1 /* 2 * This program is free software; you can redistribute it and/or 3 * modify it under the terms of the GNU General Public License 4 * as published by the Free Software Foundation; either version 5 * 2 of the License, or (at your option) any later version. 6 * 7 * Robert Olsson <robert.olsson@its.uu.se> Uppsala Universitet 8 * & Swedish University of Agricultural Sciences. 9 * 10 * Jens Laas <jens.laas@data.slu.se> Swedish University of 11 * Agricultural Sciences. 12 * 13 * Hans Liss <hans.liss@its.uu.se> Uppsala Universitet 14 * 15 * This work is based on the LPC-trie which is originally described in: 16 * 17 * An experimental study of compression methods for dynamic tries 18 * Stefan Nilsson and Matti Tikkanen. Algorithmica, 33(1):19-33, 2002. 19 * http://www.csc.kth.se/~snilsson/software/dyntrie2/ 20 * 21 * 22 * IP-address lookup using LC-tries. Stefan Nilsson and Gunnar Karlsson 23 * IEEE Journal on Selected Areas in Communications, 17(6):1083-1092, June 1999 24 * 25 * 26 * Code from fib_hash has been reused which includes the following header: 27 * 28 * 29 * INET An implementation of the TCP/IP protocol suite for the LINUX 30 * operating system. INET is implemented using the BSD Socket 31 * interface as the means of communication with the user level. 32 * 33 * IPv4 FIB: lookup engine and maintenance routines. 34 * 35 * 36 * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> 37 * 38 * This program is free software; you can redistribute it and/or 39 * modify it under the terms of the GNU General Public License 40 * as published by the Free Software Foundation; either version 41 * 2 of the License, or (at your option) any later version. 42 * 43 * Substantial contributions to this work comes from: 44 * 45 * David S. Miller, <davem@davemloft.net> 46 * Stephen Hemminger <shemminger@osdl.org> 47 * Paul E. McKenney <paulmck@us.ibm.com> 48 * Patrick McHardy <kaber@trash.net> 49 */ 50 51 #define VERSION "0.409" 52 53 #include <asm/uaccess.h> 54 #include <asm/system.h> 55 #include <linux/bitops.h> 56 #include <linux/types.h> 57 #include <linux/kernel.h> 58 #include <linux/mm.h> 59 #include <linux/string.h> 60 #include <linux/socket.h> 61 #include <linux/sockios.h> 62 #include <linux/errno.h> 63 #include <linux/in.h> 64 #include <linux/inet.h> 65 #include <linux/inetdevice.h> 66 #include <linux/netdevice.h> 67 #include <linux/if_arp.h> 68 #include <linux/proc_fs.h> 69 #include <linux/rcupdate.h> 70 #include <linux/skbuff.h> 71 #include <linux/netlink.h> 72 #include <linux/init.h> 73 #include <linux/list.h> 74 #include <linux/slab.h> 75 #include <linux/prefetch.h> 76 #include <linux/export.h> 77 #include <net/net_namespace.h> 78 #include <net/ip.h> 79 #include <net/protocol.h> 80 #include <net/route.h> 81 #include <net/tcp.h> 82 #include <net/sock.h> 83 #include <net/ip_fib.h> 84 #include "fib_lookup.h" 85 86 #define MAX_STAT_DEPTH 32 87 88 #define KEYLENGTH (8*sizeof(t_key)) 89 90 typedef unsigned int t_key; 91 92 #define T_TNODE 0 93 #define T_LEAF 1 94 #define NODE_TYPE_MASK 0x1UL 95 #define NODE_TYPE(node) ((node)->parent & NODE_TYPE_MASK) 96 97 #define IS_TNODE(n) (!(n->parent & T_LEAF)) 98 #define IS_LEAF(n) (n->parent & T_LEAF) 99 100 struct rt_trie_node { 101 unsigned long parent; 102 t_key key; 103 }; 104 105 struct leaf { 106 unsigned long parent; 107 t_key key; 108 struct hlist_head list; 109 struct rcu_head rcu; 110 }; 111 112 struct leaf_info { 113 struct hlist_node hlist; 114 int plen; 115 u32 mask_plen; /* ntohl(inet_make_mask(plen)) */ 116 struct list_head falh; 117 struct rcu_head rcu; 118 }; 119 120 struct tnode { 121 unsigned long parent; 122 t_key key; 123 unsigned char pos; /* 2log(KEYLENGTH) bits needed */ 124 unsigned char bits; /* 2log(KEYLENGTH) bits needed */ 125 unsigned int full_children; /* KEYLENGTH bits needed */ 126 unsigned int empty_children; /* KEYLENGTH bits needed */ 127 union { 128 struct rcu_head rcu; 129 struct work_struct work; 130 struct tnode *tnode_free; 131 }; 132 struct rt_trie_node __rcu *child[0]; 133 }; 134 135 #ifdef CONFIG_IP_FIB_TRIE_STATS 136 struct trie_use_stats { 137 unsigned int gets; 138 unsigned int backtrack; 139 unsigned int semantic_match_passed; 140 unsigned int semantic_match_miss; 141 unsigned int null_node_hit; 142 unsigned int resize_node_skipped; 143 }; 144 #endif 145 146 struct trie_stat { 147 unsigned int totdepth; 148 unsigned int maxdepth; 149 unsigned int tnodes; 150 unsigned int leaves; 151 unsigned int nullpointers; 152 unsigned int prefixes; 153 unsigned int nodesizes[MAX_STAT_DEPTH]; 154 }; 155 156 struct trie { 157 struct rt_trie_node __rcu *trie; 158 #ifdef CONFIG_IP_FIB_TRIE_STATS 159 struct trie_use_stats stats; 160 #endif 161 }; 162 163 static void put_child(struct trie *t, struct tnode *tn, int i, struct rt_trie_node *n); 164 static void tnode_put_child_reorg(struct tnode *tn, int i, struct rt_trie_node *n, 165 int wasfull); 166 static struct rt_trie_node *resize(struct trie *t, struct tnode *tn); 167 static struct tnode *inflate(struct trie *t, struct tnode *tn); 168 static struct tnode *halve(struct trie *t, struct tnode *tn); 169 /* tnodes to free after resize(); protected by RTNL */ 170 static struct tnode *tnode_free_head; 171 static size_t tnode_free_size; 172 173 /* 174 * synchronize_rcu after call_rcu for that many pages; it should be especially 175 * useful before resizing the root node with PREEMPT_NONE configs; the value was 176 * obtained experimentally, aiming to avoid visible slowdown. 177 */ 178 static const int sync_pages = 128; 179 180 static struct kmem_cache *fn_alias_kmem __read_mostly; 181 static struct kmem_cache *trie_leaf_kmem __read_mostly; 182 183 /* 184 * caller must hold RTNL 185 */ 186 static inline struct tnode *node_parent(const struct rt_trie_node *node) 187 { 188 unsigned long parent; 189 190 parent = rcu_dereference_index_check(node->parent, lockdep_rtnl_is_held()); 191 192 return (struct tnode *)(parent & ~NODE_TYPE_MASK); 193 } 194 195 /* 196 * caller must hold RCU read lock or RTNL 197 */ 198 static inline struct tnode *node_parent_rcu(const struct rt_trie_node *node) 199 { 200 unsigned long parent; 201 202 parent = rcu_dereference_index_check(node->parent, rcu_read_lock_held() || 203 lockdep_rtnl_is_held()); 204 205 return (struct tnode *)(parent & ~NODE_TYPE_MASK); 206 } 207 208 /* Same as RCU_INIT_POINTER 209 * but that macro() assumes that value is a pointer. 210 */ 211 static inline void node_set_parent(struct rt_trie_node *node, struct tnode *ptr) 212 { 213 smp_wmb(); 214 node->parent = (unsigned long)ptr | NODE_TYPE(node); 215 } 216 217 /* 218 * caller must hold RTNL 219 */ 220 static inline struct rt_trie_node *tnode_get_child(const struct tnode *tn, unsigned int i) 221 { 222 BUG_ON(i >= 1U << tn->bits); 223 224 return rtnl_dereference(tn->child[i]); 225 } 226 227 /* 228 * caller must hold RCU read lock or RTNL 229 */ 230 static inline struct rt_trie_node *tnode_get_child_rcu(const struct tnode *tn, unsigned int i) 231 { 232 BUG_ON(i >= 1U << tn->bits); 233 234 return rcu_dereference_rtnl(tn->child[i]); 235 } 236 237 static inline int tnode_child_length(const struct tnode *tn) 238 { 239 return 1 << tn->bits; 240 } 241 242 static inline t_key mask_pfx(t_key k, unsigned int l) 243 { 244 return (l == 0) ? 0 : k >> (KEYLENGTH-l) << (KEYLENGTH-l); 245 } 246 247 static inline t_key tkey_extract_bits(t_key a, unsigned int offset, unsigned int bits) 248 { 249 if (offset < KEYLENGTH) 250 return ((t_key)(a << offset)) >> (KEYLENGTH - bits); 251 else 252 return 0; 253 } 254 255 static inline int tkey_equals(t_key a, t_key b) 256 { 257 return a == b; 258 } 259 260 static inline int tkey_sub_equals(t_key a, int offset, int bits, t_key b) 261 { 262 if (bits == 0 || offset >= KEYLENGTH) 263 return 1; 264 bits = bits > KEYLENGTH ? KEYLENGTH : bits; 265 return ((a ^ b) << offset) >> (KEYLENGTH - bits) == 0; 266 } 267 268 static inline int tkey_mismatch(t_key a, int offset, t_key b) 269 { 270 t_key diff = a ^ b; 271 int i = offset; 272 273 if (!diff) 274 return 0; 275 while ((diff << i) >> (KEYLENGTH-1) == 0) 276 i++; 277 return i; 278 } 279 280 /* 281 To understand this stuff, an understanding of keys and all their bits is 282 necessary. Every node in the trie has a key associated with it, but not 283 all of the bits in that key are significant. 284 285 Consider a node 'n' and its parent 'tp'. 286 287 If n is a leaf, every bit in its key is significant. Its presence is 288 necessitated by path compression, since during a tree traversal (when 289 searching for a leaf - unless we are doing an insertion) we will completely 290 ignore all skipped bits we encounter. Thus we need to verify, at the end of 291 a potentially successful search, that we have indeed been walking the 292 correct key path. 293 294 Note that we can never "miss" the correct key in the tree if present by 295 following the wrong path. Path compression ensures that segments of the key 296 that are the same for all keys with a given prefix are skipped, but the 297 skipped part *is* identical for each node in the subtrie below the skipped 298 bit! trie_insert() in this implementation takes care of that - note the 299 call to tkey_sub_equals() in trie_insert(). 300 301 if n is an internal node - a 'tnode' here, the various parts of its key 302 have many different meanings. 303 304 Example: 305 _________________________________________________________________ 306 | i | i | i | i | i | i | i | N | N | N | S | S | S | S | S | C | 307 ----------------------------------------------------------------- 308 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 309 310 _________________________________________________________________ 311 | C | C | C | u | u | u | u | u | u | u | u | u | u | u | u | u | 312 ----------------------------------------------------------------- 313 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 314 315 tp->pos = 7 316 tp->bits = 3 317 n->pos = 15 318 n->bits = 4 319 320 First, let's just ignore the bits that come before the parent tp, that is 321 the bits from 0 to (tp->pos-1). They are *known* but at this point we do 322 not use them for anything. 323 324 The bits from (tp->pos) to (tp->pos + tp->bits - 1) - "N", above - are the 325 index into the parent's child array. That is, they will be used to find 326 'n' among tp's children. 327 328 The bits from (tp->pos + tp->bits) to (n->pos - 1) - "S" - are skipped bits 329 for the node n. 330 331 All the bits we have seen so far are significant to the node n. The rest 332 of the bits are really not needed or indeed known in n->key. 333 334 The bits from (n->pos) to (n->pos + n->bits - 1) - "C" - are the index into 335 n's child array, and will of course be different for each child. 336 337 338 The rest of the bits, from (n->pos + n->bits) onward, are completely unknown 339 at this point. 340 341 */ 342 343 static inline void check_tnode(const struct tnode *tn) 344 { 345 WARN_ON(tn && tn->pos+tn->bits > 32); 346 } 347 348 static const int halve_threshold = 25; 349 static const int inflate_threshold = 50; 350 static const int halve_threshold_root = 15; 351 static const int inflate_threshold_root = 30; 352 353 static void __alias_free_mem(struct rcu_head *head) 354 { 355 struct fib_alias *fa = container_of(head, struct fib_alias, rcu); 356 kmem_cache_free(fn_alias_kmem, fa); 357 } 358 359 static inline void alias_free_mem_rcu(struct fib_alias *fa) 360 { 361 call_rcu(&fa->rcu, __alias_free_mem); 362 } 363 364 static void __leaf_free_rcu(struct rcu_head *head) 365 { 366 struct leaf *l = container_of(head, struct leaf, rcu); 367 kmem_cache_free(trie_leaf_kmem, l); 368 } 369 370 static inline void free_leaf(struct leaf *l) 371 { 372 call_rcu_bh(&l->rcu, __leaf_free_rcu); 373 } 374 375 static inline void free_leaf_info(struct leaf_info *leaf) 376 { 377 kfree_rcu(leaf, rcu); 378 } 379 380 static struct tnode *tnode_alloc(size_t size) 381 { 382 if (size <= PAGE_SIZE) 383 return kzalloc(size, GFP_KERNEL); 384 else 385 return vzalloc(size); 386 } 387 388 static void __tnode_vfree(struct work_struct *arg) 389 { 390 struct tnode *tn = container_of(arg, struct tnode, work); 391 vfree(tn); 392 } 393 394 static void __tnode_free_rcu(struct rcu_head *head) 395 { 396 struct tnode *tn = container_of(head, struct tnode, rcu); 397 size_t size = sizeof(struct tnode) + 398 (sizeof(struct rt_trie_node *) << tn->bits); 399 400 if (size <= PAGE_SIZE) 401 kfree(tn); 402 else { 403 INIT_WORK(&tn->work, __tnode_vfree); 404 schedule_work(&tn->work); 405 } 406 } 407 408 static inline void tnode_free(struct tnode *tn) 409 { 410 if (IS_LEAF(tn)) 411 free_leaf((struct leaf *) tn); 412 else 413 call_rcu(&tn->rcu, __tnode_free_rcu); 414 } 415 416 static void tnode_free_safe(struct tnode *tn) 417 { 418 BUG_ON(IS_LEAF(tn)); 419 tn->tnode_free = tnode_free_head; 420 tnode_free_head = tn; 421 tnode_free_size += sizeof(struct tnode) + 422 (sizeof(struct rt_trie_node *) << tn->bits); 423 } 424 425 static void tnode_free_flush(void) 426 { 427 struct tnode *tn; 428 429 while ((tn = tnode_free_head)) { 430 tnode_free_head = tn->tnode_free; 431 tn->tnode_free = NULL; 432 tnode_free(tn); 433 } 434 435 if (tnode_free_size >= PAGE_SIZE * sync_pages) { 436 tnode_free_size = 0; 437 synchronize_rcu(); 438 } 439 } 440 441 static struct leaf *leaf_new(void) 442 { 443 struct leaf *l = kmem_cache_alloc(trie_leaf_kmem, GFP_KERNEL); 444 if (l) { 445 l->parent = T_LEAF; 446 INIT_HLIST_HEAD(&l->list); 447 } 448 return l; 449 } 450 451 static struct leaf_info *leaf_info_new(int plen) 452 { 453 struct leaf_info *li = kmalloc(sizeof(struct leaf_info), GFP_KERNEL); 454 if (li) { 455 li->plen = plen; 456 li->mask_plen = ntohl(inet_make_mask(plen)); 457 INIT_LIST_HEAD(&li->falh); 458 } 459 return li; 460 } 461 462 static struct tnode *tnode_new(t_key key, int pos, int bits) 463 { 464 size_t sz = sizeof(struct tnode) + (sizeof(struct rt_trie_node *) << bits); 465 struct tnode *tn = tnode_alloc(sz); 466 467 if (tn) { 468 tn->parent = T_TNODE; 469 tn->pos = pos; 470 tn->bits = bits; 471 tn->key = key; 472 tn->full_children = 0; 473 tn->empty_children = 1<<bits; 474 } 475 476 pr_debug("AT %p s=%zu %zu\n", tn, sizeof(struct tnode), 477 sizeof(struct rt_trie_node) << bits); 478 return tn; 479 } 480 481 /* 482 * Check whether a tnode 'n' is "full", i.e. it is an internal node 483 * and no bits are skipped. See discussion in dyntree paper p. 6 484 */ 485 486 static inline int tnode_full(const struct tnode *tn, const struct rt_trie_node *n) 487 { 488 if (n == NULL || IS_LEAF(n)) 489 return 0; 490 491 return ((struct tnode *) n)->pos == tn->pos + tn->bits; 492 } 493 494 static inline void put_child(struct trie *t, struct tnode *tn, int i, 495 struct rt_trie_node *n) 496 { 497 tnode_put_child_reorg(tn, i, n, -1); 498 } 499 500 /* 501 * Add a child at position i overwriting the old value. 502 * Update the value of full_children and empty_children. 503 */ 504 505 static void tnode_put_child_reorg(struct tnode *tn, int i, struct rt_trie_node *n, 506 int wasfull) 507 { 508 struct rt_trie_node *chi = rtnl_dereference(tn->child[i]); 509 int isfull; 510 511 BUG_ON(i >= 1<<tn->bits); 512 513 /* update emptyChildren */ 514 if (n == NULL && chi != NULL) 515 tn->empty_children++; 516 else if (n != NULL && chi == NULL) 517 tn->empty_children--; 518 519 /* update fullChildren */ 520 if (wasfull == -1) 521 wasfull = tnode_full(tn, chi); 522 523 isfull = tnode_full(tn, n); 524 if (wasfull && !isfull) 525 tn->full_children--; 526 else if (!wasfull && isfull) 527 tn->full_children++; 528 529 if (n) 530 node_set_parent(n, tn); 531 532 RCU_INIT_POINTER(tn->child[i], n); 533 } 534 535 #define MAX_WORK 10 536 static struct rt_trie_node *resize(struct trie *t, struct tnode *tn) 537 { 538 int i; 539 struct tnode *old_tn; 540 int inflate_threshold_use; 541 int halve_threshold_use; 542 int max_work; 543 544 if (!tn) 545 return NULL; 546 547 pr_debug("In tnode_resize %p inflate_threshold=%d threshold=%d\n", 548 tn, inflate_threshold, halve_threshold); 549 550 /* No children */ 551 if (tn->empty_children == tnode_child_length(tn)) { 552 tnode_free_safe(tn); 553 return NULL; 554 } 555 /* One child */ 556 if (tn->empty_children == tnode_child_length(tn) - 1) 557 goto one_child; 558 /* 559 * Double as long as the resulting node has a number of 560 * nonempty nodes that are above the threshold. 561 */ 562 563 /* 564 * From "Implementing a dynamic compressed trie" by Stefan Nilsson of 565 * the Helsinki University of Technology and Matti Tikkanen of Nokia 566 * Telecommunications, page 6: 567 * "A node is doubled if the ratio of non-empty children to all 568 * children in the *doubled* node is at least 'high'." 569 * 570 * 'high' in this instance is the variable 'inflate_threshold'. It 571 * is expressed as a percentage, so we multiply it with 572 * tnode_child_length() and instead of multiplying by 2 (since the 573 * child array will be doubled by inflate()) and multiplying 574 * the left-hand side by 100 (to handle the percentage thing) we 575 * multiply the left-hand side by 50. 576 * 577 * The left-hand side may look a bit weird: tnode_child_length(tn) 578 * - tn->empty_children is of course the number of non-null children 579 * in the current node. tn->full_children is the number of "full" 580 * children, that is non-null tnodes with a skip value of 0. 581 * All of those will be doubled in the resulting inflated tnode, so 582 * we just count them one extra time here. 583 * 584 * A clearer way to write this would be: 585 * 586 * to_be_doubled = tn->full_children; 587 * not_to_be_doubled = tnode_child_length(tn) - tn->empty_children - 588 * tn->full_children; 589 * 590 * new_child_length = tnode_child_length(tn) * 2; 591 * 592 * new_fill_factor = 100 * (not_to_be_doubled + 2*to_be_doubled) / 593 * new_child_length; 594 * if (new_fill_factor >= inflate_threshold) 595 * 596 * ...and so on, tho it would mess up the while () loop. 597 * 598 * anyway, 599 * 100 * (not_to_be_doubled + 2*to_be_doubled) / new_child_length >= 600 * inflate_threshold 601 * 602 * avoid a division: 603 * 100 * (not_to_be_doubled + 2*to_be_doubled) >= 604 * inflate_threshold * new_child_length 605 * 606 * expand not_to_be_doubled and to_be_doubled, and shorten: 607 * 100 * (tnode_child_length(tn) - tn->empty_children + 608 * tn->full_children) >= inflate_threshold * new_child_length 609 * 610 * expand new_child_length: 611 * 100 * (tnode_child_length(tn) - tn->empty_children + 612 * tn->full_children) >= 613 * inflate_threshold * tnode_child_length(tn) * 2 614 * 615 * shorten again: 616 * 50 * (tn->full_children + tnode_child_length(tn) - 617 * tn->empty_children) >= inflate_threshold * 618 * tnode_child_length(tn) 619 * 620 */ 621 622 check_tnode(tn); 623 624 /* Keep root node larger */ 625 626 if (!node_parent((struct rt_trie_node *)tn)) { 627 inflate_threshold_use = inflate_threshold_root; 628 halve_threshold_use = halve_threshold_root; 629 } else { 630 inflate_threshold_use = inflate_threshold; 631 halve_threshold_use = halve_threshold; 632 } 633 634 max_work = MAX_WORK; 635 while ((tn->full_children > 0 && max_work-- && 636 50 * (tn->full_children + tnode_child_length(tn) 637 - tn->empty_children) 638 >= inflate_threshold_use * tnode_child_length(tn))) { 639 640 old_tn = tn; 641 tn = inflate(t, tn); 642 643 if (IS_ERR(tn)) { 644 tn = old_tn; 645 #ifdef CONFIG_IP_FIB_TRIE_STATS 646 t->stats.resize_node_skipped++; 647 #endif 648 break; 649 } 650 } 651 652 check_tnode(tn); 653 654 /* Return if at least one inflate is run */ 655 if (max_work != MAX_WORK) 656 return (struct rt_trie_node *) tn; 657 658 /* 659 * Halve as long as the number of empty children in this 660 * node is above threshold. 661 */ 662 663 max_work = MAX_WORK; 664 while (tn->bits > 1 && max_work-- && 665 100 * (tnode_child_length(tn) - tn->empty_children) < 666 halve_threshold_use * tnode_child_length(tn)) { 667 668 old_tn = tn; 669 tn = halve(t, tn); 670 if (IS_ERR(tn)) { 671 tn = old_tn; 672 #ifdef CONFIG_IP_FIB_TRIE_STATS 673 t->stats.resize_node_skipped++; 674 #endif 675 break; 676 } 677 } 678 679 680 /* Only one child remains */ 681 if (tn->empty_children == tnode_child_length(tn) - 1) { 682 one_child: 683 for (i = 0; i < tnode_child_length(tn); i++) { 684 struct rt_trie_node *n; 685 686 n = rtnl_dereference(tn->child[i]); 687 if (!n) 688 continue; 689 690 /* compress one level */ 691 692 node_set_parent(n, NULL); 693 tnode_free_safe(tn); 694 return n; 695 } 696 } 697 return (struct rt_trie_node *) tn; 698 } 699 700 701 static void tnode_clean_free(struct tnode *tn) 702 { 703 int i; 704 struct tnode *tofree; 705 706 for (i = 0; i < tnode_child_length(tn); i++) { 707 tofree = (struct tnode *)rtnl_dereference(tn->child[i]); 708 if (tofree) 709 tnode_free(tofree); 710 } 711 tnode_free(tn); 712 } 713 714 static struct tnode *inflate(struct trie *t, struct tnode *tn) 715 { 716 struct tnode *oldtnode = tn; 717 int olen = tnode_child_length(tn); 718 int i; 719 720 pr_debug("In inflate\n"); 721 722 tn = tnode_new(oldtnode->key, oldtnode->pos, oldtnode->bits + 1); 723 724 if (!tn) 725 return ERR_PTR(-ENOMEM); 726 727 /* 728 * Preallocate and store tnodes before the actual work so we 729 * don't get into an inconsistent state if memory allocation 730 * fails. In case of failure we return the oldnode and inflate 731 * of tnode is ignored. 732 */ 733 734 for (i = 0; i < olen; i++) { 735 struct tnode *inode; 736 737 inode = (struct tnode *) tnode_get_child(oldtnode, i); 738 if (inode && 739 IS_TNODE(inode) && 740 inode->pos == oldtnode->pos + oldtnode->bits && 741 inode->bits > 1) { 742 struct tnode *left, *right; 743 t_key m = ~0U << (KEYLENGTH - 1) >> inode->pos; 744 745 left = tnode_new(inode->key&(~m), inode->pos + 1, 746 inode->bits - 1); 747 if (!left) 748 goto nomem; 749 750 right = tnode_new(inode->key|m, inode->pos + 1, 751 inode->bits - 1); 752 753 if (!right) { 754 tnode_free(left); 755 goto nomem; 756 } 757 758 put_child(t, tn, 2*i, (struct rt_trie_node *) left); 759 put_child(t, tn, 2*i+1, (struct rt_trie_node *) right); 760 } 761 } 762 763 for (i = 0; i < olen; i++) { 764 struct tnode *inode; 765 struct rt_trie_node *node = tnode_get_child(oldtnode, i); 766 struct tnode *left, *right; 767 int size, j; 768 769 /* An empty child */ 770 if (node == NULL) 771 continue; 772 773 /* A leaf or an internal node with skipped bits */ 774 775 if (IS_LEAF(node) || ((struct tnode *) node)->pos > 776 tn->pos + tn->bits - 1) { 777 if (tkey_extract_bits(node->key, 778 oldtnode->pos + oldtnode->bits, 779 1) == 0) 780 put_child(t, tn, 2*i, node); 781 else 782 put_child(t, tn, 2*i+1, node); 783 continue; 784 } 785 786 /* An internal node with two children */ 787 inode = (struct tnode *) node; 788 789 if (inode->bits == 1) { 790 put_child(t, tn, 2*i, rtnl_dereference(inode->child[0])); 791 put_child(t, tn, 2*i+1, rtnl_dereference(inode->child[1])); 792 793 tnode_free_safe(inode); 794 continue; 795 } 796 797 /* An internal node with more than two children */ 798 799 /* We will replace this node 'inode' with two new 800 * ones, 'left' and 'right', each with half of the 801 * original children. The two new nodes will have 802 * a position one bit further down the key and this 803 * means that the "significant" part of their keys 804 * (see the discussion near the top of this file) 805 * will differ by one bit, which will be "0" in 806 * left's key and "1" in right's key. Since we are 807 * moving the key position by one step, the bit that 808 * we are moving away from - the bit at position 809 * (inode->pos) - is the one that will differ between 810 * left and right. So... we synthesize that bit in the 811 * two new keys. 812 * The mask 'm' below will be a single "one" bit at 813 * the position (inode->pos) 814 */ 815 816 /* Use the old key, but set the new significant 817 * bit to zero. 818 */ 819 820 left = (struct tnode *) tnode_get_child(tn, 2*i); 821 put_child(t, tn, 2*i, NULL); 822 823 BUG_ON(!left); 824 825 right = (struct tnode *) tnode_get_child(tn, 2*i+1); 826 put_child(t, tn, 2*i+1, NULL); 827 828 BUG_ON(!right); 829 830 size = tnode_child_length(left); 831 for (j = 0; j < size; j++) { 832 put_child(t, left, j, rtnl_dereference(inode->child[j])); 833 put_child(t, right, j, rtnl_dereference(inode->child[j + size])); 834 } 835 put_child(t, tn, 2*i, resize(t, left)); 836 put_child(t, tn, 2*i+1, resize(t, right)); 837 838 tnode_free_safe(inode); 839 } 840 tnode_free_safe(oldtnode); 841 return tn; 842 nomem: 843 tnode_clean_free(tn); 844 return ERR_PTR(-ENOMEM); 845 } 846 847 static struct tnode *halve(struct trie *t, struct tnode *tn) 848 { 849 struct tnode *oldtnode = tn; 850 struct rt_trie_node *left, *right; 851 int i; 852 int olen = tnode_child_length(tn); 853 854 pr_debug("In halve\n"); 855 856 tn = tnode_new(oldtnode->key, oldtnode->pos, oldtnode->bits - 1); 857 858 if (!tn) 859 return ERR_PTR(-ENOMEM); 860 861 /* 862 * Preallocate and store tnodes before the actual work so we 863 * don't get into an inconsistent state if memory allocation 864 * fails. In case of failure we return the oldnode and halve 865 * of tnode is ignored. 866 */ 867 868 for (i = 0; i < olen; i += 2) { 869 left = tnode_get_child(oldtnode, i); 870 right = tnode_get_child(oldtnode, i+1); 871 872 /* Two nonempty children */ 873 if (left && right) { 874 struct tnode *newn; 875 876 newn = tnode_new(left->key, tn->pos + tn->bits, 1); 877 878 if (!newn) 879 goto nomem; 880 881 put_child(t, tn, i/2, (struct rt_trie_node *)newn); 882 } 883 884 } 885 886 for (i = 0; i < olen; i += 2) { 887 struct tnode *newBinNode; 888 889 left = tnode_get_child(oldtnode, i); 890 right = tnode_get_child(oldtnode, i+1); 891 892 /* At least one of the children is empty */ 893 if (left == NULL) { 894 if (right == NULL) /* Both are empty */ 895 continue; 896 put_child(t, tn, i/2, right); 897 continue; 898 } 899 900 if (right == NULL) { 901 put_child(t, tn, i/2, left); 902 continue; 903 } 904 905 /* Two nonempty children */ 906 newBinNode = (struct tnode *) tnode_get_child(tn, i/2); 907 put_child(t, tn, i/2, NULL); 908 put_child(t, newBinNode, 0, left); 909 put_child(t, newBinNode, 1, right); 910 put_child(t, tn, i/2, resize(t, newBinNode)); 911 } 912 tnode_free_safe(oldtnode); 913 return tn; 914 nomem: 915 tnode_clean_free(tn); 916 return ERR_PTR(-ENOMEM); 917 } 918 919 /* readside must use rcu_read_lock currently dump routines 920 via get_fa_head and dump */ 921 922 static struct leaf_info *find_leaf_info(struct leaf *l, int plen) 923 { 924 struct hlist_head *head = &l->list; 925 struct hlist_node *node; 926 struct leaf_info *li; 927 928 hlist_for_each_entry_rcu(li, node, head, hlist) 929 if (li->plen == plen) 930 return li; 931 932 return NULL; 933 } 934 935 static inline struct list_head *get_fa_head(struct leaf *l, int plen) 936 { 937 struct leaf_info *li = find_leaf_info(l, plen); 938 939 if (!li) 940 return NULL; 941 942 return &li->falh; 943 } 944 945 static void insert_leaf_info(struct hlist_head *head, struct leaf_info *new) 946 { 947 struct leaf_info *li = NULL, *last = NULL; 948 struct hlist_node *node; 949 950 if (hlist_empty(head)) { 951 hlist_add_head_rcu(&new->hlist, head); 952 } else { 953 hlist_for_each_entry(li, node, head, hlist) { 954 if (new->plen > li->plen) 955 break; 956 957 last = li; 958 } 959 if (last) 960 hlist_add_after_rcu(&last->hlist, &new->hlist); 961 else 962 hlist_add_before_rcu(&new->hlist, &li->hlist); 963 } 964 } 965 966 /* rcu_read_lock needs to be hold by caller from readside */ 967 968 static struct leaf * 969 fib_find_node(struct trie *t, u32 key) 970 { 971 int pos; 972 struct tnode *tn; 973 struct rt_trie_node *n; 974 975 pos = 0; 976 n = rcu_dereference_rtnl(t->trie); 977 978 while (n != NULL && NODE_TYPE(n) == T_TNODE) { 979 tn = (struct tnode *) n; 980 981 check_tnode(tn); 982 983 if (tkey_sub_equals(tn->key, pos, tn->pos-pos, key)) { 984 pos = tn->pos + tn->bits; 985 n = tnode_get_child_rcu(tn, 986 tkey_extract_bits(key, 987 tn->pos, 988 tn->bits)); 989 } else 990 break; 991 } 992 /* Case we have found a leaf. Compare prefixes */ 993 994 if (n != NULL && IS_LEAF(n) && tkey_equals(key, n->key)) 995 return (struct leaf *)n; 996 997 return NULL; 998 } 999 1000 static void trie_rebalance(struct trie *t, struct tnode *tn) 1001 { 1002 int wasfull; 1003 t_key cindex, key; 1004 struct tnode *tp; 1005 1006 key = tn->key; 1007 1008 while (tn != NULL && (tp = node_parent((struct rt_trie_node *)tn)) != NULL) { 1009 cindex = tkey_extract_bits(key, tp->pos, tp->bits); 1010 wasfull = tnode_full(tp, tnode_get_child(tp, cindex)); 1011 tn = (struct tnode *) resize(t, (struct tnode *)tn); 1012 1013 tnode_put_child_reorg((struct tnode *)tp, cindex, 1014 (struct rt_trie_node *)tn, wasfull); 1015 1016 tp = node_parent((struct rt_trie_node *) tn); 1017 if (!tp) 1018 RCU_INIT_POINTER(t->trie, (struct rt_trie_node *)tn); 1019 1020 tnode_free_flush(); 1021 if (!tp) 1022 break; 1023 tn = tp; 1024 } 1025 1026 /* Handle last (top) tnode */ 1027 if (IS_TNODE(tn)) 1028 tn = (struct tnode *)resize(t, (struct tnode *)tn); 1029 1030 RCU_INIT_POINTER(t->trie, (struct rt_trie_node *)tn); 1031 tnode_free_flush(); 1032 } 1033 1034 /* only used from updater-side */ 1035 1036 static struct list_head *fib_insert_node(struct trie *t, u32 key, int plen) 1037 { 1038 int pos, newpos; 1039 struct tnode *tp = NULL, *tn = NULL; 1040 struct rt_trie_node *n; 1041 struct leaf *l; 1042 int missbit; 1043 struct list_head *fa_head = NULL; 1044 struct leaf_info *li; 1045 t_key cindex; 1046 1047 pos = 0; 1048 n = rtnl_dereference(t->trie); 1049 1050 /* If we point to NULL, stop. Either the tree is empty and we should 1051 * just put a new leaf in if, or we have reached an empty child slot, 1052 * and we should just put our new leaf in that. 1053 * If we point to a T_TNODE, check if it matches our key. Note that 1054 * a T_TNODE might be skipping any number of bits - its 'pos' need 1055 * not be the parent's 'pos'+'bits'! 1056 * 1057 * If it does match the current key, get pos/bits from it, extract 1058 * the index from our key, push the T_TNODE and walk the tree. 1059 * 1060 * If it doesn't, we have to replace it with a new T_TNODE. 1061 * 1062 * If we point to a T_LEAF, it might or might not have the same key 1063 * as we do. If it does, just change the value, update the T_LEAF's 1064 * value, and return it. 1065 * If it doesn't, we need to replace it with a T_TNODE. 1066 */ 1067 1068 while (n != NULL && NODE_TYPE(n) == T_TNODE) { 1069 tn = (struct tnode *) n; 1070 1071 check_tnode(tn); 1072 1073 if (tkey_sub_equals(tn->key, pos, tn->pos-pos, key)) { 1074 tp = tn; 1075 pos = tn->pos + tn->bits; 1076 n = tnode_get_child(tn, 1077 tkey_extract_bits(key, 1078 tn->pos, 1079 tn->bits)); 1080 1081 BUG_ON(n && node_parent(n) != tn); 1082 } else 1083 break; 1084 } 1085 1086 /* 1087 * n ----> NULL, LEAF or TNODE 1088 * 1089 * tp is n's (parent) ----> NULL or TNODE 1090 */ 1091 1092 BUG_ON(tp && IS_LEAF(tp)); 1093 1094 /* Case 1: n is a leaf. Compare prefixes */ 1095 1096 if (n != NULL && IS_LEAF(n) && tkey_equals(key, n->key)) { 1097 l = (struct leaf *) n; 1098 li = leaf_info_new(plen); 1099 1100 if (!li) 1101 return NULL; 1102 1103 fa_head = &li->falh; 1104 insert_leaf_info(&l->list, li); 1105 goto done; 1106 } 1107 l = leaf_new(); 1108 1109 if (!l) 1110 return NULL; 1111 1112 l->key = key; 1113 li = leaf_info_new(plen); 1114 1115 if (!li) { 1116 free_leaf(l); 1117 return NULL; 1118 } 1119 1120 fa_head = &li->falh; 1121 insert_leaf_info(&l->list, li); 1122 1123 if (t->trie && n == NULL) { 1124 /* Case 2: n is NULL, and will just insert a new leaf */ 1125 1126 node_set_parent((struct rt_trie_node *)l, tp); 1127 1128 cindex = tkey_extract_bits(key, tp->pos, tp->bits); 1129 put_child(t, (struct tnode *)tp, cindex, (struct rt_trie_node *)l); 1130 } else { 1131 /* Case 3: n is a LEAF or a TNODE and the key doesn't match. */ 1132 /* 1133 * Add a new tnode here 1134 * first tnode need some special handling 1135 */ 1136 1137 if (tp) 1138 pos = tp->pos+tp->bits; 1139 else 1140 pos = 0; 1141 1142 if (n) { 1143 newpos = tkey_mismatch(key, pos, n->key); 1144 tn = tnode_new(n->key, newpos, 1); 1145 } else { 1146 newpos = 0; 1147 tn = tnode_new(key, newpos, 1); /* First tnode */ 1148 } 1149 1150 if (!tn) { 1151 free_leaf_info(li); 1152 free_leaf(l); 1153 return NULL; 1154 } 1155 1156 node_set_parent((struct rt_trie_node *)tn, tp); 1157 1158 missbit = tkey_extract_bits(key, newpos, 1); 1159 put_child(t, tn, missbit, (struct rt_trie_node *)l); 1160 put_child(t, tn, 1-missbit, n); 1161 1162 if (tp) { 1163 cindex = tkey_extract_bits(key, tp->pos, tp->bits); 1164 put_child(t, (struct tnode *)tp, cindex, 1165 (struct rt_trie_node *)tn); 1166 } else { 1167 RCU_INIT_POINTER(t->trie, (struct rt_trie_node *)tn); 1168 tp = tn; 1169 } 1170 } 1171 1172 if (tp && tp->pos + tp->bits > 32) 1173 pr_warning("fib_trie" 1174 " tp=%p pos=%d, bits=%d, key=%0x plen=%d\n", 1175 tp, tp->pos, tp->bits, key, plen); 1176 1177 /* Rebalance the trie */ 1178 1179 trie_rebalance(t, tp); 1180 done: 1181 return fa_head; 1182 } 1183 1184 /* 1185 * Caller must hold RTNL. 1186 */ 1187 int fib_table_insert(struct fib_table *tb, struct fib_config *cfg) 1188 { 1189 struct trie *t = (struct trie *) tb->tb_data; 1190 struct fib_alias *fa, *new_fa; 1191 struct list_head *fa_head = NULL; 1192 struct fib_info *fi; 1193 int plen = cfg->fc_dst_len; 1194 u8 tos = cfg->fc_tos; 1195 u32 key, mask; 1196 int err; 1197 struct leaf *l; 1198 1199 if (plen > 32) 1200 return -EINVAL; 1201 1202 key = ntohl(cfg->fc_dst); 1203 1204 pr_debug("Insert table=%u %08x/%d\n", tb->tb_id, key, plen); 1205 1206 mask = ntohl(inet_make_mask(plen)); 1207 1208 if (key & ~mask) 1209 return -EINVAL; 1210 1211 key = key & mask; 1212 1213 fi = fib_create_info(cfg); 1214 if (IS_ERR(fi)) { 1215 err = PTR_ERR(fi); 1216 goto err; 1217 } 1218 1219 l = fib_find_node(t, key); 1220 fa = NULL; 1221 1222 if (l) { 1223 fa_head = get_fa_head(l, plen); 1224 fa = fib_find_alias(fa_head, tos, fi->fib_priority); 1225 } 1226 1227 /* Now fa, if non-NULL, points to the first fib alias 1228 * with the same keys [prefix,tos,priority], if such key already 1229 * exists or to the node before which we will insert new one. 1230 * 1231 * If fa is NULL, we will need to allocate a new one and 1232 * insert to the head of f. 1233 * 1234 * If f is NULL, no fib node matched the destination key 1235 * and we need to allocate a new one of those as well. 1236 */ 1237 1238 if (fa && fa->fa_tos == tos && 1239 fa->fa_info->fib_priority == fi->fib_priority) { 1240 struct fib_alias *fa_first, *fa_match; 1241 1242 err = -EEXIST; 1243 if (cfg->fc_nlflags & NLM_F_EXCL) 1244 goto out; 1245 1246 /* We have 2 goals: 1247 * 1. Find exact match for type, scope, fib_info to avoid 1248 * duplicate routes 1249 * 2. Find next 'fa' (or head), NLM_F_APPEND inserts before it 1250 */ 1251 fa_match = NULL; 1252 fa_first = fa; 1253 fa = list_entry(fa->fa_list.prev, struct fib_alias, fa_list); 1254 list_for_each_entry_continue(fa, fa_head, fa_list) { 1255 if (fa->fa_tos != tos) 1256 break; 1257 if (fa->fa_info->fib_priority != fi->fib_priority) 1258 break; 1259 if (fa->fa_type == cfg->fc_type && 1260 fa->fa_info == fi) { 1261 fa_match = fa; 1262 break; 1263 } 1264 } 1265 1266 if (cfg->fc_nlflags & NLM_F_REPLACE) { 1267 struct fib_info *fi_drop; 1268 u8 state; 1269 1270 fa = fa_first; 1271 if (fa_match) { 1272 if (fa == fa_match) 1273 err = 0; 1274 goto out; 1275 } 1276 err = -ENOBUFS; 1277 new_fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL); 1278 if (new_fa == NULL) 1279 goto out; 1280 1281 fi_drop = fa->fa_info; 1282 new_fa->fa_tos = fa->fa_tos; 1283 new_fa->fa_info = fi; 1284 new_fa->fa_type = cfg->fc_type; 1285 state = fa->fa_state; 1286 new_fa->fa_state = state & ~FA_S_ACCESSED; 1287 1288 list_replace_rcu(&fa->fa_list, &new_fa->fa_list); 1289 alias_free_mem_rcu(fa); 1290 1291 fib_release_info(fi_drop); 1292 if (state & FA_S_ACCESSED) 1293 rt_cache_flush(cfg->fc_nlinfo.nl_net, -1); 1294 rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen, 1295 tb->tb_id, &cfg->fc_nlinfo, NLM_F_REPLACE); 1296 1297 goto succeeded; 1298 } 1299 /* Error if we find a perfect match which 1300 * uses the same scope, type, and nexthop 1301 * information. 1302 */ 1303 if (fa_match) 1304 goto out; 1305 1306 if (!(cfg->fc_nlflags & NLM_F_APPEND)) 1307 fa = fa_first; 1308 } 1309 err = -ENOENT; 1310 if (!(cfg->fc_nlflags & NLM_F_CREATE)) 1311 goto out; 1312 1313 err = -ENOBUFS; 1314 new_fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL); 1315 if (new_fa == NULL) 1316 goto out; 1317 1318 new_fa->fa_info = fi; 1319 new_fa->fa_tos = tos; 1320 new_fa->fa_type = cfg->fc_type; 1321 new_fa->fa_state = 0; 1322 /* 1323 * Insert new entry to the list. 1324 */ 1325 1326 if (!fa_head) { 1327 fa_head = fib_insert_node(t, key, plen); 1328 if (unlikely(!fa_head)) { 1329 err = -ENOMEM; 1330 goto out_free_new_fa; 1331 } 1332 } 1333 1334 if (!plen) 1335 tb->tb_num_default++; 1336 1337 list_add_tail_rcu(&new_fa->fa_list, 1338 (fa ? &fa->fa_list : fa_head)); 1339 1340 rt_cache_flush(cfg->fc_nlinfo.nl_net, -1); 1341 rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen, tb->tb_id, 1342 &cfg->fc_nlinfo, 0); 1343 succeeded: 1344 return 0; 1345 1346 out_free_new_fa: 1347 kmem_cache_free(fn_alias_kmem, new_fa); 1348 out: 1349 fib_release_info(fi); 1350 err: 1351 return err; 1352 } 1353 1354 /* should be called with rcu_read_lock */ 1355 static int check_leaf(struct fib_table *tb, struct trie *t, struct leaf *l, 1356 t_key key, const struct flowi4 *flp, 1357 struct fib_result *res, int fib_flags) 1358 { 1359 struct leaf_info *li; 1360 struct hlist_head *hhead = &l->list; 1361 struct hlist_node *node; 1362 1363 hlist_for_each_entry_rcu(li, node, hhead, hlist) { 1364 struct fib_alias *fa; 1365 1366 if (l->key != (key & li->mask_plen)) 1367 continue; 1368 1369 list_for_each_entry_rcu(fa, &li->falh, fa_list) { 1370 struct fib_info *fi = fa->fa_info; 1371 int nhsel, err; 1372 1373 if (fa->fa_tos && fa->fa_tos != flp->flowi4_tos) 1374 continue; 1375 if (fa->fa_info->fib_scope < flp->flowi4_scope) 1376 continue; 1377 fib_alias_accessed(fa); 1378 err = fib_props[fa->fa_type].error; 1379 if (err) { 1380 #ifdef CONFIG_IP_FIB_TRIE_STATS 1381 t->stats.semantic_match_passed++; 1382 #endif 1383 return err; 1384 } 1385 if (fi->fib_flags & RTNH_F_DEAD) 1386 continue; 1387 for (nhsel = 0; nhsel < fi->fib_nhs; nhsel++) { 1388 const struct fib_nh *nh = &fi->fib_nh[nhsel]; 1389 1390 if (nh->nh_flags & RTNH_F_DEAD) 1391 continue; 1392 if (flp->flowi4_oif && flp->flowi4_oif != nh->nh_oif) 1393 continue; 1394 1395 #ifdef CONFIG_IP_FIB_TRIE_STATS 1396 t->stats.semantic_match_passed++; 1397 #endif 1398 res->prefixlen = li->plen; 1399 res->nh_sel = nhsel; 1400 res->type = fa->fa_type; 1401 res->scope = fa->fa_info->fib_scope; 1402 res->fi = fi; 1403 res->table = tb; 1404 res->fa_head = &li->falh; 1405 if (!(fib_flags & FIB_LOOKUP_NOREF)) 1406 atomic_inc(&fi->fib_clntref); 1407 return 0; 1408 } 1409 } 1410 1411 #ifdef CONFIG_IP_FIB_TRIE_STATS 1412 t->stats.semantic_match_miss++; 1413 #endif 1414 } 1415 1416 return 1; 1417 } 1418 1419 int fib_table_lookup(struct fib_table *tb, const struct flowi4 *flp, 1420 struct fib_result *res, int fib_flags) 1421 { 1422 struct trie *t = (struct trie *) tb->tb_data; 1423 int ret; 1424 struct rt_trie_node *n; 1425 struct tnode *pn; 1426 unsigned int pos, bits; 1427 t_key key = ntohl(flp->daddr); 1428 unsigned int chopped_off; 1429 t_key cindex = 0; 1430 unsigned int current_prefix_length = KEYLENGTH; 1431 struct tnode *cn; 1432 t_key pref_mismatch; 1433 1434 rcu_read_lock(); 1435 1436 n = rcu_dereference(t->trie); 1437 if (!n) 1438 goto failed; 1439 1440 #ifdef CONFIG_IP_FIB_TRIE_STATS 1441 t->stats.gets++; 1442 #endif 1443 1444 /* Just a leaf? */ 1445 if (IS_LEAF(n)) { 1446 ret = check_leaf(tb, t, (struct leaf *)n, key, flp, res, fib_flags); 1447 goto found; 1448 } 1449 1450 pn = (struct tnode *) n; 1451 chopped_off = 0; 1452 1453 while (pn) { 1454 pos = pn->pos; 1455 bits = pn->bits; 1456 1457 if (!chopped_off) 1458 cindex = tkey_extract_bits(mask_pfx(key, current_prefix_length), 1459 pos, bits); 1460 1461 n = tnode_get_child_rcu(pn, cindex); 1462 1463 if (n == NULL) { 1464 #ifdef CONFIG_IP_FIB_TRIE_STATS 1465 t->stats.null_node_hit++; 1466 #endif 1467 goto backtrace; 1468 } 1469 1470 if (IS_LEAF(n)) { 1471 ret = check_leaf(tb, t, (struct leaf *)n, key, flp, res, fib_flags); 1472 if (ret > 0) 1473 goto backtrace; 1474 goto found; 1475 } 1476 1477 cn = (struct tnode *)n; 1478 1479 /* 1480 * It's a tnode, and we can do some extra checks here if we 1481 * like, to avoid descending into a dead-end branch. 1482 * This tnode is in the parent's child array at index 1483 * key[p_pos..p_pos+p_bits] but potentially with some bits 1484 * chopped off, so in reality the index may be just a 1485 * subprefix, padded with zero at the end. 1486 * We can also take a look at any skipped bits in this 1487 * tnode - everything up to p_pos is supposed to be ok, 1488 * and the non-chopped bits of the index (se previous 1489 * paragraph) are also guaranteed ok, but the rest is 1490 * considered unknown. 1491 * 1492 * The skipped bits are key[pos+bits..cn->pos]. 1493 */ 1494 1495 /* If current_prefix_length < pos+bits, we are already doing 1496 * actual prefix matching, which means everything from 1497 * pos+(bits-chopped_off) onward must be zero along some 1498 * branch of this subtree - otherwise there is *no* valid 1499 * prefix present. Here we can only check the skipped 1500 * bits. Remember, since we have already indexed into the 1501 * parent's child array, we know that the bits we chopped of 1502 * *are* zero. 1503 */ 1504 1505 /* NOTA BENE: Checking only skipped bits 1506 for the new node here */ 1507 1508 if (current_prefix_length < pos+bits) { 1509 if (tkey_extract_bits(cn->key, current_prefix_length, 1510 cn->pos - current_prefix_length) 1511 || !(cn->child[0])) 1512 goto backtrace; 1513 } 1514 1515 /* 1516 * If chopped_off=0, the index is fully validated and we 1517 * only need to look at the skipped bits for this, the new, 1518 * tnode. What we actually want to do is to find out if 1519 * these skipped bits match our key perfectly, or if we will 1520 * have to count on finding a matching prefix further down, 1521 * because if we do, we would like to have some way of 1522 * verifying the existence of such a prefix at this point. 1523 */ 1524 1525 /* The only thing we can do at this point is to verify that 1526 * any such matching prefix can indeed be a prefix to our 1527 * key, and if the bits in the node we are inspecting that 1528 * do not match our key are not ZERO, this cannot be true. 1529 * Thus, find out where there is a mismatch (before cn->pos) 1530 * and verify that all the mismatching bits are zero in the 1531 * new tnode's key. 1532 */ 1533 1534 /* 1535 * Note: We aren't very concerned about the piece of 1536 * the key that precede pn->pos+pn->bits, since these 1537 * have already been checked. The bits after cn->pos 1538 * aren't checked since these are by definition 1539 * "unknown" at this point. Thus, what we want to see 1540 * is if we are about to enter the "prefix matching" 1541 * state, and in that case verify that the skipped 1542 * bits that will prevail throughout this subtree are 1543 * zero, as they have to be if we are to find a 1544 * matching prefix. 1545 */ 1546 1547 pref_mismatch = mask_pfx(cn->key ^ key, cn->pos); 1548 1549 /* 1550 * In short: If skipped bits in this node do not match 1551 * the search key, enter the "prefix matching" 1552 * state.directly. 1553 */ 1554 if (pref_mismatch) { 1555 int mp = KEYLENGTH - fls(pref_mismatch); 1556 1557 if (tkey_extract_bits(cn->key, mp, cn->pos - mp) != 0) 1558 goto backtrace; 1559 1560 if (current_prefix_length >= cn->pos) 1561 current_prefix_length = mp; 1562 } 1563 1564 pn = (struct tnode *)n; /* Descend */ 1565 chopped_off = 0; 1566 continue; 1567 1568 backtrace: 1569 chopped_off++; 1570 1571 /* As zero don't change the child key (cindex) */ 1572 while ((chopped_off <= pn->bits) 1573 && !(cindex & (1<<(chopped_off-1)))) 1574 chopped_off++; 1575 1576 /* Decrease current_... with bits chopped off */ 1577 if (current_prefix_length > pn->pos + pn->bits - chopped_off) 1578 current_prefix_length = pn->pos + pn->bits 1579 - chopped_off; 1580 1581 /* 1582 * Either we do the actual chop off according or if we have 1583 * chopped off all bits in this tnode walk up to our parent. 1584 */ 1585 1586 if (chopped_off <= pn->bits) { 1587 cindex &= ~(1 << (chopped_off-1)); 1588 } else { 1589 struct tnode *parent = node_parent_rcu((struct rt_trie_node *) pn); 1590 if (!parent) 1591 goto failed; 1592 1593 /* Get Child's index */ 1594 cindex = tkey_extract_bits(pn->key, parent->pos, parent->bits); 1595 pn = parent; 1596 chopped_off = 0; 1597 1598 #ifdef CONFIG_IP_FIB_TRIE_STATS 1599 t->stats.backtrack++; 1600 #endif 1601 goto backtrace; 1602 } 1603 } 1604 failed: 1605 ret = 1; 1606 found: 1607 rcu_read_unlock(); 1608 return ret; 1609 } 1610 1611 /* 1612 * Remove the leaf and return parent. 1613 */ 1614 static void trie_leaf_remove(struct trie *t, struct leaf *l) 1615 { 1616 struct tnode *tp = node_parent((struct rt_trie_node *) l); 1617 1618 pr_debug("entering trie_leaf_remove(%p)\n", l); 1619 1620 if (tp) { 1621 t_key cindex = tkey_extract_bits(l->key, tp->pos, tp->bits); 1622 put_child(t, (struct tnode *)tp, cindex, NULL); 1623 trie_rebalance(t, tp); 1624 } else 1625 RCU_INIT_POINTER(t->trie, NULL); 1626 1627 free_leaf(l); 1628 } 1629 1630 /* 1631 * Caller must hold RTNL. 1632 */ 1633 int fib_table_delete(struct fib_table *tb, struct fib_config *cfg) 1634 { 1635 struct trie *t = (struct trie *) tb->tb_data; 1636 u32 key, mask; 1637 int plen = cfg->fc_dst_len; 1638 u8 tos = cfg->fc_tos; 1639 struct fib_alias *fa, *fa_to_delete; 1640 struct list_head *fa_head; 1641 struct leaf *l; 1642 struct leaf_info *li; 1643 1644 if (plen > 32) 1645 return -EINVAL; 1646 1647 key = ntohl(cfg->fc_dst); 1648 mask = ntohl(inet_make_mask(plen)); 1649 1650 if (key & ~mask) 1651 return -EINVAL; 1652 1653 key = key & mask; 1654 l = fib_find_node(t, key); 1655 1656 if (!l) 1657 return -ESRCH; 1658 1659 fa_head = get_fa_head(l, plen); 1660 fa = fib_find_alias(fa_head, tos, 0); 1661 1662 if (!fa) 1663 return -ESRCH; 1664 1665 pr_debug("Deleting %08x/%d tos=%d t=%p\n", key, plen, tos, t); 1666 1667 fa_to_delete = NULL; 1668 fa = list_entry(fa->fa_list.prev, struct fib_alias, fa_list); 1669 list_for_each_entry_continue(fa, fa_head, fa_list) { 1670 struct fib_info *fi = fa->fa_info; 1671 1672 if (fa->fa_tos != tos) 1673 break; 1674 1675 if ((!cfg->fc_type || fa->fa_type == cfg->fc_type) && 1676 (cfg->fc_scope == RT_SCOPE_NOWHERE || 1677 fa->fa_info->fib_scope == cfg->fc_scope) && 1678 (!cfg->fc_prefsrc || 1679 fi->fib_prefsrc == cfg->fc_prefsrc) && 1680 (!cfg->fc_protocol || 1681 fi->fib_protocol == cfg->fc_protocol) && 1682 fib_nh_match(cfg, fi) == 0) { 1683 fa_to_delete = fa; 1684 break; 1685 } 1686 } 1687 1688 if (!fa_to_delete) 1689 return -ESRCH; 1690 1691 fa = fa_to_delete; 1692 rtmsg_fib(RTM_DELROUTE, htonl(key), fa, plen, tb->tb_id, 1693 &cfg->fc_nlinfo, 0); 1694 1695 l = fib_find_node(t, key); 1696 li = find_leaf_info(l, plen); 1697 1698 list_del_rcu(&fa->fa_list); 1699 1700 if (!plen) 1701 tb->tb_num_default--; 1702 1703 if (list_empty(fa_head)) { 1704 hlist_del_rcu(&li->hlist); 1705 free_leaf_info(li); 1706 } 1707 1708 if (hlist_empty(&l->list)) 1709 trie_leaf_remove(t, l); 1710 1711 if (fa->fa_state & FA_S_ACCESSED) 1712 rt_cache_flush(cfg->fc_nlinfo.nl_net, -1); 1713 1714 fib_release_info(fa->fa_info); 1715 alias_free_mem_rcu(fa); 1716 return 0; 1717 } 1718 1719 static int trie_flush_list(struct list_head *head) 1720 { 1721 struct fib_alias *fa, *fa_node; 1722 int found = 0; 1723 1724 list_for_each_entry_safe(fa, fa_node, head, fa_list) { 1725 struct fib_info *fi = fa->fa_info; 1726 1727 if (fi && (fi->fib_flags & RTNH_F_DEAD)) { 1728 list_del_rcu(&fa->fa_list); 1729 fib_release_info(fa->fa_info); 1730 alias_free_mem_rcu(fa); 1731 found++; 1732 } 1733 } 1734 return found; 1735 } 1736 1737 static int trie_flush_leaf(struct leaf *l) 1738 { 1739 int found = 0; 1740 struct hlist_head *lih = &l->list; 1741 struct hlist_node *node, *tmp; 1742 struct leaf_info *li = NULL; 1743 1744 hlist_for_each_entry_safe(li, node, tmp, lih, hlist) { 1745 found += trie_flush_list(&li->falh); 1746 1747 if (list_empty(&li->falh)) { 1748 hlist_del_rcu(&li->hlist); 1749 free_leaf_info(li); 1750 } 1751 } 1752 return found; 1753 } 1754 1755 /* 1756 * Scan for the next right leaf starting at node p->child[idx] 1757 * Since we have back pointer, no recursion necessary. 1758 */ 1759 static struct leaf *leaf_walk_rcu(struct tnode *p, struct rt_trie_node *c) 1760 { 1761 do { 1762 t_key idx; 1763 1764 if (c) 1765 idx = tkey_extract_bits(c->key, p->pos, p->bits) + 1; 1766 else 1767 idx = 0; 1768 1769 while (idx < 1u << p->bits) { 1770 c = tnode_get_child_rcu(p, idx++); 1771 if (!c) 1772 continue; 1773 1774 if (IS_LEAF(c)) { 1775 prefetch(rcu_dereference_rtnl(p->child[idx])); 1776 return (struct leaf *) c; 1777 } 1778 1779 /* Rescan start scanning in new node */ 1780 p = (struct tnode *) c; 1781 idx = 0; 1782 } 1783 1784 /* Node empty, walk back up to parent */ 1785 c = (struct rt_trie_node *) p; 1786 } while ((p = node_parent_rcu(c)) != NULL); 1787 1788 return NULL; /* Root of trie */ 1789 } 1790 1791 static struct leaf *trie_firstleaf(struct trie *t) 1792 { 1793 struct tnode *n = (struct tnode *)rcu_dereference_rtnl(t->trie); 1794 1795 if (!n) 1796 return NULL; 1797 1798 if (IS_LEAF(n)) /* trie is just a leaf */ 1799 return (struct leaf *) n; 1800 1801 return leaf_walk_rcu(n, NULL); 1802 } 1803 1804 static struct leaf *trie_nextleaf(struct leaf *l) 1805 { 1806 struct rt_trie_node *c = (struct rt_trie_node *) l; 1807 struct tnode *p = node_parent_rcu(c); 1808 1809 if (!p) 1810 return NULL; /* trie with just one leaf */ 1811 1812 return leaf_walk_rcu(p, c); 1813 } 1814 1815 static struct leaf *trie_leafindex(struct trie *t, int index) 1816 { 1817 struct leaf *l = trie_firstleaf(t); 1818 1819 while (l && index-- > 0) 1820 l = trie_nextleaf(l); 1821 1822 return l; 1823 } 1824 1825 1826 /* 1827 * Caller must hold RTNL. 1828 */ 1829 int fib_table_flush(struct fib_table *tb) 1830 { 1831 struct trie *t = (struct trie *) tb->tb_data; 1832 struct leaf *l, *ll = NULL; 1833 int found = 0; 1834 1835 for (l = trie_firstleaf(t); l; l = trie_nextleaf(l)) { 1836 found += trie_flush_leaf(l); 1837 1838 if (ll && hlist_empty(&ll->list)) 1839 trie_leaf_remove(t, ll); 1840 ll = l; 1841 } 1842 1843 if (ll && hlist_empty(&ll->list)) 1844 trie_leaf_remove(t, ll); 1845 1846 pr_debug("trie_flush found=%d\n", found); 1847 return found; 1848 } 1849 1850 void fib_free_table(struct fib_table *tb) 1851 { 1852 kfree(tb); 1853 } 1854 1855 static int fn_trie_dump_fa(t_key key, int plen, struct list_head *fah, 1856 struct fib_table *tb, 1857 struct sk_buff *skb, struct netlink_callback *cb) 1858 { 1859 int i, s_i; 1860 struct fib_alias *fa; 1861 __be32 xkey = htonl(key); 1862 1863 s_i = cb->args[5]; 1864 i = 0; 1865 1866 /* rcu_read_lock is hold by caller */ 1867 1868 list_for_each_entry_rcu(fa, fah, fa_list) { 1869 if (i < s_i) { 1870 i++; 1871 continue; 1872 } 1873 1874 if (fib_dump_info(skb, NETLINK_CB(cb->skb).pid, 1875 cb->nlh->nlmsg_seq, 1876 RTM_NEWROUTE, 1877 tb->tb_id, 1878 fa->fa_type, 1879 xkey, 1880 plen, 1881 fa->fa_tos, 1882 fa->fa_info, NLM_F_MULTI) < 0) { 1883 cb->args[5] = i; 1884 return -1; 1885 } 1886 i++; 1887 } 1888 cb->args[5] = i; 1889 return skb->len; 1890 } 1891 1892 static int fn_trie_dump_leaf(struct leaf *l, struct fib_table *tb, 1893 struct sk_buff *skb, struct netlink_callback *cb) 1894 { 1895 struct leaf_info *li; 1896 struct hlist_node *node; 1897 int i, s_i; 1898 1899 s_i = cb->args[4]; 1900 i = 0; 1901 1902 /* rcu_read_lock is hold by caller */ 1903 hlist_for_each_entry_rcu(li, node, &l->list, hlist) { 1904 if (i < s_i) { 1905 i++; 1906 continue; 1907 } 1908 1909 if (i > s_i) 1910 cb->args[5] = 0; 1911 1912 if (list_empty(&li->falh)) 1913 continue; 1914 1915 if (fn_trie_dump_fa(l->key, li->plen, &li->falh, tb, skb, cb) < 0) { 1916 cb->args[4] = i; 1917 return -1; 1918 } 1919 i++; 1920 } 1921 1922 cb->args[4] = i; 1923 return skb->len; 1924 } 1925 1926 int fib_table_dump(struct fib_table *tb, struct sk_buff *skb, 1927 struct netlink_callback *cb) 1928 { 1929 struct leaf *l; 1930 struct trie *t = (struct trie *) tb->tb_data; 1931 t_key key = cb->args[2]; 1932 int count = cb->args[3]; 1933 1934 rcu_read_lock(); 1935 /* Dump starting at last key. 1936 * Note: 0.0.0.0/0 (ie default) is first key. 1937 */ 1938 if (count == 0) 1939 l = trie_firstleaf(t); 1940 else { 1941 /* Normally, continue from last key, but if that is missing 1942 * fallback to using slow rescan 1943 */ 1944 l = fib_find_node(t, key); 1945 if (!l) 1946 l = trie_leafindex(t, count); 1947 } 1948 1949 while (l) { 1950 cb->args[2] = l->key; 1951 if (fn_trie_dump_leaf(l, tb, skb, cb) < 0) { 1952 cb->args[3] = count; 1953 rcu_read_unlock(); 1954 return -1; 1955 } 1956 1957 ++count; 1958 l = trie_nextleaf(l); 1959 memset(&cb->args[4], 0, 1960 sizeof(cb->args) - 4*sizeof(cb->args[0])); 1961 } 1962 cb->args[3] = count; 1963 rcu_read_unlock(); 1964 1965 return skb->len; 1966 } 1967 1968 void __init fib_trie_init(void) 1969 { 1970 fn_alias_kmem = kmem_cache_create("ip_fib_alias", 1971 sizeof(struct fib_alias), 1972 0, SLAB_PANIC, NULL); 1973 1974 trie_leaf_kmem = kmem_cache_create("ip_fib_trie", 1975 max(sizeof(struct leaf), 1976 sizeof(struct leaf_info)), 1977 0, SLAB_PANIC, NULL); 1978 } 1979 1980 1981 struct fib_table *fib_trie_table(u32 id) 1982 { 1983 struct fib_table *tb; 1984 struct trie *t; 1985 1986 tb = kmalloc(sizeof(struct fib_table) + sizeof(struct trie), 1987 GFP_KERNEL); 1988 if (tb == NULL) 1989 return NULL; 1990 1991 tb->tb_id = id; 1992 tb->tb_default = -1; 1993 tb->tb_num_default = 0; 1994 1995 t = (struct trie *) tb->tb_data; 1996 memset(t, 0, sizeof(*t)); 1997 1998 return tb; 1999 } 2000 2001 #ifdef CONFIG_PROC_FS 2002 /* Depth first Trie walk iterator */ 2003 struct fib_trie_iter { 2004 struct seq_net_private p; 2005 struct fib_table *tb; 2006 struct tnode *tnode; 2007 unsigned int index; 2008 unsigned int depth; 2009 }; 2010 2011 static struct rt_trie_node *fib_trie_get_next(struct fib_trie_iter *iter) 2012 { 2013 struct tnode *tn = iter->tnode; 2014 unsigned int cindex = iter->index; 2015 struct tnode *p; 2016 2017 /* A single entry routing table */ 2018 if (!tn) 2019 return NULL; 2020 2021 pr_debug("get_next iter={node=%p index=%d depth=%d}\n", 2022 iter->tnode, iter->index, iter->depth); 2023 rescan: 2024 while (cindex < (1<<tn->bits)) { 2025 struct rt_trie_node *n = tnode_get_child_rcu(tn, cindex); 2026 2027 if (n) { 2028 if (IS_LEAF(n)) { 2029 iter->tnode = tn; 2030 iter->index = cindex + 1; 2031 } else { 2032 /* push down one level */ 2033 iter->tnode = (struct tnode *) n; 2034 iter->index = 0; 2035 ++iter->depth; 2036 } 2037 return n; 2038 } 2039 2040 ++cindex; 2041 } 2042 2043 /* Current node exhausted, pop back up */ 2044 p = node_parent_rcu((struct rt_trie_node *)tn); 2045 if (p) { 2046 cindex = tkey_extract_bits(tn->key, p->pos, p->bits)+1; 2047 tn = p; 2048 --iter->depth; 2049 goto rescan; 2050 } 2051 2052 /* got root? */ 2053 return NULL; 2054 } 2055 2056 static struct rt_trie_node *fib_trie_get_first(struct fib_trie_iter *iter, 2057 struct trie *t) 2058 { 2059 struct rt_trie_node *n; 2060 2061 if (!t) 2062 return NULL; 2063 2064 n = rcu_dereference(t->trie); 2065 if (!n) 2066 return NULL; 2067 2068 if (IS_TNODE(n)) { 2069 iter->tnode = (struct tnode *) n; 2070 iter->index = 0; 2071 iter->depth = 1; 2072 } else { 2073 iter->tnode = NULL; 2074 iter->index = 0; 2075 iter->depth = 0; 2076 } 2077 2078 return n; 2079 } 2080 2081 static void trie_collect_stats(struct trie *t, struct trie_stat *s) 2082 { 2083 struct rt_trie_node *n; 2084 struct fib_trie_iter iter; 2085 2086 memset(s, 0, sizeof(*s)); 2087 2088 rcu_read_lock(); 2089 for (n = fib_trie_get_first(&iter, t); n; n = fib_trie_get_next(&iter)) { 2090 if (IS_LEAF(n)) { 2091 struct leaf *l = (struct leaf *)n; 2092 struct leaf_info *li; 2093 struct hlist_node *tmp; 2094 2095 s->leaves++; 2096 s->totdepth += iter.depth; 2097 if (iter.depth > s->maxdepth) 2098 s->maxdepth = iter.depth; 2099 2100 hlist_for_each_entry_rcu(li, tmp, &l->list, hlist) 2101 ++s->prefixes; 2102 } else { 2103 const struct tnode *tn = (const struct tnode *) n; 2104 int i; 2105 2106 s->tnodes++; 2107 if (tn->bits < MAX_STAT_DEPTH) 2108 s->nodesizes[tn->bits]++; 2109 2110 for (i = 0; i < (1<<tn->bits); i++) 2111 if (!tn->child[i]) 2112 s->nullpointers++; 2113 } 2114 } 2115 rcu_read_unlock(); 2116 } 2117 2118 /* 2119 * This outputs /proc/net/fib_triestats 2120 */ 2121 static void trie_show_stats(struct seq_file *seq, struct trie_stat *stat) 2122 { 2123 unsigned int i, max, pointers, bytes, avdepth; 2124 2125 if (stat->leaves) 2126 avdepth = stat->totdepth*100 / stat->leaves; 2127 else 2128 avdepth = 0; 2129 2130 seq_printf(seq, "\tAver depth: %u.%02d\n", 2131 avdepth / 100, avdepth % 100); 2132 seq_printf(seq, "\tMax depth: %u\n", stat->maxdepth); 2133 2134 seq_printf(seq, "\tLeaves: %u\n", stat->leaves); 2135 bytes = sizeof(struct leaf) * stat->leaves; 2136 2137 seq_printf(seq, "\tPrefixes: %u\n", stat->prefixes); 2138 bytes += sizeof(struct leaf_info) * stat->prefixes; 2139 2140 seq_printf(seq, "\tInternal nodes: %u\n\t", stat->tnodes); 2141 bytes += sizeof(struct tnode) * stat->tnodes; 2142 2143 max = MAX_STAT_DEPTH; 2144 while (max > 0 && stat->nodesizes[max-1] == 0) 2145 max--; 2146 2147 pointers = 0; 2148 for (i = 1; i <= max; i++) 2149 if (stat->nodesizes[i] != 0) { 2150 seq_printf(seq, " %u: %u", i, stat->nodesizes[i]); 2151 pointers += (1<<i) * stat->nodesizes[i]; 2152 } 2153 seq_putc(seq, '\n'); 2154 seq_printf(seq, "\tPointers: %u\n", pointers); 2155 2156 bytes += sizeof(struct rt_trie_node *) * pointers; 2157 seq_printf(seq, "Null ptrs: %u\n", stat->nullpointers); 2158 seq_printf(seq, "Total size: %u kB\n", (bytes + 1023) / 1024); 2159 } 2160 2161 #ifdef CONFIG_IP_FIB_TRIE_STATS 2162 static void trie_show_usage(struct seq_file *seq, 2163 const struct trie_use_stats *stats) 2164 { 2165 seq_printf(seq, "\nCounters:\n---------\n"); 2166 seq_printf(seq, "gets = %u\n", stats->gets); 2167 seq_printf(seq, "backtracks = %u\n", stats->backtrack); 2168 seq_printf(seq, "semantic match passed = %u\n", 2169 stats->semantic_match_passed); 2170 seq_printf(seq, "semantic match miss = %u\n", 2171 stats->semantic_match_miss); 2172 seq_printf(seq, "null node hit= %u\n", stats->null_node_hit); 2173 seq_printf(seq, "skipped node resize = %u\n\n", 2174 stats->resize_node_skipped); 2175 } 2176 #endif /* CONFIG_IP_FIB_TRIE_STATS */ 2177 2178 static void fib_table_print(struct seq_file *seq, struct fib_table *tb) 2179 { 2180 if (tb->tb_id == RT_TABLE_LOCAL) 2181 seq_puts(seq, "Local:\n"); 2182 else if (tb->tb_id == RT_TABLE_MAIN) 2183 seq_puts(seq, "Main:\n"); 2184 else 2185 seq_printf(seq, "Id %d:\n", tb->tb_id); 2186 } 2187 2188 2189 static int fib_triestat_seq_show(struct seq_file *seq, void *v) 2190 { 2191 struct net *net = (struct net *)seq->private; 2192 unsigned int h; 2193 2194 seq_printf(seq, 2195 "Basic info: size of leaf:" 2196 " %Zd bytes, size of tnode: %Zd bytes.\n", 2197 sizeof(struct leaf), sizeof(struct tnode)); 2198 2199 for (h = 0; h < FIB_TABLE_HASHSZ; h++) { 2200 struct hlist_head *head = &net->ipv4.fib_table_hash[h]; 2201 struct hlist_node *node; 2202 struct fib_table *tb; 2203 2204 hlist_for_each_entry_rcu(tb, node, head, tb_hlist) { 2205 struct trie *t = (struct trie *) tb->tb_data; 2206 struct trie_stat stat; 2207 2208 if (!t) 2209 continue; 2210 2211 fib_table_print(seq, tb); 2212 2213 trie_collect_stats(t, &stat); 2214 trie_show_stats(seq, &stat); 2215 #ifdef CONFIG_IP_FIB_TRIE_STATS 2216 trie_show_usage(seq, &t->stats); 2217 #endif 2218 } 2219 } 2220 2221 return 0; 2222 } 2223 2224 static int fib_triestat_seq_open(struct inode *inode, struct file *file) 2225 { 2226 return single_open_net(inode, file, fib_triestat_seq_show); 2227 } 2228 2229 static const struct file_operations fib_triestat_fops = { 2230 .owner = THIS_MODULE, 2231 .open = fib_triestat_seq_open, 2232 .read = seq_read, 2233 .llseek = seq_lseek, 2234 .release = single_release_net, 2235 }; 2236 2237 static struct rt_trie_node *fib_trie_get_idx(struct seq_file *seq, loff_t pos) 2238 { 2239 struct fib_trie_iter *iter = seq->private; 2240 struct net *net = seq_file_net(seq); 2241 loff_t idx = 0; 2242 unsigned int h; 2243 2244 for (h = 0; h < FIB_TABLE_HASHSZ; h++) { 2245 struct hlist_head *head = &net->ipv4.fib_table_hash[h]; 2246 struct hlist_node *node; 2247 struct fib_table *tb; 2248 2249 hlist_for_each_entry_rcu(tb, node, head, tb_hlist) { 2250 struct rt_trie_node *n; 2251 2252 for (n = fib_trie_get_first(iter, 2253 (struct trie *) tb->tb_data); 2254 n; n = fib_trie_get_next(iter)) 2255 if (pos == idx++) { 2256 iter->tb = tb; 2257 return n; 2258 } 2259 } 2260 } 2261 2262 return NULL; 2263 } 2264 2265 static void *fib_trie_seq_start(struct seq_file *seq, loff_t *pos) 2266 __acquires(RCU) 2267 { 2268 rcu_read_lock(); 2269 return fib_trie_get_idx(seq, *pos); 2270 } 2271 2272 static void *fib_trie_seq_next(struct seq_file *seq, void *v, loff_t *pos) 2273 { 2274 struct fib_trie_iter *iter = seq->private; 2275 struct net *net = seq_file_net(seq); 2276 struct fib_table *tb = iter->tb; 2277 struct hlist_node *tb_node; 2278 unsigned int h; 2279 struct rt_trie_node *n; 2280 2281 ++*pos; 2282 /* next node in same table */ 2283 n = fib_trie_get_next(iter); 2284 if (n) 2285 return n; 2286 2287 /* walk rest of this hash chain */ 2288 h = tb->tb_id & (FIB_TABLE_HASHSZ - 1); 2289 while ((tb_node = rcu_dereference(hlist_next_rcu(&tb->tb_hlist)))) { 2290 tb = hlist_entry(tb_node, struct fib_table, tb_hlist); 2291 n = fib_trie_get_first(iter, (struct trie *) tb->tb_data); 2292 if (n) 2293 goto found; 2294 } 2295 2296 /* new hash chain */ 2297 while (++h < FIB_TABLE_HASHSZ) { 2298 struct hlist_head *head = &net->ipv4.fib_table_hash[h]; 2299 hlist_for_each_entry_rcu(tb, tb_node, head, tb_hlist) { 2300 n = fib_trie_get_first(iter, (struct trie *) tb->tb_data); 2301 if (n) 2302 goto found; 2303 } 2304 } 2305 return NULL; 2306 2307 found: 2308 iter->tb = tb; 2309 return n; 2310 } 2311 2312 static void fib_trie_seq_stop(struct seq_file *seq, void *v) 2313 __releases(RCU) 2314 { 2315 rcu_read_unlock(); 2316 } 2317 2318 static void seq_indent(struct seq_file *seq, int n) 2319 { 2320 while (n-- > 0) 2321 seq_puts(seq, " "); 2322 } 2323 2324 static inline const char *rtn_scope(char *buf, size_t len, enum rt_scope_t s) 2325 { 2326 switch (s) { 2327 case RT_SCOPE_UNIVERSE: return "universe"; 2328 case RT_SCOPE_SITE: return "site"; 2329 case RT_SCOPE_LINK: return "link"; 2330 case RT_SCOPE_HOST: return "host"; 2331 case RT_SCOPE_NOWHERE: return "nowhere"; 2332 default: 2333 snprintf(buf, len, "scope=%d", s); 2334 return buf; 2335 } 2336 } 2337 2338 static const char *const rtn_type_names[__RTN_MAX] = { 2339 [RTN_UNSPEC] = "UNSPEC", 2340 [RTN_UNICAST] = "UNICAST", 2341 [RTN_LOCAL] = "LOCAL", 2342 [RTN_BROADCAST] = "BROADCAST", 2343 [RTN_ANYCAST] = "ANYCAST", 2344 [RTN_MULTICAST] = "MULTICAST", 2345 [RTN_BLACKHOLE] = "BLACKHOLE", 2346 [RTN_UNREACHABLE] = "UNREACHABLE", 2347 [RTN_PROHIBIT] = "PROHIBIT", 2348 [RTN_THROW] = "THROW", 2349 [RTN_NAT] = "NAT", 2350 [RTN_XRESOLVE] = "XRESOLVE", 2351 }; 2352 2353 static inline const char *rtn_type(char *buf, size_t len, unsigned int t) 2354 { 2355 if (t < __RTN_MAX && rtn_type_names[t]) 2356 return rtn_type_names[t]; 2357 snprintf(buf, len, "type %u", t); 2358 return buf; 2359 } 2360 2361 /* Pretty print the trie */ 2362 static int fib_trie_seq_show(struct seq_file *seq, void *v) 2363 { 2364 const struct fib_trie_iter *iter = seq->private; 2365 struct rt_trie_node *n = v; 2366 2367 if (!node_parent_rcu(n)) 2368 fib_table_print(seq, iter->tb); 2369 2370 if (IS_TNODE(n)) { 2371 struct tnode *tn = (struct tnode *) n; 2372 __be32 prf = htonl(mask_pfx(tn->key, tn->pos)); 2373 2374 seq_indent(seq, iter->depth-1); 2375 seq_printf(seq, " +-- %pI4/%d %d %d %d\n", 2376 &prf, tn->pos, tn->bits, tn->full_children, 2377 tn->empty_children); 2378 2379 } else { 2380 struct leaf *l = (struct leaf *) n; 2381 struct leaf_info *li; 2382 struct hlist_node *node; 2383 __be32 val = htonl(l->key); 2384 2385 seq_indent(seq, iter->depth); 2386 seq_printf(seq, " |-- %pI4\n", &val); 2387 2388 hlist_for_each_entry_rcu(li, node, &l->list, hlist) { 2389 struct fib_alias *fa; 2390 2391 list_for_each_entry_rcu(fa, &li->falh, fa_list) { 2392 char buf1[32], buf2[32]; 2393 2394 seq_indent(seq, iter->depth+1); 2395 seq_printf(seq, " /%d %s %s", li->plen, 2396 rtn_scope(buf1, sizeof(buf1), 2397 fa->fa_info->fib_scope), 2398 rtn_type(buf2, sizeof(buf2), 2399 fa->fa_type)); 2400 if (fa->fa_tos) 2401 seq_printf(seq, " tos=%d", fa->fa_tos); 2402 seq_putc(seq, '\n'); 2403 } 2404 } 2405 } 2406 2407 return 0; 2408 } 2409 2410 static const struct seq_operations fib_trie_seq_ops = { 2411 .start = fib_trie_seq_start, 2412 .next = fib_trie_seq_next, 2413 .stop = fib_trie_seq_stop, 2414 .show = fib_trie_seq_show, 2415 }; 2416 2417 static int fib_trie_seq_open(struct inode *inode, struct file *file) 2418 { 2419 return seq_open_net(inode, file, &fib_trie_seq_ops, 2420 sizeof(struct fib_trie_iter)); 2421 } 2422 2423 static const struct file_operations fib_trie_fops = { 2424 .owner = THIS_MODULE, 2425 .open = fib_trie_seq_open, 2426 .read = seq_read, 2427 .llseek = seq_lseek, 2428 .release = seq_release_net, 2429 }; 2430 2431 struct fib_route_iter { 2432 struct seq_net_private p; 2433 struct trie *main_trie; 2434 loff_t pos; 2435 t_key key; 2436 }; 2437 2438 static struct leaf *fib_route_get_idx(struct fib_route_iter *iter, loff_t pos) 2439 { 2440 struct leaf *l = NULL; 2441 struct trie *t = iter->main_trie; 2442 2443 /* use cache location of last found key */ 2444 if (iter->pos > 0 && pos >= iter->pos && (l = fib_find_node(t, iter->key))) 2445 pos -= iter->pos; 2446 else { 2447 iter->pos = 0; 2448 l = trie_firstleaf(t); 2449 } 2450 2451 while (l && pos-- > 0) { 2452 iter->pos++; 2453 l = trie_nextleaf(l); 2454 } 2455 2456 if (l) 2457 iter->key = pos; /* remember it */ 2458 else 2459 iter->pos = 0; /* forget it */ 2460 2461 return l; 2462 } 2463 2464 static void *fib_route_seq_start(struct seq_file *seq, loff_t *pos) 2465 __acquires(RCU) 2466 { 2467 struct fib_route_iter *iter = seq->private; 2468 struct fib_table *tb; 2469 2470 rcu_read_lock(); 2471 tb = fib_get_table(seq_file_net(seq), RT_TABLE_MAIN); 2472 if (!tb) 2473 return NULL; 2474 2475 iter->main_trie = (struct trie *) tb->tb_data; 2476 if (*pos == 0) 2477 return SEQ_START_TOKEN; 2478 else 2479 return fib_route_get_idx(iter, *pos - 1); 2480 } 2481 2482 static void *fib_route_seq_next(struct seq_file *seq, void *v, loff_t *pos) 2483 { 2484 struct fib_route_iter *iter = seq->private; 2485 struct leaf *l = v; 2486 2487 ++*pos; 2488 if (v == SEQ_START_TOKEN) { 2489 iter->pos = 0; 2490 l = trie_firstleaf(iter->main_trie); 2491 } else { 2492 iter->pos++; 2493 l = trie_nextleaf(l); 2494 } 2495 2496 if (l) 2497 iter->key = l->key; 2498 else 2499 iter->pos = 0; 2500 return l; 2501 } 2502 2503 static void fib_route_seq_stop(struct seq_file *seq, void *v) 2504 __releases(RCU) 2505 { 2506 rcu_read_unlock(); 2507 } 2508 2509 static unsigned int fib_flag_trans(int type, __be32 mask, const struct fib_info *fi) 2510 { 2511 unsigned int flags = 0; 2512 2513 if (type == RTN_UNREACHABLE || type == RTN_PROHIBIT) 2514 flags = RTF_REJECT; 2515 if (fi && fi->fib_nh->nh_gw) 2516 flags |= RTF_GATEWAY; 2517 if (mask == htonl(0xFFFFFFFF)) 2518 flags |= RTF_HOST; 2519 flags |= RTF_UP; 2520 return flags; 2521 } 2522 2523 /* 2524 * This outputs /proc/net/route. 2525 * The format of the file is not supposed to be changed 2526 * and needs to be same as fib_hash output to avoid breaking 2527 * legacy utilities 2528 */ 2529 static int fib_route_seq_show(struct seq_file *seq, void *v) 2530 { 2531 struct leaf *l = v; 2532 struct leaf_info *li; 2533 struct hlist_node *node; 2534 2535 if (v == SEQ_START_TOKEN) { 2536 seq_printf(seq, "%-127s\n", "Iface\tDestination\tGateway " 2537 "\tFlags\tRefCnt\tUse\tMetric\tMask\t\tMTU" 2538 "\tWindow\tIRTT"); 2539 return 0; 2540 } 2541 2542 hlist_for_each_entry_rcu(li, node, &l->list, hlist) { 2543 struct fib_alias *fa; 2544 __be32 mask, prefix; 2545 2546 mask = inet_make_mask(li->plen); 2547 prefix = htonl(l->key); 2548 2549 list_for_each_entry_rcu(fa, &li->falh, fa_list) { 2550 const struct fib_info *fi = fa->fa_info; 2551 unsigned int flags = fib_flag_trans(fa->fa_type, mask, fi); 2552 int len; 2553 2554 if (fa->fa_type == RTN_BROADCAST 2555 || fa->fa_type == RTN_MULTICAST) 2556 continue; 2557 2558 if (fi) 2559 seq_printf(seq, 2560 "%s\t%08X\t%08X\t%04X\t%d\t%u\t" 2561 "%d\t%08X\t%d\t%u\t%u%n", 2562 fi->fib_dev ? fi->fib_dev->name : "*", 2563 prefix, 2564 fi->fib_nh->nh_gw, flags, 0, 0, 2565 fi->fib_priority, 2566 mask, 2567 (fi->fib_advmss ? 2568 fi->fib_advmss + 40 : 0), 2569 fi->fib_window, 2570 fi->fib_rtt >> 3, &len); 2571 else 2572 seq_printf(seq, 2573 "*\t%08X\t%08X\t%04X\t%d\t%u\t" 2574 "%d\t%08X\t%d\t%u\t%u%n", 2575 prefix, 0, flags, 0, 0, 0, 2576 mask, 0, 0, 0, &len); 2577 2578 seq_printf(seq, "%*s\n", 127 - len, ""); 2579 } 2580 } 2581 2582 return 0; 2583 } 2584 2585 static const struct seq_operations fib_route_seq_ops = { 2586 .start = fib_route_seq_start, 2587 .next = fib_route_seq_next, 2588 .stop = fib_route_seq_stop, 2589 .show = fib_route_seq_show, 2590 }; 2591 2592 static int fib_route_seq_open(struct inode *inode, struct file *file) 2593 { 2594 return seq_open_net(inode, file, &fib_route_seq_ops, 2595 sizeof(struct fib_route_iter)); 2596 } 2597 2598 static const struct file_operations fib_route_fops = { 2599 .owner = THIS_MODULE, 2600 .open = fib_route_seq_open, 2601 .read = seq_read, 2602 .llseek = seq_lseek, 2603 .release = seq_release_net, 2604 }; 2605 2606 int __net_init fib_proc_init(struct net *net) 2607 { 2608 if (!proc_net_fops_create(net, "fib_trie", S_IRUGO, &fib_trie_fops)) 2609 goto out1; 2610 2611 if (!proc_net_fops_create(net, "fib_triestat", S_IRUGO, 2612 &fib_triestat_fops)) 2613 goto out2; 2614 2615 if (!proc_net_fops_create(net, "route", S_IRUGO, &fib_route_fops)) 2616 goto out3; 2617 2618 return 0; 2619 2620 out3: 2621 proc_net_remove(net, "fib_triestat"); 2622 out2: 2623 proc_net_remove(net, "fib_trie"); 2624 out1: 2625 return -ENOMEM; 2626 } 2627 2628 void __net_exit fib_proc_exit(struct net *net) 2629 { 2630 proc_net_remove(net, "fib_trie"); 2631 proc_net_remove(net, "fib_triestat"); 2632 proc_net_remove(net, "route"); 2633 } 2634 2635 #endif /* CONFIG_PROC_FS */ 2636