net/ipv4/fib_hash.c

   1 /*
   2  * INET         An implementation of the TCP/IP protocol suite for the LINUX
   3  *              operating system.  INET is implemented using the  BSD Socket
   4  *              interface as the means of communication with the user level.
   5  *
   6  *              IPv4 FIB: lookup engine and maintenance routines.
   7  *
   8  * Authors:     Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
   9  *
  10  *              This program is free software; you can redistribute it and/or
  11  *              modify it under the terms of the GNU General Public License
  12  *              as published by the Free Software Foundation; either version
  13  *              2 of the License, or (at your option) any later version.
  14  */
  15
  16 #include <asm/uaccess.h>
  17 #include <asm/system.h>
  18 #include <linux/bitops.h>
  19 #include <linux/types.h>
  20 #include <linux/kernel.h>
  21 #include <linux/mm.h>
  22 #include <linux/string.h>
  23 #include <linux/socket.h>
  24 #include <linux/sockios.h>
  25 #include <linux/errno.h>
  26 #include <linux/in.h>
  27 #include <linux/inet.h>
  28 #include <linux/inetdevice.h>
  29 #include <linux/netdevice.h>
  30 #include <linux/if_arp.h>
  31 #include <linux/proc_fs.h>
  32 #include <linux/skbuff.h>
  33 #include <linux/netlink.h>
  34 #include <linux/init.h>
  35 #include <linux/slab.h>
  36
  37 #include <net/net_namespace.h>
  38 #include <net/ip.h>
  39 #include <net/protocol.h>
  40 #include <net/route.h>
  41 #include <net/tcp.h>
  42 #include <net/sock.h>
  43 #include <net/ip_fib.h>
  44
  45 #include "fib_lookup.h"
  46
  47 static struct kmem_cache *fn_hash_kmem __read_mostly;
  48 static struct kmem_cache *fn_alias_kmem __read_mostly;
  49
  50 struct fib_node {
  51         struct hlist_node       fn_hash;
  52         struct list_head        fn_alias;
  53         __be32                  fn_key;
  54         struct fib_alias        fn_embedded_alias;
  55 };
  56
  57 #define EMBEDDED_HASH_SIZE (L1_CACHE_BYTES / sizeof(struct hlist_head))
  58
  59 struct fn_zone {
  60         struct fn_zone __rcu    *fz_next;       /* Next not empty zone  */
  61         struct hlist_head __rcu *fz_hash;       /* Hash table pointer   */
  62         seqlock_t               fz_lock;
  63         u32                     fz_hashmask;    /* (fz_divisor - 1)     */
  64
  65         u8                      fz_order;       /* Zone order (0..32)   */
  66         u8                      fz_revorder;    /* 32 - fz_order        */
  67         __be32                  fz_mask;        /* inet_make_mask(order) */
  68 #define FZ_MASK(fz)             ((fz)->fz_mask)
  69
  70         struct hlist_head       fz_embedded_hash[EMBEDDED_HASH_SIZE];
  71
  72         int                     fz_nent;        /* Number of entries    */
  73         int                     fz_divisor;     /* Hash size (mask+1)   */
  74 };
  75
  76 struct fn_hash {
  77         struct fn_zone          *fn_zones[33];
  78         struct fn_zone __rcu    *fn_zone_list;
  79 };
  80
  81 static inline u32 fn_hash(__be32 key, struct fn_zone *fz)
  82 {
  83         u32 h = ntohl(key) >> fz->fz_revorder;
  84         h ^= (h>>20);
  85         h ^= (h>>10);
  86         h ^= (h>>5);
  87         h &= fz->fz_hashmask;
  88         return h;
  89 }
  90
  91 static inline __be32 fz_key(__be32 dst, struct fn_zone *fz)
  92 {
  93         return dst & FZ_MASK(fz);
  94 }
  95
  96 static unsigned int fib_hash_genid;
  97
  98 #define FZ_MAX_DIVISOR ((PAGE_SIZE<<MAX_ORDER) / sizeof(struct hlist_head))
  99
 100 static struct hlist_head *fz_hash_alloc(int divisor)
 101 {
 102         unsigned long size = divisor * sizeof(struct hlist_head);
 103
 104         if (size <= PAGE_SIZE)
 105                 return kzalloc(size, GFP_KERNEL);
 106
 107         return (struct hlist_head *)
 108                 __get_free_pages(GFP_KERNEL | __GFP_ZERO, get_order(size));
 109 }
 110
 111 /* The fib hash lock must be held when this is called. */
 112 static inline void fn_rebuild_zone(struct fn_zone *fz,
 113                                    struct hlist_head *old_ht,
 114                                    int old_divisor)
 115 {
 116         int i;
 117
 118         for (i = 0; i < old_divisor; i++) {
 119                 struct hlist_node *node, *n;
 120                 struct fib_node *f;
 121
 122                 hlist_for_each_entry_safe(f, node, n, &old_ht[i], fn_hash) {
 123                         struct hlist_head *new_head;
 124
 125                         hlist_del_rcu(&f->fn_hash);
 126
 127                         new_head = rcu_dereference_protected(fz->fz_hash, 1) +
 128                                    fn_hash(f->fn_key, fz);
 129                         hlist_add_head_rcu(&f->fn_hash, new_head);
 130                 }
 131         }
 132 }
 133
 134 static void fz_hash_free(struct hlist_head *hash, int divisor)
 135 {
 136         unsigned long size = divisor * sizeof(struct hlist_head);
 137
 138         if (size <= PAGE_SIZE)
 139                 kfree(hash);
 140         else
 141                 free_pages((unsigned long)hash, get_order(size));
 142 }
 143
 144 static void fn_rehash_zone(struct fn_zone *fz)
 145 {
 146         struct hlist_head *ht, *old_ht;
 147         int old_divisor, new_divisor;
 148         u32 new_hashmask;
 149
 150         new_divisor = old_divisor = fz->fz_divisor;
 151
 152         switch (old_divisor) {
 153         case EMBEDDED_HASH_SIZE:
 154                 new_divisor *= EMBEDDED_HASH_SIZE;
 155                 break;
 156         case EMBEDDED_HASH_SIZE*EMBEDDED_HASH_SIZE:
 157                 new_divisor *= (EMBEDDED_HASH_SIZE/2);
 158                 break;
 159         default:
 160                 if ((old_divisor << 1) > FZ_MAX_DIVISOR) {
 161                         printk(KERN_CRIT "route.c: bad divisor %d!\n", old_divisor);
 162                         return;
 163                 }
 164                 new_divisor = (old_divisor << 1);
 165                 break;
 166         }
 167
 168         new_hashmask = (new_divisor - 1);
 169
 170 #if RT_CACHE_DEBUG >= 2
 171         printk(KERN_DEBUG "fn_rehash_zone: hash for zone %d grows from %d\n",
 172                fz->fz_order, old_divisor);
 173 #endif
 174
 175         ht = fz_hash_alloc(new_divisor);
 176
 177         if (ht) {
 178                 struct fn_zone nfz;
 179
 180                 memcpy(&nfz, fz, sizeof(nfz));
 181
 182                 write_seqlock_bh(&fz->fz_lock);
 183                 old_ht = rcu_dereference_protected(fz->fz_hash, 1);
 184                 RCU_INIT_POINTER(nfz.fz_hash, ht);
 185                 nfz.fz_hashmask = new_hashmask;
 186                 nfz.fz_divisor = new_divisor;
 187                 fn_rebuild_zone(&nfz, old_ht, old_divisor);
 188                 fib_hash_genid++;
 189                 rcu_assign_pointer(fz->fz_hash, ht);
 190                 fz->fz_hashmask = new_hashmask;
 191                 fz->fz_divisor = new_divisor;
 192                 write_sequnlock_bh(&fz->fz_lock);
 193
 194                 if (old_ht != fz->fz_embedded_hash) {
 195                         synchronize_rcu();
 196                         fz_hash_free(old_ht, old_divisor);
 197                 }
 198         }
 199 }
 200
 201 static void fn_free_node_rcu(struct rcu_head *head)
 202 {
 203         struct fib_node *f = container_of(head, struct fib_node, fn_embedded_alias.rcu);
 204
 205         kmem_cache_free(fn_hash_kmem, f);
 206 }
 207
 208 static inline void fn_free_node(struct fib_node *f)
 209 {
 210         call_rcu(&f->fn_embedded_alias.rcu, fn_free_node_rcu);
 211 }
 212
 213 static void fn_free_alias_rcu(struct rcu_head *head)
 214 {
 215         struct fib_alias *fa = container_of(head, struct fib_alias, rcu);
 216
 217         kmem_cache_free(fn_alias_kmem, fa);
 218 }
 219
 220 static inline void fn_free_alias(struct fib_alias *fa, struct fib_node *f)
 221 {
 222         fib_release_info(fa->fa_info);
 223         if (fa == &f->fn_embedded_alias)
 224                 fa->fa_info = NULL;
 225         else
 226                 call_rcu(&fa->rcu, fn_free_alias_rcu);
 227 }
 228
 229 static struct fn_zone *
 230 fn_new_zone(struct fn_hash *table, int z)
 231 {
 232         int i;
 233         struct fn_zone *fz = kzalloc(sizeof(struct fn_zone), GFP_KERNEL);
 234         if (!fz)
 235                 return NULL;
 236
 237         seqlock_init(&fz->fz_lock);
 238         fz->fz_divisor = z ? EMBEDDED_HASH_SIZE : 1;
 239         fz->fz_hashmask = fz->fz_divisor - 1;
 240         RCU_INIT_POINTER(fz->fz_hash, fz->fz_embedded_hash);
 241         fz->fz_order = z;
 242         fz->fz_revorder = 32 - z;
 243         fz->fz_mask = inet_make_mask(z);
 244
 245         /* Find the first not empty zone with more specific mask */
 246         for (i = z + 1; i <= 32; i++)
 247                 if (table->fn_zones[i])
 248                         break;
 249         if (i > 32) {
 250                 /* No more specific masks, we are the first. */
 251                 rcu_assign_pointer(fz->fz_next,
 252                                    rtnl_dereference(table->fn_zone_list));
 253                 rcu_assign_pointer(table->fn_zone_list, fz);
 254         } else {
 255                 rcu_assign_pointer(fz->fz_next,
 256                                    rtnl_dereference(table->fn_zones[i]->fz_next));
 257                 rcu_assign_pointer(table->fn_zones[i]->fz_next, fz);
 258         }
 259         table->fn_zones[z] = fz;
 260         fib_hash_genid++;
 261         return fz;
 262 }
 263
 264 int fib_table_lookup(struct fib_table *tb,
 265                      const struct flowi *flp, struct fib_result *res,
 266                      int fib_flags)
 267 {
 268         int err;
 269         struct fn_zone *fz;
 270         struct fn_hash *t = (struct fn_hash *)tb->tb_data;
 271
 272         rcu_read_lock();
 273         for (fz = rcu_dereference(t->fn_zone_list);
 274              fz != NULL;
 275              fz = rcu_dereference(fz->fz_next)) {
 276                 struct hlist_head *head;
 277                 struct hlist_node *node;
 278                 struct fib_node *f;
 279                 __be32 k;
 280                 unsigned int seq;
 281
 282                 do {
 283                         seq = read_seqbegin(&fz->fz_lock);
 284                         k = fz_key(flp->fl4_dst, fz);
 285
 286                         head = rcu_dereference(fz->fz_hash) + fn_hash(k, fz);
 287                         hlist_for_each_entry_rcu(f, node, head, fn_hash) {
 288                                 if (f->fn_key != k)
 289                                         continue;
 290
 291                                 err = fib_semantic_match(tb, &f->fn_alias,
 292                                                  flp, res,
 293                                                  fz->fz_order, fib_flags);
 294                                 if (err <= 0)
 295                                         goto out;
 296                         }
 297                 } while (read_seqretry(&fz->fz_lock, seq));
 298         }
 299         err = 1;
 300 out:
 301         rcu_read_unlock();
 302         return err;
 303 }
 304
 305 /* Insert node F to FZ. */
 306 static inline void fib_insert_node(struct fn_zone *fz, struct fib_node *f)
 307 {
 308         struct hlist_head *head = rtnl_dereference(fz->fz_hash) + fn_hash(f->fn_key, fz);
 309
 310         hlist_add_head_rcu(&f->fn_hash, head);
 311 }
 312
 313 /* Return the node in FZ matching KEY. */
 314 static struct fib_node *fib_find_node(struct fn_zone *fz, __be32 key)
 315 {
 316         struct hlist_head *head = rtnl_dereference(fz->fz_hash) + fn_hash(key, fz);
 317         struct hlist_node *node;
 318         struct fib_node *f;
 319
 320         hlist_for_each_entry_rcu(f, node, head, fn_hash) {
 321                 if (f->fn_key == key)
 322                         return f;
 323         }
 324
 325         return NULL;
 326 }
 327
 328
 329 static struct fib_alias *fib_fast_alloc(struct fib_node *f)
 330 {
 331         struct fib_alias *fa = &f->fn_embedded_alias;
 332
 333         if (fa->fa_info != NULL)
 334                 fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL);
 335         return fa;
 336 }
 337
 338 /* Caller must hold RTNL. */
 339 int fib_table_insert(struct fib_table *tb, struct fib_config *cfg)
 340 {
 341         struct fn_hash *table = (struct fn_hash *) tb->tb_data;
 342         struct fib_node *new_f = NULL;
 343         struct fib_node *f;
 344         struct fib_alias *fa, *new_fa;
 345         struct fn_zone *fz;
 346         struct fib_info *fi;
 347         u8 tos = cfg->fc_tos;
 348         __be32 key;
 349         int err;
 350
 351         if (cfg->fc_dst_len > 32)
 352                 return -EINVAL;
 353
 354         fz = table->fn_zones[cfg->fc_dst_len];
 355         if (!fz && !(fz = fn_new_zone(table, cfg->fc_dst_len)))
 356                 return -ENOBUFS;
 357
 358         key = 0;
 359         if (cfg->fc_dst) {
 360                 if (cfg->fc_dst & ~FZ_MASK(fz))
 361                         return -EINVAL;
 362                 key = fz_key(cfg->fc_dst, fz);
 363         }
 364
 365         fi = fib_create_info(cfg);
 366         if (IS_ERR(fi))
 367                 return PTR_ERR(fi);
 368
 369         if (fz->fz_nent > (fz->fz_divisor<<1) &&
 370             fz->fz_divisor < FZ_MAX_DIVISOR &&
 371             (cfg->fc_dst_len == 32 ||
 372              (1 << cfg->fc_dst_len) > fz->fz_divisor))
 373                 fn_rehash_zone(fz);
 374
 375         f = fib_find_node(fz, key);
 376
 377         if (!f)
 378                 fa = NULL;
 379         else
 380                 fa = fib_find_alias(&f->fn_alias, tos, fi->fib_priority);
 381
 382         /* Now fa, if non-NULL, points to the first fib alias
 383          * with the same keys [prefix,tos,priority], if such key already
 384          * exists or to the node before which we will insert new one.
 385          *
 386          * If fa is NULL, we will need to allocate a new one and
 387          * insert to the head of f.
 388          *
 389          * If f is NULL, no fib node matched the destination key
 390          * and we need to allocate a new one of those as well.
 391          */
 392
 393         if (fa && fa->fa_tos == tos &&
 394             fa->fa_info->fib_priority == fi->fib_priority) {
 395                 struct fib_alias *fa_first, *fa_match;
 396
 397                 err = -EEXIST;
 398                 if (cfg->fc_nlflags & NLM_F_EXCL)
 399                         goto out;
 400
 401                 /* We have 2 goals:
 402                  * 1. Find exact match for type, scope, fib_info to avoid
 403                  * duplicate routes
 404                  * 2. Find next 'fa' (or head), NLM_F_APPEND inserts before it
 405                  */
 406                 fa_match = NULL;
 407                 fa_first = fa;
 408                 fa = list_entry(fa->fa_list.prev, struct fib_alias, fa_list);
 409                 list_for_each_entry_continue(fa, &f->fn_alias, fa_list) {
 410                         if (fa->fa_tos != tos)
 411                                 break;
 412                         if (fa->fa_info->fib_priority != fi->fib_priority)
 413                                 break;
 414                         if (fa->fa_type == cfg->fc_type &&
 415                             fa->fa_scope == cfg->fc_scope &&
 416                             fa->fa_info == fi) {
 417                                 fa_match = fa;
 418                                 break;
 419                         }
 420                 }
 421
 422                 if (cfg->fc_nlflags & NLM_F_REPLACE) {
 423                         u8 state;
 424
 425                         fa = fa_first;
 426                         if (fa_match) {
 427                                 if (fa == fa_match)
 428                                         err = 0;
 429                                 goto out;
 430                         }
 431                         err = -ENOBUFS;
 432                         new_fa = fib_fast_alloc(f);
 433                         if (new_fa == NULL)
 434                                 goto out;
 435
 436                         new_fa->fa_tos = fa->fa_tos;
 437                         new_fa->fa_info = fi;
 438                         new_fa->fa_type = cfg->fc_type;
 439                         new_fa->fa_scope = cfg->fc_scope;
 440                         state = fa->fa_state;
 441                         new_fa->fa_state = state & ~FA_S_ACCESSED;
 442                         fib_hash_genid++;
 443                         list_replace_rcu(&fa->fa_list, &new_fa->fa_list);
 444
 445                         fn_free_alias(fa, f);
 446                         if (state & FA_S_ACCESSED)
 447                                 rt_cache_flush(cfg->fc_nlinfo.nl_net, -1);
 448                         rtmsg_fib(RTM_NEWROUTE, key, new_fa, cfg->fc_dst_len,
 449                                   tb->tb_id, &cfg->fc_nlinfo, NLM_F_REPLACE);
 450                         return 0;
 451                 }
 452
 453                 /* Error if we find a perfect match which
 454                  * uses the same scope, type, and nexthop
 455                  * information.
 456                  */
 457                 if (fa_match)
 458                         goto out;
 459
 460                 if (!(cfg->fc_nlflags & NLM_F_APPEND))
 461                         fa = fa_first;
 462         }
 463
 464         err = -ENOENT;
 465         if (!(cfg->fc_nlflags & NLM_F_CREATE))
 466                 goto out;
 467
 468         err = -ENOBUFS;
 469
 470         if (!f) {
 471                 new_f = kmem_cache_zalloc(fn_hash_kmem, GFP_KERNEL);
 472                 if (new_f == NULL)
 473                         goto out;
 474
 475                 INIT_HLIST_NODE(&new_f->fn_hash);
 476                 INIT_LIST_HEAD(&new_f->fn_alias);
 477                 new_f->fn_key = key;
 478                 f = new_f;
 479         }
 480
 481         new_fa = fib_fast_alloc(f);
 482         if (new_fa == NULL)
 483                 goto out;
 484
 485         new_fa->fa_info = fi;
 486         new_fa->fa_tos = tos;
 487         new_fa->fa_type = cfg->fc_type;
 488         new_fa->fa_scope = cfg->fc_scope;
 489         new_fa->fa_state = 0;
 490
 491         /*
 492          * Insert new entry to the list.
 493          */
 494
 495         if (new_f)
 496                 fib_insert_node(fz, new_f);
 497         list_add_tail_rcu(&new_fa->fa_list,
 498                  (fa ? &fa->fa_list : &f->fn_alias));
 499         fib_hash_genid++;
 500
 501         if (new_f)
 502                 fz->fz_nent++;
 503         rt_cache_flush(cfg->fc_nlinfo.nl_net, -1);
 504
 505         rtmsg_fib(RTM_NEWROUTE, key, new_fa, cfg->fc_dst_len, tb->tb_id,
 506                   &cfg->fc_nlinfo, 0);
 507         return 0;
 508
 509 out:
 510         if (new_f)
 511                 kmem_cache_free(fn_hash_kmem, new_f);
 512         fib_release_info(fi);
 513         return err;
 514 }
 515
 516 int fib_table_delete(struct fib_table *tb, struct fib_config *cfg)
 517 {
 518         struct fn_hash *table = (struct fn_hash *)tb->tb_data;
 519         struct fib_node *f;
 520         struct fib_alias *fa, *fa_to_delete;
 521         struct fn_zone *fz;
 522         __be32 key;
 523
 524         if (cfg->fc_dst_len > 32)
 525                 return -EINVAL;
 526
 527         if ((fz  = table->fn_zones[cfg->fc_dst_len]) == NULL)
 528                 return -ESRCH;
 529
 530         key = 0;
 531         if (cfg->fc_dst) {
 532                 if (cfg->fc_dst & ~FZ_MASK(fz))
 533                         return -EINVAL;
 534                 key = fz_key(cfg->fc_dst, fz);
 535         }
 536
 537         f = fib_find_node(fz, key);
 538
 539         if (!f)
 540                 fa = NULL;
 541         else
 542                 fa = fib_find_alias(&f->fn_alias, cfg->fc_tos, 0);
 543         if (!fa)
 544                 return -ESRCH;
 545
 546         fa_to_delete = NULL;
 547         fa = list_entry(fa->fa_list.prev, struct fib_alias, fa_list);
 548         list_for_each_entry_continue(fa, &f->fn_alias, fa_list) {
 549                 struct fib_info *fi = fa->fa_info;
 550
 551                 if (fa->fa_tos != cfg->fc_tos)
 552                         break;
 553
 554                 if ((!cfg->fc_type ||
 555                      fa->fa_type == cfg->fc_type) &&
 556                     (cfg->fc_scope == RT_SCOPE_NOWHERE ||
 557                      fa->fa_scope == cfg->fc_scope) &&
 558                     (!cfg->fc_protocol ||
 559                      fi->fib_protocol == cfg->fc_protocol) &&
 560                     fib_nh_match(cfg, fi) == 0) {
 561                         fa_to_delete = fa;
 562                         break;
 563                 }
 564         }
 565
 566         if (fa_to_delete) {
 567                 int kill_fn;
 568
 569                 fa = fa_to_delete;
 570                 rtmsg_fib(RTM_DELROUTE, key, fa, cfg->fc_dst_len,
 571                           tb->tb_id, &cfg->fc_nlinfo, 0);
 572
 573                 kill_fn = 0;
 574                 list_del_rcu(&fa->fa_list);
 575                 if (list_empty(&f->fn_alias)) {
 576                         hlist_del_rcu(&f->fn_hash);
 577                         kill_fn = 1;
 578                 }
 579                 fib_hash_genid++;
 580
 581                 if (fa->fa_state & FA_S_ACCESSED)
 582                         rt_cache_flush(cfg->fc_nlinfo.nl_net, -1);
 583                 fn_free_alias(fa, f);
 584                 if (kill_fn) {
 585                         fn_free_node(f);
 586                         fz->fz_nent--;
 587                 }
 588
 589                 return 0;
 590         }
 591         return -ESRCH;
 592 }
 593
 594 static int fn_flush_list(struct fn_zone *fz, int idx)
 595 {
 596         struct hlist_head *head = rtnl_dereference(fz->fz_hash) + idx;
 597         struct hlist_node *node, *n;
 598         struct fib_node *f;
 599         int found = 0;
 600
 601         hlist_for_each_entry_safe(f, node, n, head, fn_hash) {
 602                 struct fib_alias *fa, *fa_node;
 603                 int kill_f;
 604
 605                 kill_f = 0;
 606                 list_for_each_entry_safe(fa, fa_node, &f->fn_alias, fa_list) {
 607                         struct fib_info *fi = fa->fa_info;
 608
 609                         if (fi && (fi->fib_flags&RTNH_F_DEAD)) {
 610                                 list_del_rcu(&fa->fa_list);
 611                                 if (list_empty(&f->fn_alias)) {
 612                                         hlist_del_rcu(&f->fn_hash);
 613                                         kill_f = 1;
 614                                 }
 615                                 fib_hash_genid++;
 616
 617                                 fn_free_alias(fa, f);
 618                                 found++;
 619                         }
 620                 }
 621                 if (kill_f) {
 622                         fn_free_node(f);
 623                         fz->fz_nent--;
 624                 }
 625         }
 626         return found;
 627 }
 628
 629 /* caller must hold RTNL. */
 630 int fib_table_flush(struct fib_table *tb)
 631 {
 632         struct fn_hash *table = (struct fn_hash *) tb->tb_data;
 633         struct fn_zone *fz;
 634         int found = 0;
 635
 636         for (fz = rtnl_dereference(table->fn_zone_list);
 637              fz != NULL;
 638              fz = rtnl_dereference(fz->fz_next)) {
 639                 int i;
 640
 641                 for (i = fz->fz_divisor - 1; i >= 0; i--)
 642                         found += fn_flush_list(fz, i);
 643         }
 644         return found;
 645 }
 646
 647 void fib_free_table(struct fib_table *tb)
 648 {
 649         struct fn_hash *table = (struct fn_hash *) tb->tb_data;
 650         struct fn_zone *fz, *next;
 651
 652         next = table->fn_zone_list;
 653         while (next != NULL) {
 654                 fz = next;
 655                 next = fz->fz_next;
 656
 657                 if (fz->fz_hash != fz->fz_embedded_hash)
 658                         fz_hash_free(fz->fz_hash, fz->fz_divisor);
 659
 660                 kfree(fz);
 661         }
 662
 663         kfree(tb);
 664 }
 665
 666 static inline int
 667 fn_hash_dump_bucket(struct sk_buff *skb, struct netlink_callback *cb,
 668                      struct fib_table *tb,
 669                      struct fn_zone *fz,
 670                      struct hlist_head *head)
 671 {
 672         struct hlist_node *node;
 673         struct fib_node *f;
 674         int i, s_i;
 675
 676         s_i = cb->args[4];
 677         i = 0;
 678         hlist_for_each_entry_rcu(f, node, head, fn_hash) {
 679                 struct fib_alias *fa;
 680
 681                 list_for_each_entry_rcu(fa, &f->fn_alias, fa_list) {
 682                         if (i < s_i)
 683                                 goto next;
 684
 685                         if (fib_dump_info(skb, NETLINK_CB(cb->skb).pid,
 686                                           cb->nlh->nlmsg_seq,
 687                                           RTM_NEWROUTE,
 688                                           tb->tb_id,
 689                                           fa->fa_type,
 690                                           fa->fa_scope,
 691                                           f->fn_key,
 692                                           fz->fz_order,
 693                                           fa->fa_tos,
 694                                           fa->fa_info,
 695                                           NLM_F_MULTI) < 0) {
 696                                 cb->args[4] = i;
 697                                 return -1;
 698                         }
 699 next:
 700                         i++;
 701                 }
 702         }
 703         cb->args[4] = i;
 704         return skb->len;
 705 }
 706
 707 static inline int
 708 fn_hash_dump_zone(struct sk_buff *skb, struct netlink_callback *cb,
 709                    struct fib_table *tb,
 710                    struct fn_zone *fz)
 711 {
 712         int h, s_h;
 713         struct hlist_head *head = rcu_dereference(fz->fz_hash);
 714
 715         if (head == NULL)
 716                 return skb->len;
 717         s_h = cb->args[3];
 718         for (h = s_h; h < fz->fz_divisor; h++) {
 719                 if (hlist_empty(head + h))
 720                         continue;
 721                 if (fn_hash_dump_bucket(skb, cb, tb, fz, head + h) < 0) {
 722                         cb->args[3] = h;
 723                         return -1;
 724                 }
 725                 memset(&cb->args[4], 0,
 726                        sizeof(cb->args) - 4*sizeof(cb->args[0]));
 727         }
 728         cb->args[3] = h;
 729         return skb->len;
 730 }
 731
 732 int fib_table_dump(struct fib_table *tb, struct sk_buff *skb,
 733                    struct netlink_callback *cb)
 734 {
 735         int m = 0, s_m;
 736         struct fn_zone *fz;
 737         struct fn_hash *table = (struct fn_hash *)tb->tb_data;
 738
 739         s_m = cb->args[2];
 740         rcu_read_lock();
 741         for (fz = rcu_dereference(table->fn_zone_list);
 742              fz != NULL;
 743              fz = rcu_dereference(fz->fz_next), m++) {
 744                 if (m < s_m)
 745                         continue;
 746                 if (fn_hash_dump_zone(skb, cb, tb, fz) < 0) {
 747                         cb->args[2] = m;
 748                         rcu_read_unlock();
 749                         return -1;
 750                 }
 751                 memset(&cb->args[3], 0,
 752                        sizeof(cb->args) - 3*sizeof(cb->args[0]));
 753         }
 754         rcu_read_unlock();
 755         cb->args[2] = m;
 756         return skb->len;
 757 }
 758
 759 void __init fib_hash_init(void)
 760 {
 761         fn_hash_kmem = kmem_cache_create("ip_fib_hash", sizeof(struct fib_node),
 762                                          0, SLAB_PANIC, NULL);
 763
 764         fn_alias_kmem = kmem_cache_create("ip_fib_alias", sizeof(struct fib_alias),
 765                                           0, SLAB_PANIC, NULL);
 766
 767 }
 768
 769 struct fib_table *fib_hash_table(u32 id)
 770 {
 771         struct fib_table *tb;
 772
 773         tb = kmalloc(sizeof(struct fib_table) + sizeof(struct fn_hash),
 774                      GFP_KERNEL);
 775         if (tb == NULL)
 776                 return NULL;
 777
 778         tb->tb_id = id;
 779         tb->tb_default = -1;
 780
 781         memset(tb->tb_data, 0, sizeof(struct fn_hash));
 782         return tb;
 783 }
 784
 785 /* ------------------------------------------------------------------------ */
 786 #ifdef CONFIG_PROC_FS
 787
 788 struct fib_iter_state {
 789         struct seq_net_private p;
 790         struct fn_zone  *zone;
 791         int             bucket;
 792         struct hlist_head *hash_head;
 793         struct fib_node *fn;
 794         struct fib_alias *fa;
 795         loff_t pos;
 796         unsigned int genid;
 797         int valid;
 798 };
 799
 800 static struct fib_alias *fib_get_first(struct seq_file *seq)
 801 {
 802         struct fib_iter_state *iter = seq->private;
 803         struct fib_table *main_table;
 804         struct fn_hash *table;
 805
 806         main_table = fib_get_table(seq_file_net(seq), RT_TABLE_MAIN);
 807         table = (struct fn_hash *)main_table->tb_data;
 808
 809         iter->bucket    = 0;
 810         iter->hash_head = NULL;
 811         iter->fn        = NULL;
 812         iter->fa        = NULL;
 813         iter->pos       = 0;
 814         iter->genid     = fib_hash_genid;
 815         iter->valid     = 1;
 816
 817         for (iter->zone = rcu_dereference(table->fn_zone_list);
 818              iter->zone != NULL;
 819              iter->zone = rcu_dereference(iter->zone->fz_next)) {
 820                 int maxslot;
 821
 822                 if (!iter->zone->fz_nent)
 823                         continue;
 824
 825                 iter->hash_head = rcu_dereference(iter->zone->fz_hash);
 826                 maxslot = iter->zone->fz_divisor;
 827
 828                 for (iter->bucket = 0; iter->bucket < maxslot;
 829                      ++iter->bucket, ++iter->hash_head) {
 830                         struct hlist_node *node;
 831                         struct fib_node *fn;
 832
 833                         hlist_for_each_entry(fn, node, iter->hash_head, fn_hash) {
 834                                 struct fib_alias *fa;
 835
 836                                 list_for_each_entry(fa, &fn->fn_alias, fa_list) {
 837                                         iter->fn = fn;
 838                                         iter->fa = fa;
 839                                         goto out;
 840                                 }
 841                         }
 842                 }
 843         }
 844 out:
 845         return iter->fa;
 846 }
 847
 848 static struct fib_alias *fib_get_next(struct seq_file *seq)
 849 {
 850         struct fib_iter_state *iter = seq->private;
 851         struct fib_node *fn;
 852         struct fib_alias *fa;
 853
 854         /* Advance FA, if any. */
 855         fn = iter->fn;
 856         fa = iter->fa;
 857         if (fa) {
 858                 BUG_ON(!fn);
 859                 list_for_each_entry_continue(fa, &fn->fn_alias, fa_list) {
 860                         iter->fa = fa;
 861                         goto out;
 862                 }
 863         }
 864
 865         fa = iter->fa = NULL;
 866
 867         /* Advance FN. */
 868         if (fn) {
 869                 struct hlist_node *node = &fn->fn_hash;
 870                 hlist_for_each_entry_continue(fn, node, fn_hash) {
 871                         iter->fn = fn;
 872
 873                         list_for_each_entry(fa, &fn->fn_alias, fa_list) {
 874                                 iter->fa = fa;
 875                                 goto out;
 876                         }
 877                 }
 878         }
 879
 880         fn = iter->fn = NULL;
 881
 882         /* Advance hash chain. */
 883         if (!iter->zone)
 884                 goto out;
 885
 886         for (;;) {
 887                 struct hlist_node *node;
 888                 int maxslot;
 889
 890                 maxslot = iter->zone->fz_divisor;
 891
 892                 while (++iter->bucket < maxslot) {
 893                         iter->hash_head++;
 894
 895                         hlist_for_each_entry(fn, node, iter->hash_head, fn_hash) {
 896                                 list_for_each_entry(fa, &fn->fn_alias, fa_list) {
 897                                         iter->fn = fn;
 898                                         iter->fa = fa;
 899                                         goto out;
 900                                 }
 901                         }
 902                 }
 903
 904                 iter->zone = rcu_dereference(iter->zone->fz_next);
 905
 906                 if (!iter->zone)
 907                         goto out;
 908
 909                 iter->bucket = 0;
 910                 iter->hash_head = rcu_dereference(iter->zone->fz_hash);
 911
 912                 hlist_for_each_entry(fn, node, iter->hash_head, fn_hash) {
 913                         list_for_each_entry(fa, &fn->fn_alias, fa_list) {
 914                                 iter->fn = fn;
 915                                 iter->fa = fa;
 916                                 goto out;
 917                         }
 918                 }
 919         }
 920 out:
 921         iter->pos++;
 922         return fa;
 923 }
 924
 925 static struct fib_alias *fib_get_idx(struct seq_file *seq, loff_t pos)
 926 {
 927         struct fib_iter_state *iter = seq->private;
 928         struct fib_alias *fa;
 929
 930         if (iter->valid && pos >= iter->pos && iter->genid == fib_hash_genid) {
 931                 fa   = iter->fa;
 932                 pos -= iter->pos;
 933         } else
 934                 fa = fib_get_first(seq);
 935
 936         if (fa)
 937                 while (pos && (fa = fib_get_next(seq)))
 938                         --pos;
 939         return pos ? NULL : fa;
 940 }
 941
 942 static void *fib_seq_start(struct seq_file *seq, loff_t *pos)
 943         __acquires(RCU)
 944 {
 945         void *v = NULL;
 946
 947         rcu_read_lock();
 948         if (fib_get_table(seq_file_net(seq), RT_TABLE_MAIN))
 949                 v = *pos ? fib_get_idx(seq, *pos - 1) : SEQ_START_TOKEN;
 950         return v;
 951 }
 952
 953 static void *fib_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 954 {
 955         ++*pos;
 956         return v == SEQ_START_TOKEN ? fib_get_first(seq) : fib_get_next(seq);
 957 }
 958
 959 static void fib_seq_stop(struct seq_file *seq, void *v)
 960         __releases(RCU)
 961 {
 962         rcu_read_unlock();
 963 }
 964
 965 static unsigned fib_flag_trans(int type, __be32 mask, struct fib_info *fi)
 966 {
 967         static const unsigned type2flags[RTN_MAX + 1] = {
 968                 [7] = RTF_REJECT,
 969                 [8] = RTF_REJECT,
 970         };
 971         unsigned flags = type2flags[type];
 972
 973         if (fi && fi->fib_nh->nh_gw)
 974                 flags |= RTF_GATEWAY;
 975         if (mask == htonl(0xFFFFFFFF))
 976                 flags |= RTF_HOST;
 977         flags |= RTF_UP;
 978         return flags;
 979 }
 980
 981 /*
 982  *      This outputs /proc/net/route.
 983  *
 984  *      It always works in backward compatibility mode.
 985  *      The format of the file is not supposed to be changed.
 986  */
 987 static int fib_seq_show(struct seq_file *seq, void *v)
 988 {
 989         struct fib_iter_state *iter;
 990         int len;
 991         __be32 prefix, mask;
 992         unsigned flags;
 993         struct fib_node *f;
 994         struct fib_alias *fa;
 995         struct fib_info *fi;
 996
 997         if (v == SEQ_START_TOKEN) {
 998                 seq_printf(seq, "%-127s\n", "Iface\tDestination\tGateway "
 999                            "\tFlags\tRefCnt\tUse\tMetric\tMask\t\tMTU"
1000                            "\tWindow\tIRTT");
1001                 goto out;
1002         }
1003
1004         iter    = seq->private;
1005         f       = iter->fn;
1006         fa      = iter->fa;
1007         fi      = fa->fa_info;
1008         prefix  = f->fn_key;
1009         mask    = FZ_MASK(iter->zone);
1010         flags   = fib_flag_trans(fa->fa_type, mask, fi);
1011         if (fi)
1012                 seq_printf(seq,
1013                          "%s\t%08X\t%08X\t%04X\t%d\t%u\t%d\t%08X\t%d\t%u\t%u%n",
1014                          fi->fib_dev ? fi->fib_dev->name : "*", prefix,
1015                          fi->fib_nh->nh_gw, flags, 0, 0, fi->fib_priority,
1016                          mask, (fi->fib_advmss ? fi->fib_advmss + 40 : 0),
1017                          fi->fib_window,
1018                          fi->fib_rtt >> 3, &len);
1019         else
1020                 seq_printf(seq,
1021                          "*\t%08X\t%08X\t%04X\t%d\t%u\t%d\t%08X\t%d\t%u\t%u%n",
1022                          prefix, 0, flags, 0, 0, 0, mask, 0, 0, 0, &len);
1023
1024         seq_printf(seq, "%*s\n", 127 - len, "");
1025 out:
1026         return 0;
1027 }
1028
1029 static const struct seq_operations fib_seq_ops = {
1030         .start  = fib_seq_start,
1031         .next   = fib_seq_next,
1032         .stop   = fib_seq_stop,
1033         .show   = fib_seq_show,
1034 };
1035
1036 static int fib_seq_open(struct inode *inode, struct file *file)
1037 {
1038         return seq_open_net(inode, file, &fib_seq_ops,
1039                             sizeof(struct fib_iter_state));
1040 }
1041
1042 static const struct file_operations fib_seq_fops = {
1043         .owner          = THIS_MODULE,
1044         .open           = fib_seq_open,
1045         .read           = seq_read,
1046         .llseek         = seq_lseek,
1047         .release        = seq_release_net,
1048 };
1049
1050 int __net_init fib_proc_init(struct net *net)
1051 {
1052         if (!proc_net_fops_create(net, "route", S_IRUGO, &fib_seq_fops))
1053                 return -ENOMEM;
1054         return 0;
1055 }
1056
1057 void __net_exit fib_proc_exit(struct net *net)
1058 {
1059         proc_net_remove(net, "route");
1060 }
1061 #endif /* CONFIG_PROC_FS */