                                 goto fail;
2108                         }
2109                 } else {
2110                         ret = pol;
2111                         break;
2112                 }
2113         }
2114         bin = xfrm_policy_inexact_lookup_rcu(net, type, family, dir, if_id);
2115         if (!bin || !xfrm_policy_find_inexact_candidates(&cand, bin, saddr,
2116                                                          daddr))
2117                 goto skip_inexact;
2118
2119         pol = xfrm_policy_eval_candidates(&cand, ret, fl, type,
2120                                           family, dir, if_id);
2121         if (pol) {
2122                 ret = pol;
2123                 if (IS_ERR(pol))
2124                         goto fail;
2125         }
2126
2127 skip_inexact:
2128         if (read_seqcount_retry(&xfrm_policy_hash_generation, sequence))
2129                 goto retry;
2130
2131         if (ret && !xfrm_pol_hold_rcu(ret))
2132                 goto retry;
2133 fail:
2134         rcu_read_unlock();
2135
2136         return ret;
2137 }
2138
2139 static struct xfrm_policy *xfrm_policy_lookup(struct net *net,
2140                                               const struct flowi *fl,
2141                                               u16 family, u8 dir, u32 if_id)
2142 {
2143 #ifdef CONFIG_XFRM_SUB_POLICY
2144         struct xfrm_policy *pol;
2145
2146         pol = xfrm_policy_lookup_bytype(net, XFRM_POLICY_TYPE_SUB, fl, family,
2147                                         dir, if_id);
2148         if (pol != NULL)
2149                 return pol;
2150 #endif
2151         return xfrm_policy_lookup_bytype(net, XFRM_POLICY_TYPE_MAIN, fl, family,
2152                                          dir, if_id);
2153 }
2154
2155 static struct xfrm_policy *xfrm_sk_policy_lookup(const struct sock *sk, int dir,
2156                                                  const struct flowi *fl,
2157                                                  u16 family, u32 if_id)
2158 {
2159         struct xfrm_policy *pol;
2160
2161         rcu_read_lock();
2162  again:
2163         pol = rcu_dereference(sk->sk_policy[dir]);
2164         if (pol != NULL) {
2165                 bool match;
2166                 int err = 0;
2167
2168                 if (pol->family != family) {
2169                         pol = NULL;
2170                         goto out;
2171                 }
2172
2173                 match = xfrm_selector_match(&pol->selector, fl, family);
2174                 if (match) {
2175                         if ((sk->sk_mark & pol->mark.m) != pol->mark.v ||
2176                             pol->if_id != if_id) {
2177                                 pol = NULL;
2178                                 goto out;
2179                         }
2180                         err = security_xfrm_policy_lookup(pol->security,
2181                                                       fl->flowi_secid,
2182                                                       dir);
2183                         if (!err) {
2184                                 if (!xfrm_pol_hold_rcu(pol))
2185                                         goto again;
2186                         } else if (err == -ESRCH) {
2187                                 pol = NULL;
2188                         } else {
2189                                 pol = ERR_PTR(err);
2190                         }
2191                 } else
2192                         pol = NULL;
2193         }
2194 out:
2195         rcu_read_unlock();
2196         return pol;
2197 }
2198
2199 static void __xfrm_policy_link(struct xfrm_policy *pol, int dir)
2200 {
2201         struct net *net = xp_net(pol);
2202
2203         list_add(&pol->walk.all, &net->xfrm.policy_all);
2204         net->xfrm.policy_count[dir]++;
2205         xfrm_pol_hold(pol);
2206 }
2207
2208 static struct xfrm_policy *__xfrm_policy_unlink(struct xfrm_policy *pol,
2209                                                 int dir)
2210 {
2211         struct net *net = xp_net(pol);
2212
2213         if (list_empty(&pol->walk.all))
2214                 return NULL;
2215
2216         /* Socket policies are not hashed. */
2217         if (!hlist_unhashed(&pol->bydst)) {
2218                 hlist_del_rcu(&pol->bydst);
2219                 hlist_del_init(&pol->bydst_inexact_list);
2220                 hlist_del(&pol->byidx);
2221         }
2222
2223         list_del_init(&pol->walk.all);
2224         net->xfrm.policy_count[dir]--;
2225
2226         return pol;
2227 }
2228
2229 static void xfrm_sk_policy_link(struct xfrm_policy *pol, int dir)
2230 {
2231         __xfrm_policy_link(pol, XFRM_POLICY_MAX + dir);
2232 }
2233
2234 static void xfrm_sk_policy_unlink(struct xfrm_policy *pol, int dir)
2235 {
2236         __xfrm_policy_unlink(pol, XFRM_POLICY_MAX + dir);
2237 }
2238
2239 int xfrm_policy_delete(struct xfrm_policy *pol, int dir)
2240 {
2241         struct net *net = xp_net(pol);
2242
2243         spin_lock_bh(&net->xfrm.xfrm_policy_lock);
2244         pol = __xfrm_policy_unlink(pol, dir);
2245         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
2246         if (pol) {
2247                 xfrm_policy_kill(pol);
2248                 return 0;
2249         }
2250         return -ENOENT;
2251 }
2252 EXPORT_SYMBOL(xfrm_policy_delete);
2253
2254 int xfrm_sk_policy_insert(struct sock *sk, int dir, struct xfrm_policy *pol)
2255 {
2256         struct net *net = sock_net(sk);
2257         struct xfrm_policy *old_pol;
2258
2259 #ifdef CONFIG_XFRM_SUB_POLICY
2260         if (pol && pol->type != XFRM_POLICY_TYPE_MAIN)
2261                 return -EINVAL;
2262 #endif
2263
2264         spin_lock_bh(&net->xfrm.xfrm_policy_lock);
2265         old_pol = rcu_dereference_protected(sk->sk_policy[dir],
2266                                 lockdep_is_held(&net->xfrm.xfrm_policy_lock));
2267         if (pol) {
2268                 pol->curlft.add_time = ktime_get_real_seconds();
2269                 pol->index = xfrm_gen_index(net, XFRM_POLICY_MAX+dir, 0);
2270                 xfrm_sk_policy_link(pol, dir);
2271         }
2272         rcu_assign_pointer(sk->sk_policy[dir], pol);
2273         if (old_pol) {
2274                 if (pol)
2275                         xfrm_policy_requeue(old_pol, pol);
2276
2277                 /* Unlinking succeeds always. This is the only function
2278                  * allowed to delete or replace socket policy.
2279                  */
2280                 xfrm_sk_policy_unlink(old_pol, dir);
2281         }
2282         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
2283
2284         if (old_pol) {
2285                 xfrm_policy_kill(old_pol);
2286         }
2287         return 0;
2288 }
2289
2290 static struct xfrm_policy *clone_policy(const struct xfrm_policy *old, int dir)
2291 {
2292         struct xfrm_policy *newp = xfrm_policy_alloc(xp_net(old), GFP_ATOMIC);
2293         struct net *net = xp_net(old);
2294
2295         if (newp) {
2296                 newp->selector = old->selector;
2297                 if (security_xfrm_policy_clone(old->security,
2298                                                &newp->security)) {
2299                         kfree(newp);
2300                         return NULL;  /* ENOMEM */
2301                 }
2302                 newp->lft = old->lft;
2303                 newp->curlft = old->curlft;
2304                 newp->mark = old->mark;
2305                 newp->if_id = old->if_id;
2306                 newp->action = old->action;
2307                 newp->flags = old->flags;
2308                 newp->xfrm_nr = old->xfrm_nr;
2309                 newp->index = old->index;
2310                 newp->type = old->type;
2311                 newp->family = old->family;
2312                 memcpy(newp->xfrm_vec, old->xfrm_vec,
2313                        newp->xfrm_nr*sizeof(struct xfrm_tmpl));
2314                 spin_lock_bh(&net->xfrm.xfrm_policy_lock);
2315                 xfrm_sk_policy_link(newp, dir);
2316                 spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
2317                 xfrm_pol_put(newp);
2318         }
2319         return newp;
2320 }
2321
2322 int __xfrm_sk_clone_policy(struct sock *sk, const struct sock *osk)
2323 {
2324         const struct xfrm_policy *p;
2325         struct xfrm_policy *np;
2326         int i, ret = 0;
2327
2328         rcu_read_lock();
2329         for (i = 0; i < 2; i++) {
2330                 p = rcu_dereference(osk->sk_policy[i]);
2331                 if (p) {
2332                         np = clone_policy(p, i);
2333                         if (unlikely(!np)) {
2334                                 ret = -ENOMEM;
2335                                 break;
2336                         }
2337                         rcu_assign_pointer(sk->sk_policy[i], np);
2338                 }
2339         }
2340         rcu_read_unlock();
2341         return ret;
2342 }
2343
2344 static int
2345 xfrm_get_saddr(struct net *net, int oif, xfrm_address_t *local,
2346                xfrm_address_t *remote, unsigned short family, u32 mark)
2347 {
2348         int err;
2349         const struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family);
2350
2351         if (unlikely(afinfo == NULL))
2352                 return -EINVAL;
2353         err = afinfo->get_saddr(net, oif, local, remote, mark);
2354         rcu_read_unlock();
2355         return err;
2356 }
2357
2358 /* Resolve list of templates for the flow, given policy. */
2359
2360 static int
2361 xfrm_tmpl_resolve_one(struct xfrm_policy *policy, const struct flowi *fl,
2362                       struct xfrm_state **xfrm, unsigned short family)
2363 {
2364         struct net *net = xp_net(policy);
2365         int nx;
2366         int i, error;
2367         xfrm_address_t *daddr = xfrm_flowi_daddr(fl, family);
2368         xfrm_address_t *saddr = xfrm_flowi_saddr(fl, family);
2369         xfrm_address_t tmp;
2370
2371         for (nx = 0, i = 0; i < policy->xfrm_nr; i++) {
2372                 struct xfrm_state *x;
2373                 xfrm_address_t *remote = daddr;
2374                 xfrm_address_t *local  = saddr;
2375                 struct xfrm_tmpl *tmpl = &policy->xfrm_vec[i];
2376
2377                 if (tmpl->mode == XFRM_MODE_TUNNEL ||
2378                     tmpl->mode == XFRM_MODE_BEET) {
2379                         remote = &tmpl->id.daddr;
2380                         local = &tmpl->saddr;
2381                         if (xfrm_addr_any(local, tmpl->encap_family)) {
2382                                 error = xfrm_get_saddr(net, fl->flowi_oif,
2383                                                        &tmp, remote,
2384                                                        tmpl->encap_family, 0);
2385                                 if (error)
2386                                         goto fail;
2387                                 local = &tmp;
2388                         }
2389                 }
2390
2391                 x = xfrm_state_find(remote, local, fl, tmpl, policy, &error,
2392                                     family, policy->if_id);
2393
2394                 if (x && x->km.state == XFRM_STATE_VALID) {
2395                         xfrm[nx++] = x;
2396                         daddr = remote;
2397                         saddr = local;
2398                         continue;
2399                 }
2400                 if (x) {
2401                         error = (x->km.state == XFRM_STATE_ERROR ?
2402                                  -EINVAL : -EAGAIN);
2403                         xfrm_state_put(x);
2404                 } else if (error == -ESRCH) {
2405                         error = -EAGAIN;
2406                 }
2407
2408                 if (!tmpl->optional)
2409                         goto fail;
2410         }
2411         return nx;
2412
2413 fail:
2414         for (nx--; nx >= 0; nx--)
2415                 xfrm_state_put(xfrm[nx]);
2416         return error;
2417 }
2418
2419 static int
2420 xfrm_tmpl_resolve(struct xfrm_policy **pols, int npols, const struct flowi *fl,
2421                   struct xfrm_state **xfrm, unsigned short family)
2422 {
2423         struct xfrm_state *tp[XFRM_MAX_DEPTH];
2424         struct xfrm_state **tpp = (npols > 1) ? tp : xfrm;
2425         int cnx = 0;
2426         int error;
2427         int ret;
2428         int i;
2429
2430         for (i = 0; i < npols; i++) {
2431                 if (cnx + pols[i]->xfrm_nr >= XFRM_MAX_DEPTH) {
2432                         error = -ENOBUFS;
2433                         goto fail;
2434                 }
2435
2436                 ret = xfrm_tmpl_resolve_one(pols[i], fl, &tpp[cnx], family);
2437                 if (ret < 0) {
2438                         error = ret;
2439                         goto fail;
2440                 } else
2441                         cnx += ret;
2442         }
2443
2444         /* found states are sorted for outbound processing */
2445         if (npols > 1)
2446                 xfrm_state_sort(xfrm, tpp, cnx, family);
2447
2448         return cnx;
2449
2450  fail:
2451         for (cnx--; cnx >= 0; cnx--)
2452                 xfrm_state_put(tpp[cnx]);
2453         return error;
2454
2455 }
2456
2457 static int xfrm_get_tos(const struct flowi *fl, int family)
2458 {
2459         if (family == AF_INET)
2460                 return IPTOS_RT_MASK & fl->u.ip4.flowi4_tos;
2461
2462         return 0;
2463 }
2464
2465 static inline struct xfrm_dst *xfrm_alloc_dst(struct net *net, int family)
2466 {
2467         const struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family);
2468         struct dst_ops *dst_ops;
2469         struct xfrm_dst *xdst;
2470
2471         if (!afinfo)
2472                 return ERR_PTR(-EINVAL);
2473
2474         switch (family) {
2475         case AF_INET:
2476                 dst_ops = &net->xfrm.xfrm4_dst_ops;
2477                 break;
2478 #if IS_ENABLED(CONFIG_IPV6)
2479         case AF_INET6:
2480                 dst_ops = &net->xfrm.xfrm6_dst_ops;
2481                 break;
2482 #endif
2483         default:
2484                 BUG();
2485         }
2486         xdst = dst_alloc(dst_ops, NULL, 1, DST_OBSOLETE_NONE, 0);
2487
2488         if (likely(xdst)) {
2489                 struct dst_entry *dst = &xdst->u.dst;
2490
2491                 memset(dst + 1, 0, sizeof(*xdst) - sizeof(*dst));
2492         } else
2493                 xdst = ERR_PTR(-ENOBUFS);
2494
2495         rcu_read_unlock();
2496
2497         return xdst;
2498 }
2499
2500 static void xfrm_init_path(struct xfrm_dst *path, struct dst_entry *dst,
2501                            int nfheader_len)
2502 {
2503         if (dst->ops->family == AF_INET6) {
2504                 struct rt6_info *rt = (struct rt6_info *)dst;
2505                 path->path_cookie = rt6_get_cookie(rt);
2506                 path->u.rt6.rt6i_nfheader_len = nfheader_len;
2507         }
2508 }
2509
2510 static inline int xfrm_fill_dst(struct xfrm_dst *xdst, struct net_device *dev,
2511                                 const struct flowi *fl)
2512 {
2513         const struct xfrm_policy_afinfo *afinfo =
2514                 xfrm_policy_get_afinfo(xdst->u.dst.ops->family);
2515         int err;
2516
2517         if (!afinfo)
2518                 return -EINVAL;
2519
2520         err = afinfo->fill_dst(xdst, dev, fl);
2521
2522         rcu_read_unlock();
2523
2524         return err;
2525 }
2526
2527
2528 /* Allocate chain of dst_entry's, attach known xfrm's, calculate
2529  * all the metrics... Shortly, bundle a bundle.
2530  */
2531
2532 static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy,
2533                                             struct xfrm_state **xfrm,
2534                                             struct xfrm_dst **bundle,
2535                                             int nx,
2536                                             const struct flowi *fl,
2537                                             struct dst_entry *dst)
2538 {
2539         const struct xfrm_state_afinfo *afinfo;
2540         const struct xfrm_mode *inner_mode;
2541         struct net *net = xp_net(policy);
2542         unsigned long now = jiffies;
2543         struct net_device *dev;
2544         struct xfrm_dst *xdst_prev = NULL;
2545         struct xfrm_dst *xdst0 = NULL;
2546         int i = 0;
2547         int err;
2548         int header_len = 0;
2549         int nfheader_len = 0;
2550         int trailer_len = 0;
2551         int tos;
2552         int family = policy->selector.family;
2553         xfrm_address_t saddr, daddr;
2554
2555         xfrm_flowi_addr_get(fl, &saddr, &daddr, family);
2556
2557         tos = xfrm_get_tos(fl, family);
2558
2559         dst_hold(dst);
2560
2561         for (; i < nx; i++) {
2562                 struct xfrm_dst *xdst = xfrm_alloc_dst(net, family);
2563                 struct dst_entry *dst1 = &xdst->u.dst;
2564
2565                 err = PTR_ERR(xdst);
2566                 if (IS_ERR(xdst)) {
2567                         dst_release(dst);
2568                         goto put_states;
2569                 }
2570
2571                 bundle[i] = xdst;
2572                 if (!xdst_prev)
2573                         xdst0 = xdst;
2574                 else
2575                         /* Ref count is taken during xfrm_alloc_dst()
2576                          * No need to do dst_clone() on dst1
2577                          */
2578                         xfrm_dst_set_child(xdst_prev, &xdst->u.dst);
2579
2580                 if (xfrm[i]->sel.family == AF_UNSPEC) {
2581                         inner_mode = xfrm_ip2inner_mode(xfrm[i],
2582                                                         xfrm_af2proto(family));
2583                         if (!inner_mode) {
2584                                 err = -EAFNOSUPPORT;
2585                                 dst_release(dst);
2586                                 goto put_states;
2587                         }
2588                 } else
2589                         inner_mode = &xfrm[i]->inner_mode;
2590
2591                 xdst->route = dst;
2592                 dst_copy_metrics(dst1, dst);
2593
2594                 if (xfrm[i]->props.mode != XFRM_MODE_TRANSPORT) {
2595                         __u32 mark = 0;
2596
2597                         if (xfrm[i]->props.smark.v || xfrm[i]->props.smark.m)
2598                                 mark = xfrm_smark_get(fl->flowi_mark, xfrm[i]);
2599
2600                         family = xfrm[i]->props.family;
2601                         dst = xfrm_dst_lookup(xfrm[i], tos, fl->flowi_oif,
2602                                               &saddr, &daddr, family, mark);
2603                         err = PTR_ERR(dst);
2604                         if (IS_ERR(dst))
2605                                 goto put_states;
2606                 } else
2607                         dst_hold(dst);
2608
2609                 dst1->xfrm = xfrm[i];
2610                 xdst->xfrm_genid = xfrm[i]->genid;
2611
2612                 dst1->obsolete = DST_OBSOLETE_FORCE_CHK;
2613                 dst1->flags |= DST_HOST;
2614                 dst1->lastuse = now;
2615
2616                 dst1->input = dst_discard;
2617
2618                 rcu_read_lock();
2619                 afinfo = xfrm_state_afinfo_get_rcu(inner_mode->family);
2620                 if (likely(afinfo))
2621                         dst1->output = afinfo->output;
2622                 else
2623                         dst1->output = dst_discard_out;
2624                 rcu_read_unlock();
2625
2626                 xdst_prev = xdst;
2627
2628                 header_len += xfrm[i]->props.header_len;
2629                 if (xfrm[i]->type->flags & XFRM_TYPE_NON_FRAGMENT)
2630                         nfheader_len += xfrm[i]->props.header_len;
2631                 trailer_len += xfrm[i]->props.trailer_len;
2632         }
2633
2634         xfrm_dst_set_child(xdst_prev, dst);
2635         xdst0->path = dst;
2636
2637         err = -ENODEV;
2638         dev = dst->dev;
2639         if (!dev)
2640                 goto free_dst;
2641
2642         xfrm_init_path(xdst0, dst, nfheader_len);
2643         xfrm_init_pmtu(bundle, nx);
2644
2645         for (xdst_prev = xdst0; xdst_prev != (struct xfrm_dst *)dst;
2646              xdst_prev = (struct xfrm_dst *) xfrm_dst_child(&xdst_prev->u.dst)) {
2647                 err = xfrm_fill_dst(xdst_prev, dev, fl);
2648                 if (err)
2649                         goto free_dst;
2650
2651                 xdst_prev->u.dst.header_len = header_len;
2652                 xdst_prev->u.dst.trailer_len = trailer_len;
2653                 header_len -= xdst_prev->u.dst.xfrm->props.header_len;
2654                 trailer_len -= xdst_prev->u.dst.xfrm->props.trailer_len;
2655         }
2656
2657         return &xdst0->u.dst;
2658
2659 put_states:
2660         for (; i < nx; i++)
2661                 xfrm_state_put(xfrm[i]);
2662 free_dst:
2663         if (xdst0)
2664                 dst_release_immediate(&xdst0->u.dst);
2665
2666         return ERR_PTR(err);
2667 }
2668
2669 static int xfrm_expand_policies(const struct flowi *fl, u16 family,
2670                                 struct xfrm_policy **pols,
2671                                 int *num_pols, int *num_xfrms)
2672 {
2673         int i;
2674
2675         if (*num_pols == 0 || !pols[0]) {
2676                 *num_pols = 0;
2677                 *num_xfrms = 0;
2678                 return 0;
2679         }
2680         if (IS_ERR(pols[0]))
2681                 return PTR_ERR(pols[0]);
2682
2683         *num_xfrms = pols[0]->xfrm_nr;
2684
2685 #ifdef CONFIG_XFRM_SUB_POLICY
2686         if (pols[0] && pols[0]->action == XFRM_POLICY_ALLOW &&
2687             pols[0]->type != XFRM_POLICY_TYPE_MAIN) {
2688                 pols[1] = xfrm_policy_lookup_bytype(xp_net(pols[0]),
2689                                                     XFRM_POLICY_TYPE_MAIN,
2690                                                     fl, family,
2691                                                     XFRM_POLICY_OUT,
2692                                                     pols[0]->if_id);
2693                 if (pols[1]) {
2694                         if (IS_ERR(pols[1])) {
2695                                 xfrm_pols_put(pols, *num_pols);
2696                                 return PTR_ERR(pols[1]);
2697                         }
2698                         (*num_pols)++;
2699                         (*num_xfrms) += pols[1]->xfrm_nr;
2700                 }
2701         }
2702 #endif
2703         for (i = 0; i < *num_pols; i++) {
2704                 if (pols[i]->action != XFRM_POLICY_ALLOW) {
2705                         *num_xfrms = -1;
2706                         break;
2707                 }
2708         }
2709
2710         return 0;
2711
2712 }
2713
2714 static struct xfrm_dst *
2715 xfrm_resolve_and_create_bundle(struct xfrm_policy **pols, int num_pols,
2716                                const struct flowi *fl, u16 family,
2717                                struct dst_entry *dst_orig)
2718 {
2719         struct net *net = xp_net(pols[0]);
2720         struct xfrm_state *xfrm[XFRM_MAX_DEPTH];
2721         struct xfrm_dst *bundle[XFRM_MAX_DEPTH];
2722         struct xfrm_dst *xdst;
2723         struct dst_entry *dst;
2724         int err;
2725
2726         /* Try to instantiate a bundle */
2727         err = xfrm_tmpl_resolve(pols, num_pols, fl, xfrm, family);
2728         if (err <= 0) {
2729                 if (err == 0)
2730                         return NULL;
2731
2732                 if (err != -EAGAIN)
2733                         XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTPOLERROR);
2734                 return ERR_PTR(err);
2735         }
2736
2737         dst = xfrm_bundle_create(pols[0], xfrm, bundle, err, fl, dst_orig);
2738         if (IS_ERR(dst)) {
2739                 XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTBUNDLEGENERROR);
2740                 return ERR_CAST(dst);
2741         }
2742
2743         xdst = (struct xfrm_dst *)dst;
2744         xdst->num_xfrms = err;
2745         xdst->num_pols = num_pols;
2746         memcpy(xdst->pols, pols, sizeof(struct xfrm_policy *) * num_pols);
2747         xdst->policy_genid = atomic_read(&pols[0]->genid);
2748
2749         return xdst;
2750 }
2751
2752 static void xfrm_policy_queue_process(struct timer_list *t)
2753 {
2754         struct sk_buff *skb;
2755         struct sock *sk;
2756         struct dst_entry *dst;
2757         struct xfrm_policy *pol = from_timer(pol, t, polq.hold_timer);
2758         struct net *net = xp_net(pol);
2759         struct xfrm_policy_queue *pq = &pol->polq;
2760         struct flowi fl;
2761         struct sk_buff_head list;
2762
2763         spin_lock(&pq->hold_queue.lock);
2764         skb = skb_peek(&pq->hold_queue);
2765         if (!skb) {
2766                 spin_unlock(&pq->hold_queue.lock);
2767                 goto out;
2768         }
2769         dst = skb_dst(skb);
2770         sk = skb->sk;
2771         xfrm_decode_session(skb, &fl, dst->ops->family);
2772         spin_unlock(&pq->hold_queue.lock);
2773
2774         dst_hold(xfrm_dst_path(dst));
2775         dst = xfrm_lookup(net, xfrm_dst_path(dst), &fl, sk, XFRM_LOOKUP_QUEUE);
2776         if (IS_ERR(dst))
2777                 goto purge_queue;
2778
2779         if (dst->flags & DST_XFRM_QUEUE) {
2780                 dst_release(dst);
2781
2782                 if (pq->timeout >= XFRM_QUEUE_TMO_MAX)
2783                         goto purge_queue;
2784
2785                 pq->timeout = pq->timeout << 1;
2786                 if (!mod_timer(&pq->hold_timer, jiffies + pq->timeout))
2787                         xfrm_pol_hold(pol);
2788                 goto out;
2789         }
2790
2791         dst_release(dst);
2792
2793         __skb_queue_head_init(&list);
2794
2795         spin_lock(&pq->hold_queue.lock);
2796         pq->timeout = 0;
2797         skb_queue_splice_init(&pq->hold_queue, &list);
2798         spin_unlock(&pq->hold_queue.lock);
2799
2800         while (!skb_queue_empty(&list)) {
2801                 skb = __skb_dequeue(&list);
2802
2803                 xfrm_decode_session(skb, &fl, skb_dst(skb)->ops->family);
2804                 dst_hold(xfrm_dst_path(skb_dst(skb)));
2805                 dst = xfrm_lookup(net, xfrm_dst_path(skb_dst(skb)), &fl, skb->sk, 0);
2806                 if (IS_ERR(dst)) {
2807                         kfree_skb(skb);
2808                         continue;
2809                 }
2810
2811                 nf_reset(skb);
2812                 skb_dst_drop(skb);
2813                 skb_dst_set(skb, dst);
2814
2815                 dst_output(net, skb->sk, skb);
2816         }
2817
2818 out:
2819         xfrm_pol_put(pol);
2820         return;
2821
2822 purge_queue:
2823         pq->timeout = 0;
2824         skb_queue_purge(&pq->hold_queue);
2825         xfrm_pol_put(pol);
2826 }
2827
2828 static int xdst_queue_output(struct net *net, struct sock *sk, struct sk_buff *skb)
2829 {
2830         unsigned long sched_next;
2831         struct dst_entry *dst = skb_dst(skb);
2832         struct xfrm_dst *xdst = (struct xfrm_dst *) dst;
2833         struct xfrm_policy *pol = xdst->pols[0];
2834         struct xfrm_policy_queue *pq = &pol->polq;
2835
2836         if (unlikely(skb_fclone_busy(sk, skb))) {
2837                 kfree_skb(skb);
2838                 return 0;
2839         }
2840
2841         if (pq->hold_queue.qlen > XFRM_MAX_QUEUE_LEN) {
2842                 kfree_skb(skb);
2843                 return -EAGAIN;
2844         }
2845
2846         skb_dst_force(skb);
2847
2848         spin_lock_bh(&pq->hold_queue.lock);
2849
2850         if (!pq->timeout)
2851                 pq->timeout = XFRM_QUEUE_TMO_MIN;
2852
2853         sched_next = jiffies + pq->timeout;
2854
2855         if (del_timer(&pq->hold_timer)) {
2856                 if (time_before(pq->hold_timer.expires, sched_next))
2857                         sched_next = pq->hold_timer.expires;
2858                 xfrm_pol_put(pol);
2859         }
2860
2861         __skb_queue_tail(&pq->hold_queue, skb);
2862         if (!mod_timer(&pq->hold_timer, sched_next))
2863                 xfrm_pol_hold(pol);
2864
2865         spin_unlock_bh(&pq->hold_queue.lock);
2866
2867         return 0;
2868 }
2869
2870 static struct xfrm_dst *xfrm_create_dummy_bundle(struct net *net,
2871                                                  struct xfrm_flo *xflo,
2872                                                  const struct flowi *fl,
2873                                                  int num_xfrms,
2874                                                  u16 family)
2875 {
2876         int err;
2877         struct net_device *dev;
2878         struct dst_entry *dst;
2879         struct dst_entry *dst1;
2880         struct xfrm_dst *xdst;
2881
2882         xdst = xfrm_alloc_dst(net, family);
2883         if (IS_ERR(xdst))
2884                 return xdst;
2885
2886         if (!(xflo->flags & XFRM_LOOKUP_QUEUE) ||
2887             net->xfrm.sysctl_larval_drop ||
2888             num_xfrms <= 0)
2889                 return xdst;
2890
2891         dst = xflo->dst_orig;
2892         dst1 = &xdst->u.dst;
2893         dst_hold(dst);
2894         xdst->route = dst;
2895
2896         dst_copy_metrics(dst1, dst);
2897
2898         dst1->obsolete = DST_OBSOLETE_FORCE_CHK;
2899         dst1->flags |= DST_HOST | DST_XFRM_QUEUE;
2900         dst1->lastuse = jiffies;
2901
2902         dst1->input = dst_discard;
2903         dst1->output = xdst_queue_output;
2904
2905         dst_hold(dst);
2906         xfrm_dst_set_child(xdst, dst);
2907         xdst->path = dst;
2908
2909         xfrm_init_path((struct xfrm_dst *)dst1, dst, 0);
2910
2911         err = -ENODEV;
2912         dev = dst->dev;
2913         if (!dev)
2914                 goto free_dst;
2915
2916         err = xfrm_fill_dst(xdst, dev, fl);
2917         if (err)
2918                 goto free_dst;
2919
2920 out:
2921         return xdst;
2922
2923 free_dst:
2924         dst_release(dst1);
2925         xdst = ERR_PTR(err);
2926         goto out;
2927 }
2928
2929 static struct xfrm_dst *xfrm_bundle_lookup(struct net *net,
2930                                            const struct flowi *fl,
2931                                            u16 family, u8 dir,
2932                                            struct xfrm_flo *xflo, u32 if_id)
2933 {
2934         struct xfrm_policy *pols[XFRM_POLICY_TYPE_MAX];
2935         int num_pols = 0, num_xfrms = 0, err;
2936         struct xfrm_dst *xdst;
2937
2938         /* Resolve policies to use if we couldn't get them from
2939          * previous cache entry */
2940         num_pols = 1;
2941         pols[0] = xfrm_policy_lookup(net, fl, family, dir, if_id);
2942         err = xfrm_expand_policies(fl, family, pols,
2943                                            &num_pols, &num_xfrms);
2944         if (err < 0)
2945                 goto inc_error;
2946         if (num_pols == 0)
2947                 return NULL;
2948         if (num_xfrms <= 0)
2949                 goto make_dummy_bundle;
2950
2951         xdst = xfrm_resolve_and_create_bundle(pols, num_pols, fl, family,
2952                                               xflo->dst_orig);
2953         if (IS_ERR(xdst)) {
2954                 err = PTR_ERR(xdst);
2955                 if (err == -EREMOTE) {
2956                         xfrm_pols_put(pols, num_pols);
2957                         return NULL;
2958                 }
2959
2960                 if (err != -EAGAIN)
2961                         goto error;
2962                 goto make_dummy_bundle;
2963         } else if (xdst == NULL) {
2964                 num_xfrms = 0;
2965                 goto make_dummy_bundle;
2966         }
2967
2968         return xdst;
2969
2970 make_dummy_bundle:
2971         /* We found policies, but there's no bundles to instantiate:
2972          * either because the policy blocks, has no transformations or
2973          * we could not build template (no xfrm_states).*/
2974         xdst = xfrm_create_dummy_bundle(net, xflo, fl, num_xfrms, family);
2975         if (IS_ERR(xdst)) {
2976                 xfrm_pols_put(pols, num_pols);
2977                 return ERR_CAST(xdst);
2978         }
2979         xdst->num_pols = num_pols;
2980         xdst->num_xfrms = num_xfrms;
2981         memcpy(xdst->pols, pols, sizeof(struct xfrm_policy *) * num_pols);
2982
2983         return xdst;
2984
2985 inc_error:
2986         XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTPOLERROR);
2987 error:
2988         xfrm_pols_put(pols, num_pols);
2989         return ERR_PTR(err);
2990 }
2991
2992 static struct dst_entry *make_blackhole(struct net *net, u16 family,
2993                                         struct dst_entry *dst_orig)
2994 {
2995         const struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family);
2996         struct dst_entry *ret;
2997
2998         if (!afinfo) {
2999                 dst_release(dst_orig);
3000                 return ERR_PTR(-EINVAL);
3001         } else {
3002                 ret = afinfo->blackhole_route(net, dst_orig);
3003         }
3004         rcu_read_unlock();
3005
3006         return ret;
3007 }
3008
3009 /* Finds/creates a bundle for given flow and if_id
3010  *
3011  * At the moment we eat a raw IP route. Mostly to speed up lookups
3012  * on interfaces with disabled IPsec.
3013  *
3014  * xfrm_lookup uses an if_id of 0 by default, and is provided for
3015  * compatibility
3016  */
3017 struct dst_entry *xfrm_lookup_with_ifid(struct net *net,
3018                                         struct dst_entry *dst_orig,
3019                                         const struct flowi *fl,
3020                                         const struct sock *sk,
3021                                         int flags, u32 if_id)
3022 {
3023         struct xfrm_policy *pols[XFRM_POLICY_TYPE_MAX];
3024         struct xfrm_dst *xdst;
3025         struct dst_entry *dst, *route;
3026         u16 family = dst_orig->ops->family;
3027         u8 dir = XFRM_POLICY_OUT;
3028         int i, err, num_pols, num_xfrms = 0, drop_pols = 0;
3029
3030         dst = NULL;
3031         xdst = NULL;
3032         route = NULL;
3033
3034         sk = sk_const_to_full_sk(sk);
3035         if (sk && sk->sk_policy[XFRM_POLICY_OUT]) {
3036                 num_pols = 1;
3037                 pols[0] = xfrm_sk_policy_lookup(sk, XFRM_POLICY_OUT, fl, family,
3038                                                 if_id);
3039                 err = xfrm_expand_policies(fl, family, pols,
3040                                            &num_pols, &num_xfrms);
3041                 if (err < 0)
3042                         goto dropdst;
3043
3044                 if (num_pols) {
3045                         if (num_xfrms <= 0) {
3046                                 drop_pols = num_pols;
3047                                 goto no_transform;
3048                         }
3049
3050                         xdst = xfrm_resolve_and_create_bundle(
3051                                         pols, num_pols, fl,
3052                                         family, dst_orig);
3053
3054                         if (IS_ERR(xdst)) {
3055                                 xfrm_pols_put(pols, num_pols);
3056                                 err = PTR_ERR(xdst);
3057                                 if (err == -EREMOTE)
3058                                         goto nopol;
3059
3060                                 goto dropdst;
3061                         } else if (xdst == NULL) {
3062                                 num_xfrms = 0;
3063                                 drop_pols = num_pols;
3064                                 goto no_transform;
3065                         }
3066
3067                         route = xdst->route;
3068                 }
3069         }
3070
3071         if (xdst == NULL) {
3072                 struct xfrm_flo xflo;
3073
3074                 xflo.dst_orig = dst_orig;
3075                 xflo.flags = flags;
3076
3077                 /* To accelerate a bit...  */
3078                 if ((dst_orig->flags & DST_NOXFRM) ||
3079                     !net->xfrm.policy_count[XFRM_POLICY_OUT])
3080                         goto nopol;
3081
3082                 xdst = xfrm_bundle_lookup(net, fl, family, dir, &xflo, if_id);
3083                 if (xdst == NULL)
3084                         goto nopol;
3085                 if (IS_ERR(xdst)) {
3086                         err = PTR_ERR(xdst);
3087                         goto dropdst;
3088                 }
3089
3090                 num_pols = xdst->num_pols;
3091                 num_xfrms = xdst->num_xfrms;
3092                 memcpy(pols, xdst->pols, sizeof(struct xfrm_policy *) * num_pols);
3093                 route = xdst->route;
3094         }
3095
3096         dst = &xdst->u.dst;
3097         if (route == NULL && num_xfrms > 0) {
3098                 /* The only case when xfrm_bundle_lookup() returns a
3099                  * bundle with null route, is when the template could
3100                  * not be resolved. It means policies are there, but
3101                  * bundle could not be created, since we don't yet
3102                  * have the xfrm_state's. We need to wait for KM to
3103                  * negotiate new SA's or bail out with error.*/
3104                 if (net->xfrm.sysctl_larval_drop) {
3105                         XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTNOSTATES);
3106                         err = -EREMOTE;
3107                         goto error;
3108                 }
3109
3110                 err = -EAGAIN;
3111
3112                 XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTNOSTATES);
3113                 goto error;
3114         }
3115
3116 no_transform:
3117         if (num_pols == 0)
3118                 goto nopol;
3119
3120         if ((flags & XFRM_LOOKUP_ICMP) &&
3121             !(pols[0]->flags & XFRM_POLICY_ICMP)) {
3122                 err = -ENOENT;
3123                 goto error;
3124         }
3125
3126         for (i = 0; i < num_pols; i++)
3127                 pols[i]->curlft.use_time = ktime_get_real_seconds();
3128
3129         if (num_xfrms < 0) {
3130                 /* Prohibit the flow */
3131                 XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTPOLBLOCK);
3132                 err = -EPERM;
3133                 goto error;
3134         } else if (num_xfrms > 0) {
3135                 /* Flow transformed */
3136                 dst_release(dst_orig);
3137         } else {
3138                 /* Flow passes untransformed */
3139                 dst_release(dst);
3140                 dst = dst_orig;
3141         }
3142 ok:
3143         xfrm_pols_put(pols, drop_pols);
3144         if (dst && dst->xfrm &&
3145             dst->xfrm->props.mode == XFRM_MODE_TUNNEL)
3146                 dst->flags |= DST_XFRM_TUNNEL;
3147         return dst;
3148
3149 nopol:
3150         if (!(flags & XFRM_LOOKUP_ICMP)) {
3151                 dst = dst_orig;
3152                 goto ok;
3153         }
3154         err = -ENOENT;
3155 error:
3156         dst_release(dst);
3157 dropdst:
3158         if (!(flags & XFRM_LOOKUP_KEEP_DST_REF))
3159                 dst_release(dst_orig);
3160         xfrm_pols_put(pols, drop_pols);
3161         return ERR_PTR(err);
3162 }
3163 EXPORT_SYMBOL(xfrm_lookup_with_ifid);
3164
3165 /* Main function: finds/creates a bundle for given flow.
3166  *
3167  * At the moment we eat a raw IP route. Mostly to speed up lookups
3168  * on interfaces with disabled IPsec.
3169  */
3170 struct dst_entry *xfrm_lookup(struct net *net, struct dst_entry *dst_orig,
3171                               const struct flowi *fl, const struct sock *sk,
3172                               int flags)
3173 {
3174         return xfrm_lookup_with_ifid(net, dst_orig, fl, sk, flags, 0);
3175 }
3176 EXPORT_SYMBOL(xfrm_lookup);
3177
3178 /* Callers of xfrm_lookup_route() must ensure a call to dst_output().
3179  * Otherwise we may send out blackholed packets.
3180  */
3181 struct dst_entry *xfrm_lookup_route(struct net *net, struct dst_entry *dst_orig,
3182                                     const struct flowi *fl,
3183                                     const struct sock *sk, int flags)
3184 {
3185         struct dst_entry *dst = xfrm_lookup(net, dst_orig, fl, sk,
3186                                             flags | XFRM_LOOKUP_QUEUE |
3187                                             XFRM_LOOKUP_KEEP_DST_REF);
3188
3189         if (IS_ERR(dst) && PTR_ERR(dst) == -EREMOTE)
3190                 return make_blackhole(net, dst_orig->ops->family, dst_orig);
3191
3192         if (IS_ERR(dst))
3193                 dst_release(dst_orig);
3194
3195         return dst;
3196 }
3197 EXPORT_SYMBOL(xfrm_lookup_route);
3198
3199 static inline int
3200 xfrm_secpath_reject(int idx, struct sk_buff *skb, const struct flowi *fl)
3201 {
3202         struct sec_path *sp = skb_sec_path(skb);
3203         struct xfrm_state *x;
3204
3205         if (!sp || idx < 0 || idx >= sp->len)
3206                 return 0;
3207         x = sp->xvec[idx];
3208         if (!x->type->reject)
3209                 return 0;
3210         return x->type->reject(x, skb, fl);
3211 }
3212
3213 /* When skb is transformed back to its "native" form, we have to
3214  * check policy restrictions. At the moment we make this in maximally
3215  * stupid way. Shame on me. :-) Of course, connected sockets must
3216  * have policy cached at them.
3217  */
3218
3219 static inline int
3220 xfrm_state_ok(const struct xfrm_tmpl *tmpl, const struct xfrm_state *x,
3221               unsigned short family)
3222 {
3223         if (xfrm_state_kern(x))
3224                 return tmpl->optional && !xfrm_state_addr_cmp(tmpl, x, tmpl->encap_family);
3225         return  x->id.proto == tmpl->id.proto &&
3226                 (x->id.spi == tmpl->id.spi || !tmpl->id.spi) &&
3227                 (x->props.reqid == tmpl->reqid || !tmpl->reqid) &&
3228                 x->props.mode == tmpl->mode &&
3229                 (tmpl->allalgs || (tmpl->aalgos & (1<<x->props.aalgo)) ||
3230                  !(xfrm_id_proto_match(tmpl->id.proto, IPSEC_PROTO_ANY))) &&
3231                 !(x->props.mode != XFRM_MODE_TRANSPORT &&
3232                   xfrm_state_addr_cmp(tmpl, x, family));
3233 }
3234
3235 /*
3236  * 0 or more than 0 is returned when validation is succeeded (either bypass
3237  * because of optional transport mode, or next index of the mathced secpath
3238  * state with the template.
3239  * -1 is returned when no matching template is found.
3240  * Otherwise "-2 - errored_index" is returned.
3241  */
3242 static inline int
3243 xfrm_policy_ok(const struct xfrm_tmpl *tmpl, const struct sec_path *sp, int start,
3244                unsigned short family)
3245 {
3246         int idx = start;
3247
3248         if (tmpl->optional) {
3249                 if (tmpl->mode == XFRM_MODE_TRANSPORT)
3250                         return start;
3251         } else
3252                 start = -1;
3253         for (; idx < sp->len; idx++) {
3254                 if (xfrm_state_ok(tmpl, sp->xvec[idx], family))
3255                         return ++idx;
3256                 if (sp->xvec[idx]->props.mode != XFRM_MODE_TRANSPORT) {
3257                         if (start == -1)
3258                                 start = -2-idx;
3259                         break;
3260                 }
3261         }
3262         return start;
3263 }
3264
3265 static void
3266 decode_session4(struct sk_buff *skb, struct flowi *fl, bool reverse)
3267 {
3268         const struct iphdr *iph = ip_hdr(skb);
3269         int ihl = iph->ihl;
3270         u8 *xprth = skb_network_header(skb) + ihl * 4;
3271         struct flowi4 *fl4 = &fl->u.ip4;
3272         int oif = 0;
3273
3274         if (skb_dst(skb) && skb_dst(skb)->dev)
3275                 oif = skb_dst(skb)->dev->ifindex;
3276
3277         memset(fl4, 0, sizeof(struct flowi4));
3278         fl4->flowi4_mark = skb->mark;
3279         fl4->flowi4_oif = reverse ? skb->skb_iif : oif;
3280
3281         fl4->flowi4_proto = iph->protocol;
3282         fl4->daddr = reverse ? iph->saddr : iph->daddr;
3283         fl4->saddr = reverse ? iph->daddr : iph->saddr;
3284         fl4->flowi4_tos = iph->tos;
3285
3286         if (!ip_is_fragment(iph)) {
3287                 switch (iph->protocol) {
3288                 case IPPROTO_UDP:
3289                 case IPPROTO_UDPLITE:
3290                 case IPPROTO_TCP:
3291                 case IPPROTO_SCTP:
3292                 case IPPROTO_DCCP:
3293                         if (xprth + 4 < skb->data ||
3294                             pskb_may_pull(skb, xprth + 4 - skb->data)) {
3295                                 __be16 *ports;
3296
3297                                 xprth = skb_network_header(skb) + ihl * 4;
3298                                 ports = (__be16 *)xprth;
3299
3300                                 fl4->fl4_sport = ports[!!reverse];
3301                                 fl4->fl4_dport = ports[!reverse];
3302                         }
3303                         break;
3304                 case IPPROTO_ICMP:
3305                         if (xprth + 2 < skb->data ||
3306                             pskb_may_pull(skb, xprth + 2 - skb->data)) {
3307                                 u8 *icmp;
3308
3309                                 xprth = skb_network_header(skb) + ihl * 4;
3310                                 icmp = xprth;
3311
3312                                 fl4->fl4_icmp_type = icmp[0];
3313                                 fl4->fl4_icmp_code = icmp[1];
3314                         }
3315                         break;
3316                 case IPPROTO_ESP:
3317                         if (xprth + 4 < skb->data ||
3318                             pskb_may_pull(skb, xprth + 4 - skb->data)) {
3319                                 __be32 *ehdr;
3320
3321                                 xprth = skb_network_header(skb) + ihl * 4;
3322                                 ehdr = (__be32 *)xprth;
3323
3324                                 fl4->fl4_ipsec_spi = ehdr[0];
3325                         }
3326                         break;
3327                 case IPPROTO_AH:
3328                         if (xprth + 8 < skb->data ||
3329                             pskb_may_pull(skb, xprth + 8 - skb->data)) {
3330                                 __be32 *ah_hdr;
3331
3332                                 xprth = skb_network_header(skb) + ihl * 4;
3333                                 ah_hdr = (__be32 *)xprth;
3334
3335                                 fl4->fl4_ipsec_spi = ah_hdr[1];
3336                         }
3337                         break;
3338                 case IPPROTO_COMP:
3339                         if (xprth + 4 < skb->data ||
3340                             pskb_may_pull(skb, xprth + 4 - skb->data)) {
3341                                 __be16 *ipcomp_hdr;
3342
3343                                 xprth = skb_network_header(skb) + ihl * 4;
3344                                 ipcomp_hdr = (__be16 *)xprth;
3345
3346                                 fl4->fl4_ipsec_spi = htonl(ntohs(ipcomp_hdr[1]));
3347                         }
3348                         break;
3349                 case IPPROTO_GRE:
3350                         if (xprth + 12 < skb->data ||
3351                             pskb_may_pull(skb, xprth + 12 - skb->data)) {
3352                                 __be16 *greflags;
3353                                 __be32 *gre_hdr;
3354
3355                                 xprth = skb_network_header(skb) + ihl * 4;
3356                                 greflags = (__be16 *)xprth;
3357                                 gre_hdr = (__be32 *)xprth;
3358
3359                                 if (greflags[0] & GRE_KEY) {
3360                                         if (greflags[0] & GRE_CSUM)
3361                                                 gre_hdr++;
3362                                         fl4->fl4_gre_key = gre_hdr[1];
3363                                 }
3364                         }
3365                         break;
3366                 default:
3367                         fl4->fl4_ipsec_spi = 0;
3368                         break;
3369                 }
3370         }
3371 }
3372
3373 #if IS_ENABLED(CONFIG_IPV6)
3374 static void
3375 decode_session6(struct sk_buff *skb, struct flowi *fl, bool reverse)
3376 {
3377         struct flowi6 *fl6 = &fl->u.ip6;
3378         int onlyproto = 0;
3379         const struct ipv6hdr *hdr = ipv6_hdr(skb);
3380         u32 offset = sizeof(*hdr);
3381         struct ipv6_opt_hdr *exthdr;
3382         const unsigned char *nh = skb_network_header(skb);
3383         u16 nhoff = IP6CB(skb)->nhoff;
3384         int oif = 0;
3385         u8 nexthdr;
3386
3387         if (!nhoff)
3388                 nhoff = offsetof(struct ipv6hdr, nexthdr);
3389
3390         nexthdr = nh[nhoff];
3391
3392         if (skb_dst(skb) && skb_dst(skb)->dev)
3393                 oif = skb_dst(skb)->dev->ifindex;
3394
3395         memset(fl6, 0, sizeof(struct flowi6));
3396         fl6->flowi6_mark = skb->mark;
3397         fl6->flowi6_oif = reverse ? skb->skb_iif : oif;
3398
3399         fl6->daddr = reverse ? hdr->saddr : hdr->daddr;
3400         fl6->saddr = reverse ? hdr->daddr : hdr->saddr;
3401
3402         while (nh + offset + sizeof(*exthdr) < skb->data ||
3403                pskb_may_pull(skb, nh + offset + sizeof(*exthdr) - skb->data)) {
3404                 nh = skb_network_header(skb);
3405                 exthdr = (struct ipv6_opt_hdr *)(nh + offset);
3406
3407                 switch (nexthdr) {
3408                 case NEXTHDR_FRAGMENT:
3409                         onlyproto = 1;
3410                         /* fall through */
3411                 case NEXTHDR_ROUTING:
3412                 case NEXTHDR_HOP:
3413                 case NEXTHDR_DEST:
3414                         offset += ipv6_optlen(exthdr);
3415                         nexthdr = exthdr->nexthdr;
3416                         exthdr = (struct ipv6_opt_hdr *)(nh + offset);
3417                         break;
3418                 case IPPROTO_UDP:
3419                 case IPPROTO_UDPLITE:
3420                 case IPPROTO_TCP:
3421                 case IPPROTO_SCTP:
3422                 case IPPROTO_DCCP:
3423                         if (!onlyproto && (nh + offset + 4 < skb->data ||
3424                              pskb_may_pull(skb, nh + offset + 4 - skb->data))) {
3425                                 __be16 *ports;
3426
3427                                 nh = skb_network_header(skb);
3428                                 ports = (__be16 *)(nh + offset);
3429                                 fl6->fl6_sport = ports[!!reverse];
3430                                 fl6->fl6_dport = ports[!reverse];
3431                         }
3432                         fl6->flowi6_proto = nexthdr;
3433                         return;
3434                 case IPPROTO_ICMPV6:
3435                         if (!onlyproto && (nh + offset + 2 < skb->data ||
3436                             pskb_may_pull(skb, nh + offset + 2 - skb->data))) {
3437                                 u8 *icmp;
3438
3439                                 nh = skb_network_header(skb);
3440                                 icmp = (u8 *)(nh + offset);
3441                                 fl6->fl6_icmp_type = icmp[0];
3442                                 fl6->fl6_icmp_code = icmp[1];
3443                         }
3444                         fl6->flowi6_proto = nexthdr;
3445                         return;
3446 #if IS_ENABLED(CONFIG_IPV6_MIP6)
3447                 case IPPROTO_MH:
3448                         offset += ipv6_optlen(exthdr);
3449                         if (!onlyproto && (nh + offset + 3 < skb->data ||
3450                             pskb_may_pull(skb, nh + offset + 3 - skb->data))) {
3451                                 struct ip6_mh *mh;
3452
3453                                 nh = skb_network_header(skb);
3454                                 mh = (struct ip6_mh *)(nh + offset);
3455                                 fl6->fl6_mh_type = mh->ip6mh_type;
3456                         }
3457                         fl6->flowi6_proto = nexthdr;
3458                         return;
3459 #endif
3460                 /* XXX Why are there these headers? */
3461                 case IPPROTO_AH:
3462                 case IPPROTO_ESP:
3463                 case IPPROTO_COMP:
3464                 default:
3465                         fl6->fl6_ipsec_spi = 0;
3466                         fl6->flowi6_proto = nexthdr;
3467                         return;
3468                 }
3469         }
3470 }
3471 #endif
3472
3473 int __xfrm_decode_session(struct sk_buff *skb, struct flowi *fl,
3474                           unsigned int family, int reverse)
3475 {
3476         switch (family) {
3477         case AF_INET:
3478                 decode_session4(skb, fl, reverse);
3479                 break;
3480 #if IS_ENABLED(CONFIG_IPV6)
3481         case AF_INET6:
3482                 decode_session6(skb, fl, reverse);
3483                 break;
3484 #endif
3485         default:
3486                 return -EAFNOSUPPORT;
3487         }
3488
3489         return security_xfrm_decode_session(skb, &fl->flowi_secid);
3490 }
3491 EXPORT_SYMBOL(__xfrm_decode_session);
3492
3493 static inline int secpath_has_nontransport(const struct sec_path *sp, int k, int *idxp)
3494 {
3495         for (; k < sp->len; k++) {
3496                 if (sp->xvec[k]->props.mode != XFRM_MODE_TRANSPORT) {
3497                         *idxp = k;
3498                         return 1;
3499                 }
3500         }
3501
3502         return 0;
3503 }
3504
3505 int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
3506                         unsigned short family)
3507 {
3508         struct net *net = dev_net(skb->dev);
3509         struct xfrm_policy *pol;
3510         struct xfrm_policy *pols[XFRM_POLICY_TYPE_MAX];
3511         int npols = 0;
3512         int xfrm_nr;
3513         int pi;
3514         int reverse;
3515         struct flowi fl;
3516         int xerr_idx = -1;
3517         const struct xfrm_if_cb *ifcb;
3518         struct sec_path *sp;
3519         struct xfrm_if *xi;
3520         u32 if_id = 0;
3521
3522         rcu_read_lock();
3523         ifcb = xfrm_if_get_cb();
3524
3525         if (ifcb) {
3526                 xi = ifcb->decode_session(skb, family);
3527                 if (xi) {
3528                         if_id = xi->p.if_id;
3529                         net = xi->net;
3530                 }
3531         }
3532         rcu_read_unlock();
3533
3534         reverse = dir & ~XFRM_POLICY_MASK;
3535         dir &= XFRM_POLICY_MASK;
3536
3537         if (__xfrm_decode_session(skb, &fl, family, reverse) < 0) {
3538                 XFRM_INC_STATS(net, LINUX_MIB_XFRMINHDRERROR);
3539                 return 0;
3540         }
3541
3542         nf_nat_decode_session(skb, &fl, family);
3543
3544         /* First, check used SA against their selectors. */
3545         sp = skb_sec_path(skb);
3546         if (sp) {
3547                 int i;
3548
3549                 for (i = sp->len - 1; i >= 0; i--) {
3550                         struct xfrm_state *x = sp->xvec[i];
3551                         if (!xfrm_selector_match(&x->sel, &fl, family)) {
3552                                 XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEMISMATCH);
3553                                 return 0;
3554                         }
3555                 }
3556         }
3557
3558         pol = NULL;
3559         sk = sk_to_full_sk(sk);
3560         if (sk && sk->sk_policy[dir]) {
3561                 pol = xfrm_sk_policy_lookup(sk, dir, &fl, family, if_id);
3562                 if (IS_ERR(pol)) {
3563                         XFRM_INC_STATS(net, LINUX_MIB_XFRMINPOLERROR);
3564                         return 0;
3565                 }
3566         }
3567
3568         if (!pol)
3569                 pol = xfrm_policy_lookup(net, &fl, family, dir, if_id);
3570
3571         if (IS_ERR(pol)) {
3572                 XFRM_INC_STATS(net, LINUX_MIB_XFRMINPOLERROR);
3573                 return 0;
3574         }
3575
3576         if (!pol) {
3577                 if (sp && secpath_has_nontransport(sp, 0, &xerr_idx)) {
3578                         xfrm_secpath_reject(xerr_idx, skb, &fl);
3579                         XFRM_INC_STATS(net, LINUX_MIB_XFRMINNOPOLS);
3580                         return 0;
3581                 }
3582                 return 1;
3583         }
3584
3585         pol->curlft.use_time = ktime_get_real_seconds();
3586
3587         pols[0] = pol;
3588         npols++;
3589 #ifdef CONFIG_XFRM_SUB_POLICY
3590         if (pols[0]->type != XFRM_POLICY_TYPE_MAIN) {
3591                 pols[1] = xfrm_policy_lookup_bytype(net, XFRM_POLICY_TYPE_MAIN,
3592                                                     &fl, family,
3593                                                     XFRM_POLICY_IN, if_id);
3594                 if (pols[1]) {
3595                         if (IS_ERR(pols[1])) {
3596                                 XFRM_INC_STATS(net, LINUX_MIB_XFRMINPOLERROR);
3597                                 return 0;
3598                         }
3599                         pols[1]->curlft.use_time = ktime_get_real_seconds();
3600                         npols++;
3601                 }
3602         }
3603 #endif
3604
3605         if (pol->action == XFRM_POLICY_ALLOW) {
3606                 static struct sec_path dummy;
3607                 struct xfrm_tmpl *tp[XFRM_MAX_DEPTH];
3608                 struct xfrm_tmpl *stp[XFRM_MAX_DEPTH];
3609                 struct xfrm_tmpl **tpp = tp;
3610                 int ti = 0;
3611                 int i, k;
3612
3613                 sp = skb_sec_path(skb);
3614                 if (!sp)
3615                         sp = &dummy;
3616
3617                 for (pi = 0; pi < npols; pi++) {
3618                         if (pols[pi] != pol &&
3619                             pols[pi]->action != XFRM_POLICY_ALLOW) {
3620                                 XFRM_INC_STATS(net, LINUX_MIB_XFRMINPOLBLOCK);
3621                                 goto reject;
3622                         }
3623                         if (ti + pols[pi]->xfrm_nr >= XFRM_MAX_DEPTH) {
3624                                 XFRM_INC_STATS(net, LINUX_MIB_XFRMINBUFFERERROR);
3625                                 goto reject_error;
3626                         }
3627                         for (i = 0; i < pols[pi]->xfrm_nr; i++)
3628                                 tpp[ti++] = &pols[pi]->xfrm_vec[i];
3629                 }
3630                 xfrm_nr = ti;
3631                 if (npols > 1) {
3632                         xfrm_tmpl_sort(stp, tpp, xfrm_nr, family);
3633                         tpp = stp;
3634                 }
3635
3636                 /* For each tunnel xfrm, find the first matching tmpl.
3637                  * For each tmpl before that, find corresponding xfrm.
3638                  * Order is _important_. Later we will implement
3639                  * some barriers, but at the moment barriers
3640                  * are implied between each two transformations.
3641                  */
3642                 for (i = xfrm_nr-1, k = 0; i >= 0; i--) {
3643                         k = xfrm_policy_ok(tpp[i], sp, k, family);
3644                         if (k < 0) {
3645                                 if (k < -1)
3646                                         /* "-2 - errored_index" returned */
3647                                         xerr_idx = -(2+k);
3648                                 XFRM_INC_STATS(net, LINUX_MIB_XFRMINTMPLMISMATCH);
3649                                 goto reject;
3650                         }
3651                 }
3652
3653                 if (secpath_has_nontransport(sp, k, &xerr_idx)) {
3654                         XFRM_INC_STATS(net, LINUX_MIB_XFRMINTMPLMISMATCH);
3655                         goto reject;
3656                 }
3657
3658                 xfrm_pols_put(pols, npols);
3659                 return 1;
3660         }
3661         XFRM_INC_STATS(net, LINUX_MIB_XFRMINPOLBLOCK);
3662
3663 reject:
3664         xfrm_secpath_reject(xerr_idx, skb, &fl);
3665 reject_error:
3666         xfrm_pols_put(pols, npols);
3667         return 0;
3668 }
3669 EXPORT_SYMBOL(__xfrm_policy_check);
3670
3671 int __xfrm_route_forward(struct sk_buff *skb, unsigned short family)
3672 {
3673         struct net *net = dev_net(skb->dev);
3674         struct flowi fl;
3675         struct dst_entry *dst;
3676         int res = 1;
3677
3678         if (xfrm_decode_session(skb, &fl, family) < 0) {
3679                 XFRM_INC_STATS(net, LINUX_MIB_XFRMFWDHDRERROR);
3680                 return 0;
3681         }
3682
3683         skb_dst_force(skb);
3684         if (!skb_dst(skb)) {
3685                 XFRM_INC_STATS(net, LINUX_MIB_XFRMFWDHDRERROR);
3686                 return 0;
3687         }
3688
3689         dst = xfrm_lookup(net, skb_dst(skb), &fl, NULL, XFRM_LOOKUP_QUEUE);
3690         if (IS_ERR(dst)) {
3691                 res = 0;
3692                 dst = NULL;
3693         }
3694         skb_dst_set(skb, dst);
3695         return res;
3696 }
3697 EXPORT_SYMBOL(__xfrm_route_forward);
3698
3699 /* Optimize later using cookies and generation ids. */
3700
3701 static struct dst_entry *xfrm_dst_check(struct dst_entry *dst, u32 cookie)
3702 {
3703         /* Code (such as __xfrm4_bundle_create()) sets dst->obsolete
3704          * to DST_OBSOLETE_FORCE_CHK to force all XFRM destinations to
3705          * get validated by dst_ops->check on every use.  We do this
3706          * because when a normal route referenced by an XFRM dst is
3707          * obsoleted we do not go looking around for all parent
3708          * referencing XFRM dsts so that we can invalidate them.  It
3709          * is just too much work.  Instead we make the checks here on
3710          * every use.  For example:
3711          *
3712          *      XFRM dst A --> IPv4 dst X
3713          *
3714          * X is the "xdst->route" of A (X is also the "dst->path" of A
3715          * in this example).  If X is marked obsolete, "A" will not
3716          * notice.  That's what we are validating here via the
3717          * stale_bundle() check.
3718          *
3719          * When a dst is removed from the fib tree, DST_OBSOLETE_DEAD will
3720          * be marked on it.
3721          * This will force stale_bundle() to fail on any xdst bundle with
3722          * this dst linked in it.
3723          */
3724         if (dst->obsolete < 0 && !stale_bundle(dst))
3725                 return dst;
3726
3727         return NULL;
3728 }
3729
3730 static int stale_bundle(struct dst_entry *dst)
3731 {
3732         return !xfrm_bundle_ok((struct xfrm_dst *)dst);
3733 }
3734
3735 void xfrm_dst_ifdown(struct dst_entry *dst, struct net_device *dev)
3736 {
3737         while ((dst = xfrm_dst_child(dst)) && dst->xfrm && dst->dev == dev) {
3738                 dst->dev = dev_net(dev)->loopback_dev;
3739                 dev_hold(dst->dev);
3740                 dev_put(dev);
3741         }
3742 }
3743 EXPORT_SYMBOL(xfrm_dst_ifdown);
3744
3745 static void xfrm_link_failure(struct sk_buff *skb)
3746 {
3747         /* Impossible. Such dst must be popped before reaches point of failure. */
3748 }
3749
3750 static struct dst_entry *xfrm_negative_advice(struct dst_entry *dst)
3751 {
3752         if (dst) {
3753                 if (dst->obsolete) {
3754                         dst_release(dst);
3755                         dst = NULL;
3756                 }
3757         }
3758         return dst;
3759 }
3760
3761 static void xfrm_init_pmtu(struct xfrm_dst **bundle, int nr)
3762 {
3763         while (nr--) {
3764                 struct xfrm_dst *xdst = bundle[nr];
3765                 u32 pmtu, route_mtu_cached;
3766                 struct dst_entry *dst;
3767
3768                 dst = &xdst->u.dst;
3769                 pmtu = dst_mtu(xfrm_dst_child(dst));
3770                 xdst->child_mtu_cached = pmtu;
3771
3772                 pmtu = xfrm_state_mtu(dst->xfrm, pmtu);
3773
3774                 route_mtu_cached = dst_mtu(xdst->route);
3775                 xdst->route_mtu_cached = route_mtu_cached;
3776
3777                 if (pmtu > route_mtu_cached)
3778                         pmtu = route_mtu_cached;
3779
3780                 dst_metric_set(dst, RTAX_MTU, pmtu);
3781         }
3782 }
3783
3784 /* Check that the bundle accepts the flow and its components are
3785  * still valid.
3786  */
3787
3788 static int xfrm_bundle_ok(struct xfrm_dst *first)
3789 {
3790         struct xfrm_dst *bundle[XFRM_MAX_DEPTH];
3791         struct dst_entry *dst = &first->u.dst;
3792         struct xfrm_dst *xdst;
3793         int start_from, nr;
3794         u32 mtu;
3795
3796         if (!dst_check(xfrm_dst_path(dst), ((struct xfrm_dst *)dst)->path_cookie) ||
3797             (dst->dev && !netif_running(dst->dev)))
3798                 return 0;
3799
3800         if (dst->flags & DST_XFRM_QUEUE)
3801                 return 1;
3802
3803         start_from = nr = 0;
3804         do {
3805                 struct xfrm_dst *xdst = (struct xfrm_dst *)dst;
3806
3807                 if (dst->xfrm->km.state != XFRM_STATE_VALID)
3808                         return 0;
3809                 if (xdst->xfrm_genid != dst->xfrm->genid)
3810                         return 0;
3811                 if (xdst->num_pols > 0 &&
3812                     xdst->policy_genid != atomic_read(&xdst->pols[0]->genid))
3813                         return 0;
3814
3815                 bundle[nr++] = xdst;
3816
3817                 mtu = dst_mtu(xfrm_dst_child(dst));
3818                 if (xdst->child_mtu_cached != mtu) {
3819                         start_from = nr;
3820                         xdst->child_mtu_cached = mtu;
3821                 }
3822
3823                 if (!dst_check(xdst->route, xdst->route_cookie))
3824                         return 0;
3825                 mtu = dst_mtu(xdst->route);
3826                 if (xdst->route_mtu_cached != mtu) {
3827                         start_from = nr;
3828                         xdst->route_mtu_cached = mtu;
3829                 }
3830
3831                 dst = xfrm_dst_child(dst);
3832         } while (dst->xfrm);
3833
3834         if (likely(!start_from))
3835                 return 1;
3836
3837         xdst = bundle[start_from - 1];
3838         mtu = xdst->child_mtu_cached;
3839         while (start_from--) {
3840                 dst = &xdst->u.dst;
3841
3842                 mtu = xfrm_state_mtu(dst->xfrm, mtu);
3843                 if (mtu > xdst->route_mtu_cached)
3844                         mtu = xdst->route_mtu_cached;
3845                 dst_metric_set(dst, RTAX_MTU, mtu);
3846                 if (!start_from)
3847                         break;
3848
3849                 xdst = bundle[start_from - 1];
3850                 xdst->child_mtu_cached = mtu;
3851         }
3852
3853         return 1;
3854 }
3855
3856 static unsigned int xfrm_default_advmss(const struct dst_entry *dst)
3857 {
3858         return dst_metric_advmss(xfrm_dst_path(dst));
3859 }
3860
3861 static unsigned int xfrm_mtu(const struct dst_entry *dst)
3862 {
3863         unsigned int mtu = dst_metric_raw(dst, RTAX_MTU);
3864
3865         return mtu ? : dst_mtu(xfrm_dst_path(dst));
3866 }
3867
3868 static const void *xfrm_get_dst_nexthop(const struct dst_entry *dst,
3869                                         const void *daddr)
3870 {
3871         while (dst->xfrm) {
3872                 const struct xfrm_state *xfrm = dst->xfrm;
3873
3874                 dst = xfrm_dst_child(dst);
3875
3876                 if (xfrm->props.mode == XFRM_MODE_TRANSPORT)
3877                         continue;
3878                 if (xfrm->type->flags & XFRM_TYPE_REMOTE_COADDR)
3879                         daddr = xfrm->coaddr;
3880                 else if (!(xfrm->type->flags & XFRM_TYPE_LOCAL_COADDR))
3881                         daddr = &xfrm->id.daddr;
3882         }
3883         return daddr;
3884 }
3885
3886 static struct neighbour *xfrm_neigh_lookup(const struct dst_entry *dst,
3887                                            struct sk_buff *skb,
3888                                            const void *daddr)
3889 {
3890         const struct dst_entry *path = xfrm_dst_path(dst);
3891
3892         if (!skb)
3893                 daddr = xfrm_get_dst_nexthop(dst, daddr);
3894         return path->ops->neigh_lookup(path, skb, daddr);
3895 }
3896
3897 static void xfrm_confirm_neigh(const struct dst_entry *dst, const void *daddr)
3898 {
3899         const struct dst_entry *path = xfrm_dst_path(dst);
3900
3901         daddr = xfrm_get_dst_nexthop(dst, daddr);
3902         path->ops->confirm_neigh(path, daddr);
3903 }
3904
3905 int xfrm_policy_register_afinfo(const struct xfrm_policy_afinfo *afinfo, int family)
3906 {
3907         int err = 0;
3908
3909         if (WARN_ON(family >= ARRAY_SIZE(xfrm_policy_afinfo)))
3910                 return -EAFNOSUPPORT;
3911
3912         spin_lock(&xfrm_policy_afinfo_lock);
3913         if (unlikely(xfrm_policy_afinfo[family] != NULL))
3914                 err = -EEXIST;
3915         else {
3916                 struct dst_ops *dst_ops = afinfo->dst_ops;
3917                 if (likely(dst_ops->kmem_cachep == NULL))
3918                         dst_ops->kmem_cachep = xfrm_dst_cache;
3919                 if (likely(dst_ops->check == NULL))
3920                         dst_ops->check = xfrm_dst_check;
3921                 if (likely(dst_ops->default_advmss == NULL))
3922                         dst_ops->default_advmss = xfrm_default_advmss;
3923                 if (likely(dst_ops->mtu == NULL))
3924                         dst_ops->mtu = xfrm_mtu;
3925                 if (likely(dst_ops->negative_advice == NULL))
3926                         dst_ops->negative_advice = xfrm_negative_advice;
3927                 if (likely(dst_ops->link_failure == NULL))
3928                         dst_ops->link_failure = xfrm_link_failure;
3929                 if (likely(dst_ops->neigh_lookup == NULL))
3930                         dst_ops->neigh_lookup = xfrm_neigh_lookup;
3931                 if (likely(!dst_ops->confirm_neigh))
3932                         dst_ops->confirm_neigh = xfrm_confirm_neigh;
3933                 rcu_assign_pointer(xfrm_policy_afinfo[family], afinfo);
3934         }
3935         spin_unlock(&xfrm_policy_afinfo_lock);
3936
3937         return err;
3938 }
3939 EXPORT_SYMBOL(xfrm_policy_register_afinfo);
3940
3941 void xfrm_policy_unregister_afinfo(const struct xfrm_policy_afinfo *afinfo)
3942 {
3943         struct dst_ops *dst_ops = afinfo->dst_ops;
3944         int i;
3945
3946         for (i = 0; i < ARRAY_SIZE(xfrm_policy_afinfo); i++) {
3947                 if (xfrm_policy_afinfo[i] != afinfo)
3948                         continue;
3949                 RCU_INIT_POINTER(xfrm_policy_afinfo[i], NULL);
3950                 break;
3951         }
3952
3953         synchronize_rcu();
3954
3955         dst_ops->kmem_cachep = NULL;
3956         dst_ops->check = NULL;
3957         dst_ops->negative_advice = NULL;
3958         dst_ops->link_failure = NULL;
3959 }
3960 EXPORT_SYMBOL(xfrm_policy_unregister_afinfo);
3961
3962 void xfrm_if_register_cb(const struct xfrm_if_cb *ifcb)
3963 {
3964         spin_lock(&xfrm_if_cb_lock);
3965         rcu_assign_pointer(xfrm_if_cb, ifcb);
3966         spin_unlock(&xfrm_if_cb_lock);
3967 }
3968 EXPORT_SYMBOL(xfrm_if_register_cb);
3969
3970 void xfrm_if_unregister_cb(void)
3971 {
3972         RCU_INIT_POINTER(xfrm_if_cb, NULL);
3973         synchronize_rcu();
3974 }
3975 EXPORT_SYMBOL(xfrm_if_unregister_cb);
3976
3977 #ifdef CONFIG_XFRM_STATISTICS
3978 static int __net_init xfrm_statistics_init(struct net *net)
3979 {
3980         int rv;
3981         net->mib.xfrm_statistics = alloc_percpu(struct linux_xfrm_mib);
3982         if (!net->mib.xfrm_statistics)
3983                 return -ENOMEM;
3984         rv = xfrm_proc_init(net);
3985         if (rv < 0)
3986                 free_percpu(net->mib.xfrm_statistics);
3987         return rv;
3988 }
3989
3990 static void xfrm_statistics_fini(struct net *net)
3991 {
3992         xfrm_proc_fini(net);
3993         free_percpu(net->mib.xfrm_statistics);
3994 }
3995 #else
3996 static int __net_init xfrm_statistics_init(struct net *net)
3997 {
3998         return 0;
3999 }
4000
4001 static void xfrm_statistics_fini(struct net *net)
4002 {
4003 }
4004 #endif
4005
4006 static int __net_init xfrm_policy_init(struct net *net)
4007 {
4008         unsigned int hmask, sz;
4009         int dir, err;
4010
4011         if (net_eq(net, &init_net)) {
4012                 xfrm_dst_cache = kmem_cache_create("xfrm_dst_cache",
4013                                            sizeof(struct xfrm_dst),
4014                                            0, SLAB_HWCACHE_ALIGN|SLAB_PANIC,
4015                                            NULL);
4016                 err = rhashtable_init(&xfrm_policy_inexact_table,
4017                                       &xfrm_pol_inexact_params);
4018                 BUG_ON(err);
4019         }
4020
4021         hmask = 8 - 1;
4022         sz = (hmask+1) * sizeof(struct hlist_head);
4023
4024         net->xfrm.policy_byidx = xfrm_hash_alloc(sz);
4025         if (!net->xfrm.policy_byidx)
4026                 goto out_byidx;
4027         net->xfrm.policy_idx_hmask = hmask;
4028
4029         for (dir = 0; dir < XFRM_POLICY_MAX; dir++) {
4030                 struct xfrm_policy_hash *htab;
4031
4032                 net->xfrm.policy_count[dir] = 0;
4033                 net->xfrm.policy_count[XFRM_POLICY_MAX + dir] = 0;
4034                 INIT_HLIST_HEAD(&net->xfrm.policy_inexact[dir]);
4035
4036                 htab = &net->xfrm.policy_bydst[dir];
4037                 htab->table = xfrm_hash_alloc(sz);
4038                 if (!htab->table)
4039                         goto out_bydst;
4040                 htab->hmask = hmask;
4041                 htab->dbits4 = 32;
4042                 htab->sbits4 = 32;
4043                 htab->dbits6 = 128;
4044                 htab->sbits6 = 128;
4045         }
4046         net->xfrm.policy_hthresh.lbits4 = 32;
4047         net->xfrm.policy_hthresh.rbits4 = 32;
4048         net->xfrm.policy_hthresh.lbits6 = 128;
4049         net->xfrm.policy_hthresh.rbits6 = 128;
4050
4051         seqlock_init(&net->xfrm.policy_hthresh.lock);
4052
4053         INIT_LIST_HEAD(&net->xfrm.policy_all);
4054         INIT_LIST_HEAD(&net->xfrm.inexact_bins);
4055         INIT_WORK(&net->xfrm.policy_hash_work, xfrm_hash_resize);
4056         INIT_WORK(&net->xfrm.policy_hthresh.work, xfrm_hash_rebuild);
4057         return 0;
4058
4059 out_bydst:
4060         for (dir--; dir >= 0; dir--) {
4061                 struct xfrm_policy_hash *htab;
4062
4063                 htab = &net->xfrm.policy_bydst[dir];
4064                 xfrm_hash_free(htab->table, sz);
4065         }
4066         xfrm_hash_free(net->xfrm.policy_byidx, sz);
4067 out_byidx:
4068         return -ENOMEM;
4069 }
4070
4071 static void xfrm_policy_fini(struct net *net)
4072 {
4073         struct xfrm_pol_inexact_bin *b, *t;
4074         unsigned int sz;
4075         int dir;
4076
4077         flush_work(&net->xfrm.policy_hash_work);
4078 #ifdef CONFIG_XFRM_SUB_POLICY
4079         xfrm_policy_flush(net, XFRM_POLICY_TYPE_SUB, false);
4080 #endif
4081         xfrm_policy_flush(net, XFRM_POLICY_TYPE_MAIN, false);
4082
4083         WARN_ON(!list_empty(&net->xfrm.policy_all));
4084
4085         for (dir = 0; dir < XFRM_POLICY_MAX; dir++) {
4086                 struct xfrm_policy_hash *htab;
4087
4088                 WARN_ON(!hlist_empty(&net->xfrm.policy_inexact[dir]));
4089
4090                 htab = &net->xfrm.policy_bydst[dir];
4091                 sz = (htab->hmask + 1) * sizeof(struct hlist_head);
4092                 WARN_ON(!hlist_empty(htab->table));
4093                 xfrm_hash_free(htab->table, sz);
4094         }
4095
4096         sz = (net->xfrm.policy_idx_hmask + 1) * sizeof(struct hlist_head);
4097         WARN_ON(!hlist_empty(net->xfrm.policy_byidx));
4098         xfrm_hash_free(net->xfrm.policy_byidx, sz);
4099
4100         spin_lock_bh(&net->xfrm.xfrm_policy_lock);
4101         list_for_each_entry_safe(b, t, &net->xfrm.inexact_bins, inexact_bins)
4102                 __xfrm_policy_inexact_prune_bin(b, true);
4103         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
4104 }
4105
4106 static int __net_init xfrm_net_init(struct net *net)
4107 {
4108         int rv;
4109
4110         /* Initialize the per-net locks here */
4111         spin_lock_init(&net->xfrm.xfrm_state_lock);
4112         spin_lock_init(&net->xfrm.xfrm_policy_lock);
4113         mutex_init(&net->xfrm.xfrm_cfg_mutex);
4114
4115         rv = xfrm_statistics_init(net);
4116         if (rv < 0)
4117                 goto out_statistics;
4118         rv = xfrm_state_init(net);
4119         if (rv < 0)
4120                 goto out_state;
4121         rv = xfrm_policy_init(net);
4122         if (rv < 0)
4123                 goto out_policy;
4124         rv = xfrm_sysctl_init(net);
4125         if (rv < 0)
4126                 goto out_sysctl;
4127
4128         return 0;
4129
4130 out_sysctl:
4131         xfrm_policy_fini(net);
4132 out_policy:
4133         xfrm_state_fini(net);
4134 out_state:
4135         xfrm_statistics_fini(net);
4136 out_statistics:
4137         return rv;
4138 }
4139
4140 static void __net_exit xfrm_net_exit(struct net *net)
4141 {
4142         xfrm_sysctl_fini(net);
4143         xfrm_policy_fini(net);
4144         xfrm_state_fini(net);
4145         xfrm_statistics_fini(net);
4146 }
4147
4148 static struct pernet_operations __net_initdata xfrm_net_ops = {
4149         .init = xfrm_net_init,
4150         .exit = xfrm_net_exit,
4151 };
4152
4153 void __init xfrm_init(void)
4154 {
4155         register_pernet_subsys(&xfrm_net_ops);
4156         xfrm_dev_init();
4157         seqcount_init(&xfrm_policy_hash_generation);
4158         xfrm_input_init();
4159
4160         RCU_INIT_POINTER(xfrm_if_cb, NULL);
4161         synchronize_rcu();
4162 }
4163
4164 #ifdef CONFIG_AUDITSYSCALL
4165 static void xfrm_audit_common_policyinfo(struct xfrm_policy *xp,
4166                                          struct audit_buffer *audit_buf)
4167 {
4168         struct xfrm_sec_ctx *ctx = xp->security;
4169         struct xfrm_selector *sel = &xp->selector;
4170
4171         if (ctx)
4172                 audit_log_format(audit_buf, " sec_alg=%u sec_doi=%u sec_obj=%s",
4173                                  ctx->ctx_alg, ctx->ctx_doi, ctx->ctx_str);
4174
4175         switch (sel->family) {
4176         case AF_INET:
4177                 audit_log_format(audit_buf, " src=%pI4", &sel->saddr.a4);
4178                 if (sel->prefixlen_s != 32)
4179                         audit_log_format(audit_buf, " src_prefixlen=%d",
4180                                          sel->prefixlen_s);
4181                 audit_log_format(audit_buf, " dst=%pI4", &sel->daddr.a4);
4182                 if (sel->prefixlen_d != 32)
4183                         audit_log_format(audit_buf, " dst_prefixlen=%d",
4184                                          sel->prefixlen_d);
4185                 break;
4186         case AF_INET6:
4187                 audit_log_format(audit_buf, " src=%pI6", sel->saddr.a6);
4188                 if (sel->prefixlen_s != 128)
4189                         audit_log_format(audit_buf, " src_prefixlen=%d",
4190                                          sel->prefixlen_s);
4191                 audit_log_format(audit_buf, " dst=%pI6", sel->daddr.a6);
4192                 if (sel->prefixlen_d != 128)
4193                         audit_log_format(audit_buf, " dst_prefixlen=%d",
4194                                          sel->prefixlen_d);
4195                 break;
4196         }
4197 }
4198
4199 void xfrm_audit_policy_add(struct xfrm_policy *xp, int result, bool task_valid)
4200 {
4201         struct audit_buffer *audit_buf;
4202
4203         audit_buf = xfrm_audit_start("SPD-add");
4204         if (audit_buf == NULL)
4205                 return;
4206         xfrm_audit_helper_usrinfo(task_valid, audit_buf);
4207         audit_log_format(audit_buf, " res=%u", result);
4208         xfrm_audit_common_policyinfo(xp, audit_buf);
4209         audit_log_end(audit_buf);
4210 }
4211 EXPORT_SYMBOL_GPL(xfrm_audit_policy_add);
4212
4213 void xfrm_audit_policy_delete(struct xfrm_policy *xp, int result,
4214                               bool task_valid)
4215 {
4216         struct audit_buffer *audit_buf;
4217
4218         audit_buf = xfrm_audit_start("SPD-delete");
4219         if (audit_buf == NULL)
4220                 return;
4221         xfrm_audit_helper_usrinfo(task_valid, audit_buf);
4222         audit_log_format(audit_buf, " res=%u", result);
4223         xfrm_audit_common_policyinfo(xp, audit_buf);
4224         audit_log_end(audit_buf);
4225 }
4226 EXPORT_SYMBOL_GPL(xfrm_audit_policy_delete);
4227 #endif
4228
4229 #ifdef CONFIG_XFRM_MIGRATE
4230 static bool xfrm_migrate_selector_match(const struct xfrm_selector *sel_cmp,
4231                                         const struct xfrm_selector *sel_tgt)
4232 {
4233         if (sel_cmp->proto == IPSEC_ULPROTO_ANY) {
4234                 if (sel_tgt->family == sel_cmp->family &&
4235                     xfrm_addr_equal(&sel_tgt->daddr, &sel_cmp->daddr,
4236                                     sel_cmp->family) &&
4237                     xfrm_addr_equal(&sel_tgt->saddr, &sel_cmp->saddr,
4238                                     sel_cmp->family) &&
4239                     sel_tgt->prefixlen_d == sel_cmp->prefixlen_d &&
4240                     sel_tgt->prefixlen_s == sel_cmp->prefixlen_s) {
4241                         return true;
4242                 }
4243         } else {
4244                 if (memcmp(sel_tgt, sel_cmp, sizeof(*sel_tgt)) == 0) {
4245                         return true;
4246                 }
4247         }
4248         return false;
4249 }
4250
4251 static struct xfrm_policy *xfrm_migrate_policy_find(const struct xfrm_selector *sel,
4252                                                     u8 dir, u8 type, struct net *net)
4253 {
4254         struct xfrm_policy *pol, *ret = NULL;
4255         struct hlist_head *chain;
4256         u32 priority = ~0U;
4257
4258         spin_lock_bh(&net->xfrm.xfrm_policy_lock);
4259         chain = policy_hash_direct(net, &sel->daddr, &sel->saddr, sel->family, dir);
4260         hlist_for_each_entry(pol, chain, bydst) {
4261                 if (xfrm_migrate_selector_match(sel, &pol->selector) &&
4262                     pol->type == type) {
4263                         ret = pol;
4264                         priority = ret->priority;
4265                         break;
4266                 }
4267         }
4268         chain = &net->xfrm.policy_inexact[dir];
4269         hlist_for_each_entry(pol, chain, bydst_inexact_list) {
4270                 if ((pol->priority >= priority) && ret)
4271                         break;
4272
4273                 if (xfrm_migrate_selector_match(sel, &pol->selector) &&
4274                     pol->type == type) {
4275                         ret = pol;
4276                         break;
4277                 }
4278         }
4279
4280         xfrm_pol_hold(ret);
4281
4282         spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
4283
4284         return ret;
4285 }
4286
4287 static int migrate_tmpl_match(const struct xfrm_migrate *m, const struct xfrm_tmpl *t)
4288 {
4289         int match = 0;
4290
4291         if (t->mode == m->mode && t->id.proto == m->proto &&
4292             (m->reqid == 0 || t->reqid == m->reqid)) {
4293                 switch (t->mode) {
4294                 case XFRM_MODE_TUNNEL:
4295                 case XFRM_MODE_BEET:
4296                         if (xfrm_addr_equal(&t->id.daddr, &m->old_daddr,
4297                                             m->old_family) &&
4298                             xfrm_addr_equal(&t->saddr, &m->old_saddr,
4299                                             m->old_family)) {
4300                                 match = 1;
4301                         }
4302                         break;
4303                 case XFRM_MODE_TRANSPORT:
4304                         /* in case of transport mode, template does not store
4305                            any IP addresses, hence we just compare mode and
4306                            protocol */
4307                         match = 1;
4308                         break;
4309                 default:
4310                         break;
4311                 }
4312         }
4313         return match;
4314 }
4315
4316 /* update endpoint address(es) of template(s) */
4317 static int xfrm_policy_migrate(struct xfrm_policy *pol,
4318                                struct xfrm_migrate *m, int num_migrate)
4319 {
4320         struct xfrm_migrate *mp;
4321         int i, j, n = 0;
4322
4323         write_lock_bh(&pol->lock);
4324         if (unlikely(pol->walk.dead)) {
4325                 /* target policy has been deleted */
4326                 write_unlock_bh(&pol->lock);
4327                 return -ENOENT;
4328         }
4329
4330         for (i = 0; i < pol->xfrm_nr; i++) {
4331                 for (j = 0, mp = m; j < num_migrate; j++, mp++) {
4332                         if (!migrate_tmpl_match(mp, &pol->xfrm_vec[i]))
4333                                 continue;
4334                         n++;
4335                         if (pol->xfrm_vec[i].mode != XFRM_MODE_TUNNEL &&
4336                             pol->xfrm_vec[i].mode != XFRM_MODE_BEET)
4337                                 continue;
4338                         /* update endpoints */
4339                         memcpy(&pol->xfrm_vec[i].id.daddr, &mp->new_daddr,
4340                                sizeof(pol->xfrm_vec[i].id.daddr));
4341                         memcpy(&pol->xfrm_vec[i].saddr, &mp->new_saddr,
4342                                sizeof(pol->xfrm_vec[i].saddr));
4343                         pol->xfrm_vec[i].encap_family = mp->new_family;
4344                         /* flush bundles */
4345                         atomic_inc(&pol->genid);
4346                 }
4347         }
4348
4349         write_unlock_bh(&pol->lock);
4350
4351         if (!n)
4352                 return -ENODATA;
4353
4354         return 0;
4355 }
4356
4357 static int xfrm_migrate_check(const struct xfrm_migrate *m, int num_migrate)
4358 {
4359         int i, j;
4360
4361         if (num_migrate < 1 || num_migrate > XFRM_MAX_DEPTH)
4362                 return -EINVAL;
4363
4364         for (i = 0; i < num_migrate; i++) {
4365                 if (xfrm_addr_any(&m[i].new_daddr, m[i].new_family) ||
4366                     xfrm_addr_any(&m[i].new_saddr, m[i].new_family))
4367                         return -EINVAL;
4368
4369                 /* check if there is any duplicated entry */
4370                 for (j = i + 1; j < num_migrate; j++) {
4371                         if (!memcmp(&m[i].old_daddr, &m[j].old_daddr,
4372                                     sizeof(m[i].old_daddr)) &&
4373                             !memcmp(&m[i].old_saddr, &m[j].old_saddr,
4374                                     sizeof(m[i].old_saddr)) &&
4375                             m[i].proto == m[j].proto &&
4376                             m[i].mode == m[j].mode &&
4377                             m[i].reqid == m[j].reqid &&
4378                             m[i].old_family == m[j].old_family)
4379                                 return -EINVAL;
4380                 }
4381         }
4382
4383         return 0;
4384 }
4385
4386 int xfrm_migrate(const struct xfrm_selector *sel, u8 dir, u8 type,
4387                  struct xfrm_migrate *m, int num_migrate,
4388                  struct xfrm_kmaddress *k, struct net *net,
4389                  struct xfrm_encap_tmpl *encap)
4390 {
4391         int i, err, nx_cur = 0, nx_new = 0;
4392         struct xfrm_policy *pol = NULL;
4393         struct xfrm_state *x, *xc;
4394         struct xfrm_state *x_cur[XFRM_MAX_DEPTH];
4395         struct xfrm_state *x_new[XFRM_MAX_DEPTH];
4396         struct xfrm_migrate *mp;
4397
4398         /* Stage 0 - sanity checks */
4399         if ((err = xfrm_migrate_check(m, num_migrate)) < 0)
4400                 goto out;
4401
4402         if (dir >= XFRM_POLICY_MAX) {
4403                 err = -EINVAL;
4404                 goto out;
4405         }
4406
4407         /* Stage 1 - find policy */
4408         if ((pol = xfrm_migrate_policy_find(sel, dir, type, net)) == NULL) {
4409                 err = -ENOENT;
4410                 goto out;
4411         }
4412
4413         /* Stage 2 - find and update state(s) */
4414         for (i = 0, mp = m; i < num_migrate; i++, mp++) {
4415                 if ((x = xfrm_migrate_state_find(mp, net))) {
4416                         x_cur[nx_cur] = x;
4417                         nx_cur++;
4418                         xc = xfrm_state_migrate(x, mp, encap);
4419                         if (xc) {
4420                                 x_new[nx_new] = xc;
4421                                 nx_new++;
4422                         } else {
4423                                 err = -ENODATA;
4424                                 goto restore_state;
4425                         }
4426                 }
4427         }
4428
4429         /* Stage 3 - update policy */
4430         if ((err = xfrm_policy_migrate(pol, m, num_migrate)) < 0)
4431                 goto restore_state;
4432
4433         /* Stage 4 - delete old state(s) */
4434         if (nx_cur) {
4435                 xfrm_states_put(x_cur, nx_cur);
4436                 xfrm_states_delete(x_cur, nx_cur);
4437         }
4438
4439         /* Stage 5 - announce */
4440         km_migrate(sel, dir, type, m, num_migrate, k, encap);
4441
4442         xfrm_pol_put(pol);
4443
4444         return 0;
4445 out:
4446         return err;
4447
4448 restore_state:
4449         if (pol)
4450                 xfrm_pol_put(pol);
4451         if (nx_cur)
4452                 xfrm_states_put(x_cur, nx_cur);
4453         if (nx_new)
4454                 xfrm_states_delete(x_new, nx_new);
4455
4456         return err;
4457 }
4458 EXPORT_SYMBOL(xfrm_migrate);
4459 #endif