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 descibed 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.nada.kth.se/~snilsson/public/papers/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 "0.408" 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 <net/net_namespace.h> 75 #include <net/ip.h> 76 #include <net/protocol.h> 77 #include <net/route.h> 78 #include <net/tcp.h> 79 #include <net/sock.h> 80 #include <net/ip_fib.h> 81 #include "fib_lookup.h" 82 83 #define MAX_STAT_DEPTH 32 84 85 #define KEYLENGTH (8*sizeof(t_key)) 86 87 typedef unsigned int t_key; 88 89 #define T_TNODE 0 90 #define T_LEAF 1 91 #define NODE_TYPE_MASK 0x1UL 92 #define NODE_TYPE(node) ((node)->parent & NODE_TYPE_MASK) 93 94 #define IS_TNODE(n) (!(n->parent & T_LEAF)) 95 #define IS_LEAF(n) (n->parent & T_LEAF) 96 97 struct node { 98 unsigned long parent; 99 t_key key; 100 }; 101 102 struct leaf { 103 unsigned long parent; 104 t_key key; 105 struct hlist_head list; 106 struct rcu_head rcu; 107 }; 108 109 struct leaf_info { 110 struct hlist_node hlist; 111 struct rcu_head rcu; 112 int plen; 113 struct list_head falh; 114 }; 115 116 struct tnode { 117 unsigned long parent; 118 t_key key; 119 unsigned char pos; /* 2log(KEYLENGTH) bits needed */ 120 unsigned char bits; /* 2log(KEYLENGTH) bits needed */ 121 unsigned int full_children; /* KEYLENGTH bits needed */ 122 unsigned int empty_children; /* KEYLENGTH bits needed */ 123 union { 124 struct rcu_head rcu; 125 struct work_struct work; 126 struct tnode *tnode_free; 127 }; 128 struct node *child[0]; 129 }; 130 131 #ifdef CONFIG_IP_FIB_TRIE_STATS 132 struct trie_use_stats { 133 unsigned int gets; 134 unsigned int backtrack; 135 unsigned int semantic_match_passed; 136 unsigned int semantic_match_miss; 137 unsigned int null_node_hit; 138 unsigned int resize_node_skipped; 139 }; 140 #endif 141 142 struct trie_stat { 143 unsigned int totdepth; 144 unsigned int maxdepth; 145 unsigned int tnodes; 146 unsigned int leaves; 147 unsigned int nullpointers; 148 unsigned int prefixes; 149 unsigned int nodesizes[MAX_STAT_DEPTH]; 150 }; 151 152 struct trie { 153 struct node *trie; 154 #ifdef CONFIG_IP_FIB_TRIE_STATS 155 struct trie_use_stats stats; 156 #endif 157 }; 158 159 static void put_child(struct trie *t, struct tnode *tn, int i, struct node *n); 160 static void tnode_put_child_reorg(struct tnode *tn, int i, struct node *n, 161 int wasfull); 162 static struct node *resize(struct trie *t, struct tnode *tn); 163 static struct tnode *inflate(struct trie *t, struct tnode *tn); 164 static struct tnode *halve(struct trie *t, struct tnode *tn); 165 /* tnodes to free after resize(); protected by RTNL */ 166 static struct tnode *tnode_free_head; 167 168 static struct kmem_cache *fn_alias_kmem __read_mostly; 169 static struct kmem_cache *trie_leaf_kmem __read_mostly; 170 171 static inline struct tnode *node_parent(struct node *node) 172 { 173 return (struct tnode *)(node->parent & ~NODE_TYPE_MASK); 174 } 175 176 static inline struct tnode *node_parent_rcu(struct node *node) 177 { 178 struct tnode *ret = node_parent(node); 179 180 return rcu_dereference(ret); 181 } 182 183 /* Same as rcu_assign_pointer 184 * but that macro() assumes that value is a pointer. 185 */ 186 static inline void node_set_parent(struct node *node, struct tnode *ptr) 187 { 188 smp_wmb(); 189 node->parent = (unsigned long)ptr | NODE_TYPE(node); 190 } 191 192 static inline struct node *tnode_get_child(struct tnode *tn, unsigned int i) 193 { 194 BUG_ON(i >= 1U << tn->bits); 195 196 return tn->child[i]; 197 } 198 199 static inline struct node *tnode_get_child_rcu(struct tnode *tn, unsigned int i) 200 { 201 struct node *ret = tnode_get_child(tn, i); 202 203 return rcu_dereference(ret); 204 } 205 206 static inline int tnode_child_length(const struct tnode *tn) 207 { 208 return 1 << tn->bits; 209 } 210 211 static inline t_key mask_pfx(t_key k, unsigned short l) 212 { 213 return (l == 0) ? 0 : k >> (KEYLENGTH-l) << (KEYLENGTH-l); 214 } 215 216 static inline t_key tkey_extract_bits(t_key a, int offset, int bits) 217 { 218 if (offset < KEYLENGTH) 219 return ((t_key)(a << offset)) >> (KEYLENGTH - bits); 220 else 221 return 0; 222 } 223 224 static inline int tkey_equals(t_key a, t_key b) 225 { 226 return a == b; 227 } 228 229 static inline int tkey_sub_equals(t_key a, int offset, int bits, t_key b) 230 { 231 if (bits == 0 || offset >= KEYLENGTH) 232 return 1; 233 bits = bits > KEYLENGTH ? KEYLENGTH : bits; 234 return ((a ^ b) << offset) >> (KEYLENGTH - bits) == 0; 235 } 236 237 static inline int tkey_mismatch(t_key a, int offset, t_key b) 238 { 239 t_key diff = a ^ b; 240 int i = offset; 241 242 if (!diff) 243 return 0; 244 while ((diff << i) >> (KEYLENGTH-1) == 0) 245 i++; 246 return i; 247 } 248 249 /* 250 To understand this stuff, an understanding of keys and all their bits is 251 necessary. Every node in the trie has a key associated with it, but not 252 all of the bits in that key are significant. 253 254 Consider a node 'n' and its parent 'tp'. 255 256 If n is a leaf, every bit in its key is significant. Its presence is 257 necessitated by path compression, since during a tree traversal (when 258 searching for a leaf - unless we are doing an insertion) we will completely 259 ignore all skipped bits we encounter. Thus we need to verify, at the end of 260 a potentially successful search, that we have indeed been walking the 261 correct key path. 262 263 Note that we can never "miss" the correct key in the tree if present by 264 following the wrong path. Path compression ensures that segments of the key 265 that are the same for all keys with a given prefix are skipped, but the 266 skipped part *is* identical for each node in the subtrie below the skipped 267 bit! trie_insert() in this implementation takes care of that - note the 268 call to tkey_sub_equals() in trie_insert(). 269 270 if n is an internal node - a 'tnode' here, the various parts of its key 271 have many different meanings. 272 273 Example: 274 _________________________________________________________________ 275 | i | i | i | i | i | i | i | N | N | N | S | S | S | S | S | C | 276 ----------------------------------------------------------------- 277 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 278 279 _________________________________________________________________ 280 | C | C | C | u | u | u | u | u | u | u | u | u | u | u | u | u | 281 ----------------------------------------------------------------- 282 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 283 284 tp->pos = 7 285 tp->bits = 3 286 n->pos = 15 287 n->bits = 4 288 289 First, let's just ignore the bits that come before the parent tp, that is 290 the bits from 0 to (tp->pos-1). They are *known* but at this point we do 291 not use them for anything. 292 293 The bits from (tp->pos) to (tp->pos + tp->bits - 1) - "N", above - are the 294 index into the parent's child array. That is, they will be used to find 295 'n' among tp's children. 296 297 The bits from (tp->pos + tp->bits) to (n->pos - 1) - "S" - are skipped bits 298 for the node n. 299 300 All the bits we have seen so far are significant to the node n. The rest 301 of the bits are really not needed or indeed known in n->key. 302 303 The bits from (n->pos) to (n->pos + n->bits - 1) - "C" - are the index into 304 n's child array, and will of course be different for each child. 305 306 307 The rest of the bits, from (n->pos + n->bits) onward, are completely unknown 308 at this point. 309 310 */ 311 312 static inline void check_tnode(const struct tnode *tn) 313 { 314 WARN_ON(tn && tn->pos+tn->bits > 32); 315 } 316 317 static const int halve_threshold = 25; 318 static const int inflate_threshold = 50; 319 static const int halve_threshold_root = 15; 320 static const int inflate_threshold_root = 25; 321 322 323 static void __alias_free_mem(struct rcu_head *head) 324 { 325 struct fib_alias *fa = container_of(head, struct fib_alias, rcu); 326 kmem_cache_free(fn_alias_kmem, fa); 327 } 328 329 static inline void alias_free_mem_rcu(struct fib_alias *fa) 330 { 331 call_rcu(&fa->rcu, __alias_free_mem); 332 } 333 334 static void __leaf_free_rcu(struct rcu_head *head) 335 { 336 struct leaf *l = container_of(head, struct leaf, rcu); 337 kmem_cache_free(trie_leaf_kmem, l); 338 } 339 340 static inline void free_leaf(struct leaf *l) 341 { 342 call_rcu_bh(&l->rcu, __leaf_free_rcu); 343 } 344 345 static void __leaf_info_free_rcu(struct rcu_head *head) 346 { 347 kfree(container_of(head, struct leaf_info, rcu)); 348 } 349 350 static inline void free_leaf_info(struct leaf_info *leaf) 351 { 352 call_rcu(&leaf->rcu, __leaf_info_free_rcu); 353 } 354 355 static struct tnode *tnode_alloc(size_t size) 356 { 357 if (size <= PAGE_SIZE) 358 return kzalloc(size, GFP_KERNEL); 359 else 360 return __vmalloc(size, GFP_KERNEL | __GFP_ZERO, PAGE_KERNEL); 361 } 362 363 static void __tnode_vfree(struct work_struct *arg) 364 { 365 struct tnode *tn = container_of(arg, struct tnode, work); 366 vfree(tn); 367 } 368 369 static void __tnode_free_rcu(struct rcu_head *head) 370 { 371 struct tnode *tn = container_of(head, struct tnode, rcu); 372 size_t size = sizeof(struct tnode) + 373 (sizeof(struct node *) << tn->bits); 374 375 if (size <= PAGE_SIZE) 376 kfree(tn); 377 else { 378 INIT_WORK(&tn->work, __tnode_vfree); 379 schedule_work(&tn->work); 380 } 381 } 382 383 static inline void tnode_free(struct tnode *tn) 384 { 385 if (IS_LEAF(tn)) 386 free_leaf((struct leaf *) tn); 387 else 388 call_rcu(&tn->rcu, __tnode_free_rcu); 389 } 390 391 static void tnode_free_safe(struct tnode *tn) 392 { 393 BUG_ON(IS_LEAF(tn)); 394 tn->tnode_free = tnode_free_head; 395 tnode_free_head = tn; 396 } 397 398 static void tnode_free_flush(void) 399 { 400 struct tnode *tn; 401 402 while ((tn = tnode_free_head)) { 403 tnode_free_head = tn->tnode_free; 404 tn->tnode_free = NULL; 405 tnode_free(tn); 406 } 407 } 408 409 static struct leaf *leaf_new(void) 410 { 411 struct leaf *l = kmem_cache_alloc(trie_leaf_kmem, GFP_KERNEL); 412 if (l) { 413 l->parent = T_LEAF; 414 INIT_HLIST_HEAD(&l->list); 415 } 416 return l; 417 } 418 419 static struct leaf_info *leaf_info_new(int plen) 420 { 421 struct leaf_info *li = kmalloc(sizeof(struct leaf_info), GFP_KERNEL); 422 if (li) { 423 li->plen = plen; 424 INIT_LIST_HEAD(&li->falh); 425 } 426 return li; 427 } 428 429 static struct tnode *tnode_new(t_key key, int pos, int bits) 430 { 431 size_t sz = sizeof(struct tnode) + (sizeof(struct node *) << bits); 432 struct tnode *tn = tnode_alloc(sz); 433 434 if (tn) { 435 tn->parent = T_TNODE; 436 tn->pos = pos; 437 tn->bits = bits; 438 tn->key = key; 439 tn->full_children = 0; 440 tn->empty_children = 1<<bits; 441 } 442 443 pr_debug("AT %p s=%u %lu\n", tn, (unsigned int) sizeof(struct tnode), 444 (unsigned long) (sizeof(struct node) << bits)); 445 return tn; 446 } 447 448 /* 449 * Check whether a tnode 'n' is "full", i.e. it is an internal node 450 * and no bits are skipped. See discussion in dyntree paper p. 6 451 */ 452 453 static inline int tnode_full(const struct tnode *tn, const struct node *n) 454 { 455 if (n == NULL || IS_LEAF(n)) 456 return 0; 457 458 return ((struct tnode *) n)->pos == tn->pos + tn->bits; 459 } 460 461 static inline void put_child(struct trie *t, struct tnode *tn, int i, 462 struct node *n) 463 { 464 tnode_put_child_reorg(tn, i, n, -1); 465 } 466 467 /* 468 * Add a child at position i overwriting the old value. 469 * Update the value of full_children and empty_children. 470 */ 471 472 static void tnode_put_child_reorg(struct tnode *tn, int i, struct node *n, 473 int wasfull) 474 { 475 struct node *chi = tn->child[i]; 476 int isfull; 477 478 BUG_ON(i >= 1<<tn->bits); 479 480 /* update emptyChildren */ 481 if (n == NULL && chi != NULL) 482 tn->empty_children++; 483 else if (n != NULL && chi == NULL) 484 tn->empty_children--; 485 486 /* update fullChildren */ 487 if (wasfull == -1) 488 wasfull = tnode_full(tn, chi); 489 490 isfull = tnode_full(tn, n); 491 if (wasfull && !isfull) 492 tn->full_children--; 493 else if (!wasfull && isfull) 494 tn->full_children++; 495 496 if (n) 497 node_set_parent(n, tn); 498 499 rcu_assign_pointer(tn->child[i], n); 500 } 501 502 static struct node *resize(struct trie *t, struct tnode *tn) 503 { 504 int i; 505 int err = 0; 506 struct tnode *old_tn; 507 int inflate_threshold_use; 508 int halve_threshold_use; 509 int max_resize; 510 511 if (!tn) 512 return NULL; 513 514 pr_debug("In tnode_resize %p inflate_threshold=%d threshold=%d\n", 515 tn, inflate_threshold, halve_threshold); 516 517 /* No children */ 518 if (tn->empty_children == tnode_child_length(tn)) { 519 tnode_free_safe(tn); 520 return NULL; 521 } 522 /* One child */ 523 if (tn->empty_children == tnode_child_length(tn) - 1) 524 for (i = 0; i < tnode_child_length(tn); i++) { 525 struct node *n; 526 527 n = tn->child[i]; 528 if (!n) 529 continue; 530 531 /* compress one level */ 532 node_set_parent(n, NULL); 533 tnode_free_safe(tn); 534 return n; 535 } 536 /* 537 * Double as long as the resulting node has a number of 538 * nonempty nodes that are above the threshold. 539 */ 540 541 /* 542 * From "Implementing a dynamic compressed trie" by Stefan Nilsson of 543 * the Helsinki University of Technology and Matti Tikkanen of Nokia 544 * Telecommunications, page 6: 545 * "A node is doubled if the ratio of non-empty children to all 546 * children in the *doubled* node is at least 'high'." 547 * 548 * 'high' in this instance is the variable 'inflate_threshold'. It 549 * is expressed as a percentage, so we multiply it with 550 * tnode_child_length() and instead of multiplying by 2 (since the 551 * child array will be doubled by inflate()) and multiplying 552 * the left-hand side by 100 (to handle the percentage thing) we 553 * multiply the left-hand side by 50. 554 * 555 * The left-hand side may look a bit weird: tnode_child_length(tn) 556 * - tn->empty_children is of course the number of non-null children 557 * in the current node. tn->full_children is the number of "full" 558 * children, that is non-null tnodes with a skip value of 0. 559 * All of those will be doubled in the resulting inflated tnode, so 560 * we just count them one extra time here. 561 * 562 * A clearer way to write this would be: 563 * 564 * to_be_doubled = tn->full_children; 565 * not_to_be_doubled = tnode_child_length(tn) - tn->empty_children - 566 * tn->full_children; 567 * 568 * new_child_length = tnode_child_length(tn) * 2; 569 * 570 * new_fill_factor = 100 * (not_to_be_doubled + 2*to_be_doubled) / 571 * new_child_length; 572 * if (new_fill_factor >= inflate_threshold) 573 * 574 * ...and so on, tho it would mess up the while () loop. 575 * 576 * anyway, 577 * 100 * (not_to_be_doubled + 2*to_be_doubled) / new_child_length >= 578 * inflate_threshold 579 * 580 * avoid a division: 581 * 100 * (not_to_be_doubled + 2*to_be_doubled) >= 582 * inflate_threshold * new_child_length 583 * 584 * expand not_to_be_doubled and to_be_doubled, and shorten: 585 * 100 * (tnode_child_length(tn) - tn->empty_children + 586 * tn->full_children) >= inflate_threshold * new_child_length 587 * 588 * expand new_child_length: 589 * 100 * (tnode_child_length(tn) - tn->empty_children + 590 * tn->full_children) >= 591 * inflate_threshold * tnode_child_length(tn) * 2 592 * 593 * shorten again: 594 * 50 * (tn->full_children + tnode_child_length(tn) - 595 * tn->empty_children) >= inflate_threshold * 596 * tnode_child_length(tn) 597 * 598 */ 599 600 check_tnode(tn); 601 602 /* Keep root node larger */ 603 604 if (!tn->parent) 605 inflate_threshold_use = inflate_threshold_root; 606 else 607 inflate_threshold_use = inflate_threshold; 608 609 err = 0; 610 max_resize = 10; 611 while ((tn->full_children > 0 && max_resize-- && 612 50 * (tn->full_children + tnode_child_length(tn) 613 - tn->empty_children) 614 >= inflate_threshold_use * tnode_child_length(tn))) { 615 616 old_tn = tn; 617 tn = inflate(t, tn); 618 619 if (IS_ERR(tn)) { 620 tn = old_tn; 621 #ifdef CONFIG_IP_FIB_TRIE_STATS 622 t->stats.resize_node_skipped++; 623 #endif 624 break; 625 } 626 } 627 628 if (max_resize < 0) { 629 if (!tn->parent) 630 pr_warning("Fix inflate_threshold_root." 631 " Now=%d size=%d bits\n", 632 inflate_threshold_root, tn->bits); 633 else 634 pr_warning("Fix inflate_threshold." 635 " Now=%d size=%d bits\n", 636 inflate_threshold, tn->bits); 637 } 638 639 check_tnode(tn); 640 641 /* 642 * Halve as long as the number of empty children in this 643 * node is above threshold. 644 */ 645 646 647 /* Keep root node larger */ 648 649 if (!tn->parent) 650 halve_threshold_use = halve_threshold_root; 651 else 652 halve_threshold_use = halve_threshold; 653 654 err = 0; 655 max_resize = 10; 656 while (tn->bits > 1 && max_resize-- && 657 100 * (tnode_child_length(tn) - tn->empty_children) < 658 halve_threshold_use * tnode_child_length(tn)) { 659 660 old_tn = tn; 661 tn = halve(t, tn); 662 if (IS_ERR(tn)) { 663 tn = old_tn; 664 #ifdef CONFIG_IP_FIB_TRIE_STATS 665 t->stats.resize_node_skipped++; 666 #endif 667 break; 668 } 669 } 670 671 if (max_resize < 0) { 672 if (!tn->parent) 673 pr_warning("Fix halve_threshold_root." 674 " Now=%d size=%d bits\n", 675 halve_threshold_root, tn->bits); 676 else 677 pr_warning("Fix halve_threshold." 678 " Now=%d size=%d bits\n", 679 halve_threshold, tn->bits); 680 } 681 682 /* Only one child remains */ 683 if (tn->empty_children == tnode_child_length(tn) - 1) 684 for (i = 0; i < tnode_child_length(tn); i++) { 685 struct node *n; 686 687 n = tn->child[i]; 688 if (!n) 689 continue; 690 691 /* compress one level */ 692 693 node_set_parent(n, NULL); 694 tnode_free_safe(tn); 695 return n; 696 } 697 698 return (struct node *) tn; 699 } 700 701 static struct tnode *inflate(struct trie *t, struct tnode *tn) 702 { 703 struct tnode *oldtnode = tn; 704 int olen = tnode_child_length(tn); 705 int i; 706 707 pr_debug("In inflate\n"); 708 709 tn = tnode_new(oldtnode->key, oldtnode->pos, oldtnode->bits + 1); 710 711 if (!tn) 712 return ERR_PTR(-ENOMEM); 713 714 /* 715 * Preallocate and store tnodes before the actual work so we 716 * don't get into an inconsistent state if memory allocation 717 * fails. In case of failure we return the oldnode and inflate 718 * of tnode is ignored. 719 */ 720 721 for (i = 0; i < olen; i++) { 722 struct tnode *inode; 723 724 inode = (struct tnode *) tnode_get_child(oldtnode, i); 725 if (inode && 726 IS_TNODE(inode) && 727 inode->pos == oldtnode->pos + oldtnode->bits && 728 inode->bits > 1) { 729 struct tnode *left, *right; 730 t_key m = ~0U << (KEYLENGTH - 1) >> inode->pos; 731 732 left = tnode_new(inode->key&(~m), inode->pos + 1, 733 inode->bits - 1); 734 if (!left) 735 goto nomem; 736 737 right = tnode_new(inode->key|m, inode->pos + 1, 738 inode->bits - 1); 739 740 if (!right) { 741 tnode_free(left); 742 goto nomem; 743 } 744 745 put_child(t, tn, 2*i, (struct node *) left); 746 put_child(t, tn, 2*i+1, (struct node *) right); 747 } 748 } 749 750 for (i = 0; i < olen; i++) { 751 struct tnode *inode; 752 struct node *node = tnode_get_child(oldtnode, i); 753 struct tnode *left, *right; 754 int size, j; 755 756 /* An empty child */ 757 if (node == NULL) 758 continue; 759 760 /* A leaf or an internal node with skipped bits */ 761 762 if (IS_LEAF(node) || ((struct tnode *) node)->pos > 763 tn->pos + tn->bits - 1) { 764 if (tkey_extract_bits(node->key, 765 oldtnode->pos + oldtnode->bits, 766 1) == 0) 767 put_child(t, tn, 2*i, node); 768 else 769 put_child(t, tn, 2*i+1, node); 770 continue; 771 } 772 773 /* An internal node with two children */ 774 inode = (struct tnode *) node; 775 776 if (inode->bits == 1) { 777 put_child(t, tn, 2*i, inode->child[0]); 778 put_child(t, tn, 2*i+1, inode->child[1]); 779 780 tnode_free_safe(inode); 781 continue; 782 } 783 784 /* An internal node with more than two children */ 785 786 /* We will replace this node 'inode' with two new 787 * ones, 'left' and 'right', each with half of the 788 * original children. The two new nodes will have 789 * a position one bit further down the key and this 790 * means that the "significant" part of their keys 791 * (see the discussion near the top of this file) 792 * will differ by one bit, which will be "0" in 793 * left's key and "1" in right's key. Since we are 794 * moving the key position by one step, the bit that 795 * we are moving away from - the bit at position 796 * (inode->pos) - is the one that will differ between 797 * left and right. So... we synthesize that bit in the 798 * two new keys. 799 * The mask 'm' below will be a single "one" bit at 800 * the position (inode->pos) 801 */ 802 803 /* Use the old key, but set the new significant 804 * bit to zero. 805 */ 806 807 left = (struct tnode *) tnode_get_child(tn, 2*i); 808 put_child(t, tn, 2*i, NULL); 809 810 BUG_ON(!left); 811 812 right = (struct tnode *) tnode_get_child(tn, 2*i+1); 813 put_child(t, tn, 2*i+1, NULL); 814 815 BUG_ON(!right); 816 817 size = tnode_child_length(left); 818 for (j = 0; j < size; j++) { 819 put_child(t, left, j, inode->child[j]); 820 put_child(t, right, j, inode->child[j + size]); 821 } 822 put_child(t, tn, 2*i, resize(t, left)); 823 put_child(t, tn, 2*i+1, resize(t, right)); 824 825 tnode_free_safe(inode); 826 } 827 tnode_free_safe(oldtnode); 828 return tn; 829 nomem: 830 { 831 int size = tnode_child_length(tn); 832 int j; 833 834 for (j = 0; j < size; j++) 835 if (tn->child[j]) 836 tnode_free((struct tnode *)tn->child[j]); 837 838 tnode_free(tn); 839 840 return ERR_PTR(-ENOMEM); 841 } 842 } 843 844 static struct tnode *halve(struct trie *t, struct tnode *tn) 845 { 846 struct tnode *oldtnode = tn; 847 struct node *left, *right; 848 int i; 849 int olen = tnode_child_length(tn); 850 851 pr_debug("In halve\n"); 852 853 tn = tnode_new(oldtnode->key, oldtnode->pos, oldtnode->bits - 1); 854 855 if (!tn) 856 return ERR_PTR(-ENOMEM); 857 858 /* 859 * Preallocate and store tnodes before the actual work so we 860 * don't get into an inconsistent state if memory allocation 861 * fails. In case of failure we return the oldnode and halve 862 * of tnode is ignored. 863 */ 864 865 for (i = 0; i < olen; i += 2) { 866 left = tnode_get_child(oldtnode, i); 867 right = tnode_get_child(oldtnode, i+1); 868 869 /* Two nonempty children */ 870 if (left && right) { 871 struct tnode *newn; 872 873 newn = tnode_new(left->key, tn->pos + tn->bits, 1); 874 875 if (!newn) 876 goto nomem; 877 878 put_child(t, tn, i/2, (struct node *)newn); 879 } 880 881 } 882 883 for (i = 0; i < olen; i += 2) { 884 struct tnode *newBinNode; 885 886 left = tnode_get_child(oldtnode, i); 887 right = tnode_get_child(oldtnode, i+1); 888 889 /* At least one of the children is empty */ 890 if (left == NULL) { 891 if (right == NULL) /* Both are empty */ 892 continue; 893 put_child(t, tn, i/2, right); 894 continue; 895 } 896 897 if (right == NULL) { 898 put_child(t, tn, i/2, left); 899 continue; 900 } 901 902 /* Two nonempty children */ 903 newBinNode = (struct tnode *) tnode_get_child(tn, i/2); 904 put_child(t, tn, i/2, NULL); 905 put_child(t, newBinNode, 0, left); 906 put_child(t, newBinNode, 1, right); 907 put_child(t, tn, i/2, resize(t, newBinNode)); 908 } 909 tnode_free_safe(oldtnode); 910 return tn; 911 nomem: 912 { 913 int size = tnode_child_length(tn); 914 int j; 915 916 for (j = 0; j < size; j++) 917 if (tn->child[j]) 918 tnode_free((struct tnode *)tn->child[j]); 919 920 tnode_free(tn); 921 922 return ERR_PTR(-ENOMEM); 923 } 924 } 925 926 /* readside must use rcu_read_lock currently dump routines 927 via get_fa_head and dump */ 928 929 static struct leaf_info *find_leaf_info(struct leaf *l, int plen) 930 { 931 struct hlist_head *head = &l->list; 932 struct hlist_node *node; 933 struct leaf_info *li; 934 935 hlist_for_each_entry_rcu(li, node, head, hlist) 936 if (li->plen == plen) 937 return li; 938 939 return NULL; 940 } 941 942 static inline struct list_head *get_fa_head(struct leaf *l, int plen) 943 { 944 struct leaf_info *li = find_leaf_info(l, plen); 945 946 if (!li) 947 return NULL; 948 949 return &li->falh; 950 } 951 952 static void insert_leaf_info(struct hlist_head *head, struct leaf_info *new) 953 { 954 struct leaf_info *li = NULL, *last = NULL; 955 struct hlist_node *node; 956 957 if (hlist_empty(head)) { 958 hlist_add_head_rcu(&new->hlist, head); 959 } else { 960 hlist_for_each_entry(li, node, head, hlist) { 961 if (new->plen > li->plen) 962 break; 963 964 last = li; 965 } 966 if (last) 967 hlist_add_after_rcu(&last->hlist, &new->hlist); 968 else 969 hlist_add_before_rcu(&new->hlist, &li->hlist); 970 } 971 } 972 973 /* rcu_read_lock needs to be hold by caller from readside */ 974 975 static struct leaf * 976 fib_find_node(struct trie *t, u32 key) 977 { 978 int pos; 979 struct tnode *tn; 980 struct node *n; 981 982 pos = 0; 983 n = rcu_dereference(t->trie); 984 985 while (n != NULL && NODE_TYPE(n) == T_TNODE) { 986 tn = (struct tnode *) n; 987 988 check_tnode(tn); 989 990 if (tkey_sub_equals(tn->key, pos, tn->pos-pos, key)) { 991 pos = tn->pos + tn->bits; 992 n = tnode_get_child_rcu(tn, 993 tkey_extract_bits(key, 994 tn->pos, 995 tn->bits)); 996 } else 997 break; 998 } 999 /* Case we have found a leaf. Compare prefixes */ 1000 1001 if (n != NULL && IS_LEAF(n) && tkey_equals(key, n->key)) 1002 return (struct leaf *)n; 1003 1004 return NULL; 1005 } 1006 1007 static void trie_rebalance(struct trie *t, struct tnode *tn) 1008 { 1009 int wasfull; 1010 t_key cindex, key; 1011 struct tnode *tp; 1012 1013 key = tn->key; 1014 1015 while (tn != NULL && (tp = node_parent((struct node *)tn)) != NULL) { 1016 cindex = tkey_extract_bits(key, tp->pos, tp->bits); 1017 wasfull = tnode_full(tp, tnode_get_child(tp, cindex)); 1018 tn = (struct tnode *) resize(t, (struct tnode *)tn); 1019 1020 tnode_put_child_reorg((struct tnode *)tp, cindex, 1021 (struct node *)tn, wasfull); 1022 1023 tp = node_parent((struct node *) tn); 1024 if (!tp) 1025 rcu_assign_pointer(t->trie, (struct node *)tn); 1026 1027 tnode_free_flush(); 1028 if (!tp) 1029 break; 1030 tn = tp; 1031 } 1032 1033 /* Handle last (top) tnode */ 1034 if (IS_TNODE(tn)) 1035 tn = (struct tnode *)resize(t, (struct tnode *)tn); 1036 1037 rcu_assign_pointer(t->trie, (struct node *)tn); 1038 tnode_free_flush(); 1039 1040 return; 1041 } 1042 1043 /* only used from updater-side */ 1044 1045 static struct list_head *fib_insert_node(struct trie *t, u32 key, int plen) 1046 { 1047 int pos, newpos; 1048 struct tnode *tp = NULL, *tn = NULL; 1049 struct node *n; 1050 struct leaf *l; 1051 int missbit; 1052 struct list_head *fa_head = NULL; 1053 struct leaf_info *li; 1054 t_key cindex; 1055 1056 pos = 0; 1057 n = t->trie; 1058 1059 /* If we point to NULL, stop. Either the tree is empty and we should 1060 * just put a new leaf in if, or we have reached an empty child slot, 1061 * and we should just put our new leaf in that. 1062 * If we point to a T_TNODE, check if it matches our key. Note that 1063 * a T_TNODE might be skipping any number of bits - its 'pos' need 1064 * not be the parent's 'pos'+'bits'! 1065 * 1066 * If it does match the current key, get pos/bits from it, extract 1067 * the index from our key, push the T_TNODE and walk the tree. 1068 * 1069 * If it doesn't, we have to replace it with a new T_TNODE. 1070 * 1071 * If we point to a T_LEAF, it might or might not have the same key 1072 * as we do. If it does, just change the value, update the T_LEAF's 1073 * value, and return it. 1074 * If it doesn't, we need to replace it with a T_TNODE. 1075 */ 1076 1077 while (n != NULL && NODE_TYPE(n) == T_TNODE) { 1078 tn = (struct tnode *) n; 1079 1080 check_tnode(tn); 1081 1082 if (tkey_sub_equals(tn->key, pos, tn->pos-pos, key)) { 1083 tp = tn; 1084 pos = tn->pos + tn->bits; 1085 n = tnode_get_child(tn, 1086 tkey_extract_bits(key, 1087 tn->pos, 1088 tn->bits)); 1089 1090 BUG_ON(n && node_parent(n) != tn); 1091 } else 1092 break; 1093 } 1094 1095 /* 1096 * n ----> NULL, LEAF or TNODE 1097 * 1098 * tp is n's (parent) ----> NULL or TNODE 1099 */ 1100 1101 BUG_ON(tp && IS_LEAF(tp)); 1102 1103 /* Case 1: n is a leaf. Compare prefixes */ 1104 1105 if (n != NULL && IS_LEAF(n) && tkey_equals(key, n->key)) { 1106 l = (struct leaf *) n; 1107 li = leaf_info_new(plen); 1108 1109 if (!li) 1110 return NULL; 1111 1112 fa_head = &li->falh; 1113 insert_leaf_info(&l->list, li); 1114 goto done; 1115 } 1116 l = leaf_new(); 1117 1118 if (!l) 1119 return NULL; 1120 1121 l->key = key; 1122 li = leaf_info_new(plen); 1123 1124 if (!li) { 1125 free_leaf(l); 1126 return NULL; 1127 } 1128 1129 fa_head = &li->falh; 1130 insert_leaf_info(&l->list, li); 1131 1132 if (t->trie && n == NULL) { 1133 /* Case 2: n is NULL, and will just insert a new leaf */ 1134 1135 node_set_parent((struct node *)l, tp); 1136 1137 cindex = tkey_extract_bits(key, tp->pos, tp->bits); 1138 put_child(t, (struct tnode *)tp, cindex, (struct node *)l); 1139 } else { 1140 /* Case 3: n is a LEAF or a TNODE and the key doesn't match. */ 1141 /* 1142 * Add a new tnode here 1143 * first tnode need some special handling 1144 */ 1145 1146 if (tp) 1147 pos = tp->pos+tp->bits; 1148 else 1149 pos = 0; 1150 1151 if (n) { 1152 newpos = tkey_mismatch(key, pos, n->key); 1153 tn = tnode_new(n->key, newpos, 1); 1154 } else { 1155 newpos = 0; 1156 tn = tnode_new(key, newpos, 1); /* First tnode */ 1157 } 1158 1159 if (!tn) { 1160 free_leaf_info(li); 1161 free_leaf(l); 1162 return NULL; 1163 } 1164 1165 node_set_parent((struct node *)tn, tp); 1166 1167 missbit = tkey_extract_bits(key, newpos, 1); 1168 put_child(t, tn, missbit, (struct node *)l); 1169 put_child(t, tn, 1-missbit, n); 1170 1171 if (tp) { 1172 cindex = tkey_extract_bits(key, tp->pos, tp->bits); 1173 put_child(t, (struct tnode *)tp, cindex, 1174 (struct node *)tn); 1175 } else { 1176 rcu_assign_pointer(t->trie, (struct node *)tn); 1177 tp = tn; 1178 } 1179 } 1180 1181 if (tp && tp->pos + tp->bits > 32) 1182 pr_warning("fib_trie" 1183 " tp=%p pos=%d, bits=%d, key=%0x plen=%d\n", 1184 tp, tp->pos, tp->bits, key, plen); 1185 1186 /* Rebalance the trie */ 1187 1188 trie_rebalance(t, tp); 1189 done: 1190 return fa_head; 1191 } 1192 1193 /* 1194 * Caller must hold RTNL. 1195 */ 1196 static int fn_trie_insert(struct fib_table *tb, struct fib_config *cfg) 1197 { 1198 struct trie *t = (struct trie *) tb->tb_data; 1199 struct fib_alias *fa, *new_fa; 1200 struct list_head *fa_head = NULL; 1201 struct fib_info *fi; 1202 int plen = cfg->fc_dst_len; 1203 u8 tos = cfg->fc_tos; 1204 u32 key, mask; 1205 int err; 1206 struct leaf *l; 1207 1208 if (plen > 32) 1209 return -EINVAL; 1210 1211 key = ntohl(cfg->fc_dst); 1212 1213 pr_debug("Insert table=%u %08x/%d\n", tb->tb_id, key, plen); 1214 1215 mask = ntohl(inet_make_mask(plen)); 1216 1217 if (key & ~mask) 1218 return -EINVAL; 1219 1220 key = key & mask; 1221 1222 fi = fib_create_info(cfg); 1223 if (IS_ERR(fi)) { 1224 err = PTR_ERR(fi); 1225 goto err; 1226 } 1227 1228 l = fib_find_node(t, key); 1229 fa = NULL; 1230 1231 if (l) { 1232 fa_head = get_fa_head(l, plen); 1233 fa = fib_find_alias(fa_head, tos, fi->fib_priority); 1234 } 1235 1236 /* Now fa, if non-NULL, points to the first fib alias 1237 * with the same keys [prefix,tos,priority], if such key already 1238 * exists or to the node before which we will insert new one. 1239 * 1240 * If fa is NULL, we will need to allocate a new one and 1241 * insert to the head of f. 1242 * 1243 * If f is NULL, no fib node matched the destination key 1244 * and we need to allocate a new one of those as well. 1245 */ 1246 1247 if (fa && fa->fa_tos == tos && 1248 fa->fa_info->fib_priority == fi->fib_priority) { 1249 struct fib_alias *fa_first, *fa_match; 1250 1251 err = -EEXIST; 1252 if (cfg->fc_nlflags & NLM_F_EXCL) 1253 goto out; 1254 1255 /* We have 2 goals: 1256 * 1. Find exact match for type, scope, fib_info to avoid 1257 * duplicate routes 1258 * 2. Find next 'fa' (or head), NLM_F_APPEND inserts before it 1259 */ 1260 fa_match = NULL; 1261 fa_first = fa; 1262 fa = list_entry(fa->fa_list.prev, struct fib_alias, fa_list); 1263 list_for_each_entry_continue(fa, fa_head, fa_list) { 1264 if (fa->fa_tos != tos) 1265 break; 1266 if (fa->fa_info->fib_priority != fi->fib_priority) 1267 break; 1268 if (fa->fa_type == cfg->fc_type && 1269 fa->fa_scope == cfg->fc_scope && 1270 fa->fa_info == fi) { 1271 fa_match = fa; 1272 break; 1273 } 1274 } 1275 1276 if (cfg->fc_nlflags & NLM_F_REPLACE) { 1277 struct fib_info *fi_drop; 1278 u8 state; 1279 1280 fa = fa_first; 1281 if (fa_match) { 1282 if (fa == fa_match) 1283 err = 0; 1284 goto out; 1285 } 1286 err = -ENOBUFS; 1287 new_fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL); 1288 if (new_fa == NULL) 1289 goto out; 1290 1291 fi_drop = fa->fa_info; 1292 new_fa->fa_tos = fa->fa_tos; 1293 new_fa->fa_info = fi; 1294 new_fa->fa_type = cfg->fc_type; 1295 new_fa->fa_scope = cfg->fc_scope; 1296 state = fa->fa_state; 1297 new_fa->fa_state = state & ~FA_S_ACCESSED; 1298 1299 list_replace_rcu(&fa->fa_list, &new_fa->fa_list); 1300 alias_free_mem_rcu(fa); 1301 1302 fib_release_info(fi_drop); 1303 if (state & FA_S_ACCESSED) 1304 rt_cache_flush(cfg->fc_nlinfo.nl_net, -1); 1305 rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen, 1306 tb->tb_id, &cfg->fc_nlinfo, NLM_F_REPLACE); 1307 1308 goto succeeded; 1309 } 1310 /* Error if we find a perfect match which 1311 * uses the same scope, type, and nexthop 1312 * information. 1313 */ 1314 if (fa_match) 1315 goto out; 1316 1317 if (!(cfg->fc_nlflags & NLM_F_APPEND)) 1318 fa = fa_first; 1319 } 1320 err = -ENOENT; 1321 if (!(cfg->fc_nlflags & NLM_F_CREATE)) 1322 goto out; 1323 1324 err = -ENOBUFS; 1325 new_fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL); 1326 if (new_fa == NULL) 1327 goto out; 1328 1329 new_fa->fa_info = fi; 1330 new_fa->fa_tos = tos; 1331 new_fa->fa_type = cfg->fc_type; 1332 new_fa->fa_scope = cfg->fc_scope; 1333 new_fa->fa_state = 0; 1334 /* 1335 * Insert new entry to the list. 1336 */ 1337 1338 if (!fa_head) { 1339 fa_head = fib_insert_node(t, key, plen); 1340 if (unlikely(!fa_head)) { 1341 err = -ENOMEM; 1342 goto out_free_new_fa; 1343 } 1344 } 1345 1346 list_add_tail_rcu(&new_fa->fa_list, 1347 (fa ? &fa->fa_list : fa_head)); 1348 1349 rt_cache_flush(cfg->fc_nlinfo.nl_net, -1); 1350 rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen, tb->tb_id, 1351 &cfg->fc_nlinfo, 0); 1352 succeeded: 1353 return 0; 1354 1355 out_free_new_fa: 1356 kmem_cache_free(fn_alias_kmem, new_fa); 1357 out: 1358 fib_release_info(fi); 1359 err: 1360 return err; 1361 } 1362 1363 /* should be called with rcu_read_lock */ 1364 static int check_leaf(struct trie *t, struct leaf *l, 1365 t_key key, const struct flowi *flp, 1366 struct fib_result *res) 1367 { 1368 struct leaf_info *li; 1369 struct hlist_head *hhead = &l->list; 1370 struct hlist_node *node; 1371 1372 hlist_for_each_entry_rcu(li, node, hhead, hlist) { 1373 int err; 1374 int plen = li->plen; 1375 __be32 mask = inet_make_mask(plen); 1376 1377 if (l->key != (key & ntohl(mask))) 1378 continue; 1379 1380 err = fib_semantic_match(&li->falh, flp, res, plen); 1381 1382 #ifdef CONFIG_IP_FIB_TRIE_STATS 1383 if (err <= 0) 1384 t->stats.semantic_match_passed++; 1385 else 1386 t->stats.semantic_match_miss++; 1387 #endif 1388 if (err <= 0) 1389 return err; 1390 } 1391 1392 return 1; 1393 } 1394 1395 static int fn_trie_lookup(struct fib_table *tb, const struct flowi *flp, 1396 struct fib_result *res) 1397 { 1398 struct trie *t = (struct trie *) tb->tb_data; 1399 int ret; 1400 struct node *n; 1401 struct tnode *pn; 1402 int pos, bits; 1403 t_key key = ntohl(flp->fl4_dst); 1404 int chopped_off; 1405 t_key cindex = 0; 1406 int current_prefix_length = KEYLENGTH; 1407 struct tnode *cn; 1408 t_key node_prefix, key_prefix, pref_mismatch; 1409 int mp; 1410 1411 rcu_read_lock(); 1412 1413 n = rcu_dereference(t->trie); 1414 if (!n) 1415 goto failed; 1416 1417 #ifdef CONFIG_IP_FIB_TRIE_STATS 1418 t->stats.gets++; 1419 #endif 1420 1421 /* Just a leaf? */ 1422 if (IS_LEAF(n)) { 1423 ret = check_leaf(t, (struct leaf *)n, key, flp, res); 1424 goto found; 1425 } 1426 1427 pn = (struct tnode *) n; 1428 chopped_off = 0; 1429 1430 while (pn) { 1431 pos = pn->pos; 1432 bits = pn->bits; 1433 1434 if (!chopped_off) 1435 cindex = tkey_extract_bits(mask_pfx(key, current_prefix_length), 1436 pos, bits); 1437 1438 n = tnode_get_child(pn, cindex); 1439 1440 if (n == NULL) { 1441 #ifdef CONFIG_IP_FIB_TRIE_STATS 1442 t->stats.null_node_hit++; 1443 #endif 1444 goto backtrace; 1445 } 1446 1447 if (IS_LEAF(n)) { 1448 ret = check_leaf(t, (struct leaf *)n, key, flp, res); 1449 if (ret > 0) 1450 goto backtrace; 1451 goto found; 1452 } 1453 1454 cn = (struct tnode *)n; 1455 1456 /* 1457 * It's a tnode, and we can do some extra checks here if we 1458 * like, to avoid descending into a dead-end branch. 1459 * This tnode is in the parent's child array at index 1460 * key[p_pos..p_pos+p_bits] but potentially with some bits 1461 * chopped off, so in reality the index may be just a 1462 * subprefix, padded with zero at the end. 1463 * We can also take a look at any skipped bits in this 1464 * tnode - everything up to p_pos is supposed to be ok, 1465 * and the non-chopped bits of the index (se previous 1466 * paragraph) are also guaranteed ok, but the rest is 1467 * considered unknown. 1468 * 1469 * The skipped bits are key[pos+bits..cn->pos]. 1470 */ 1471 1472 /* If current_prefix_length < pos+bits, we are already doing 1473 * actual prefix matching, which means everything from 1474 * pos+(bits-chopped_off) onward must be zero along some 1475 * branch of this subtree - otherwise there is *no* valid 1476 * prefix present. Here we can only check the skipped 1477 * bits. Remember, since we have already indexed into the 1478 * parent's child array, we know that the bits we chopped of 1479 * *are* zero. 1480 */ 1481 1482 /* NOTA BENE: Checking only skipped bits 1483 for the new node here */ 1484 1485 if (current_prefix_length < pos+bits) { 1486 if (tkey_extract_bits(cn->key, current_prefix_length, 1487 cn->pos - current_prefix_length) 1488 || !(cn->child[0])) 1489 goto backtrace; 1490 } 1491 1492 /* 1493 * If chopped_off=0, the index is fully validated and we 1494 * only need to look at the skipped bits for this, the new, 1495 * tnode. What we actually want to do is to find out if 1496 * these skipped bits match our key perfectly, or if we will 1497 * have to count on finding a matching prefix further down, 1498 * because if we do, we would like to have some way of 1499 * verifying the existence of such a prefix at this point. 1500 */ 1501 1502 /* The only thing we can do at this point is to verify that 1503 * any such matching prefix can indeed be a prefix to our 1504 * key, and if the bits in the node we are inspecting that 1505 * do not match our key are not ZERO, this cannot be true. 1506 * Thus, find out where there is a mismatch (before cn->pos) 1507 * and verify that all the mismatching bits are zero in the 1508 * new tnode's key. 1509 */ 1510 1511 /* 1512 * Note: We aren't very concerned about the piece of 1513 * the key that precede pn->pos+pn->bits, since these 1514 * have already been checked. The bits after cn->pos 1515 * aren't checked since these are by definition 1516 * "unknown" at this point. Thus, what we want to see 1517 * is if we are about to enter the "prefix matching" 1518 * state, and in that case verify that the skipped 1519 * bits that will prevail throughout this subtree are 1520 * zero, as they have to be if we are to find a 1521 * matching prefix. 1522 */ 1523 1524 node_prefix = mask_pfx(cn->key, cn->pos); 1525 key_prefix = mask_pfx(key, cn->pos); 1526 pref_mismatch = key_prefix^node_prefix; 1527 mp = 0; 1528 1529 /* 1530 * In short: If skipped bits in this node do not match 1531 * the search key, enter the "prefix matching" 1532 * state.directly. 1533 */ 1534 if (pref_mismatch) { 1535 while (!(pref_mismatch & (1<<(KEYLENGTH-1)))) { 1536 mp++; 1537 pref_mismatch = pref_mismatch << 1; 1538 } 1539 key_prefix = tkey_extract_bits(cn->key, mp, cn->pos-mp); 1540 1541 if (key_prefix != 0) 1542 goto backtrace; 1543 1544 if (current_prefix_length >= cn->pos) 1545 current_prefix_length = mp; 1546 } 1547 1548 pn = (struct tnode *)n; /* Descend */ 1549 chopped_off = 0; 1550 continue; 1551 1552 backtrace: 1553 chopped_off++; 1554 1555 /* As zero don't change the child key (cindex) */ 1556 while ((chopped_off <= pn->bits) 1557 && !(cindex & (1<<(chopped_off-1)))) 1558 chopped_off++; 1559 1560 /* Decrease current_... with bits chopped off */ 1561 if (current_prefix_length > pn->pos + pn->bits - chopped_off) 1562 current_prefix_length = pn->pos + pn->bits 1563 - chopped_off; 1564 1565 /* 1566 * Either we do the actual chop off according or if we have 1567 * chopped off all bits in this tnode walk up to our parent. 1568 */ 1569 1570 if (chopped_off <= pn->bits) { 1571 cindex &= ~(1 << (chopped_off-1)); 1572 } else { 1573 struct tnode *parent = node_parent((struct node *) pn); 1574 if (!parent) 1575 goto failed; 1576 1577 /* Get Child's index */ 1578 cindex = tkey_extract_bits(pn->key, parent->pos, parent->bits); 1579 pn = parent; 1580 chopped_off = 0; 1581 1582 #ifdef CONFIG_IP_FIB_TRIE_STATS 1583 t->stats.backtrack++; 1584 #endif 1585 goto backtrace; 1586 } 1587 } 1588 failed: 1589 ret = 1; 1590 found: 1591 rcu_read_unlock(); 1592 return ret; 1593 } 1594 1595 /* 1596 * Remove the leaf and return parent. 1597 */ 1598 static void trie_leaf_remove(struct trie *t, struct leaf *l) 1599 { 1600 struct tnode *tp = node_parent((struct node *) l); 1601 1602 pr_debug("entering trie_leaf_remove(%p)\n", l); 1603 1604 if (tp) { 1605 t_key cindex = tkey_extract_bits(l->key, tp->pos, tp->bits); 1606 put_child(t, (struct tnode *)tp, cindex, NULL); 1607 trie_rebalance(t, tp); 1608 } else 1609 rcu_assign_pointer(t->trie, NULL); 1610 1611 free_leaf(l); 1612 } 1613 1614 /* 1615 * Caller must hold RTNL. 1616 */ 1617 static int fn_trie_delete(struct fib_table *tb, struct fib_config *cfg) 1618 { 1619 struct trie *t = (struct trie *) tb->tb_data; 1620 u32 key, mask; 1621 int plen = cfg->fc_dst_len; 1622 u8 tos = cfg->fc_tos; 1623 struct fib_alias *fa, *fa_to_delete; 1624 struct list_head *fa_head; 1625 struct leaf *l; 1626 struct leaf_info *li; 1627 1628 if (plen > 32) 1629 return -EINVAL; 1630 1631 key = ntohl(cfg->fc_dst); 1632 mask = ntohl(inet_make_mask(plen)); 1633 1634 if (key & ~mask) 1635 return -EINVAL; 1636 1637 key = key & mask; 1638 l = fib_find_node(t, key); 1639 1640 if (!l) 1641 return -ESRCH; 1642 1643 fa_head = get_fa_head(l, plen); 1644 fa = fib_find_alias(fa_head, tos, 0); 1645 1646 if (!fa) 1647 return -ESRCH; 1648 1649 pr_debug("Deleting %08x/%d tos=%d t=%p\n", key, plen, tos, t); 1650 1651 fa_to_delete = NULL; 1652 fa = list_entry(fa->fa_list.prev, struct fib_alias, fa_list); 1653 list_for_each_entry_continue(fa, fa_head, fa_list) { 1654 struct fib_info *fi = fa->fa_info; 1655 1656 if (fa->fa_tos != tos) 1657 break; 1658 1659 if ((!cfg->fc_type || fa->fa_type == cfg->fc_type) && 1660 (cfg->fc_scope == RT_SCOPE_NOWHERE || 1661 fa->fa_scope == cfg->fc_scope) && 1662 (!cfg->fc_protocol || 1663 fi->fib_protocol == cfg->fc_protocol) && 1664 fib_nh_match(cfg, fi) == 0) { 1665 fa_to_delete = fa; 1666 break; 1667 } 1668 } 1669 1670 if (!fa_to_delete) 1671 return -ESRCH; 1672 1673 fa = fa_to_delete; 1674 rtmsg_fib(RTM_DELROUTE, htonl(key), fa, plen, tb->tb_id, 1675 &cfg->fc_nlinfo, 0); 1676 1677 l = fib_find_node(t, key); 1678 li = find_leaf_info(l, plen); 1679 1680 list_del_rcu(&fa->fa_list); 1681 1682 if (list_empty(fa_head)) { 1683 hlist_del_rcu(&li->hlist); 1684 free_leaf_info(li); 1685 } 1686 1687 if (hlist_empty(&l->list)) 1688 trie_leaf_remove(t, l); 1689 1690 if (fa->fa_state & FA_S_ACCESSED) 1691 rt_cache_flush(cfg->fc_nlinfo.nl_net, -1); 1692 1693 fib_release_info(fa->fa_info); 1694 alias_free_mem_rcu(fa); 1695 return 0; 1696 } 1697 1698 static int trie_flush_list(struct list_head *head) 1699 { 1700 struct fib_alias *fa, *fa_node; 1701 int found = 0; 1702 1703 list_for_each_entry_safe(fa, fa_node, head, fa_list) { 1704 struct fib_info *fi = fa->fa_info; 1705 1706 if (fi && (fi->fib_flags & RTNH_F_DEAD)) { 1707 list_del_rcu(&fa->fa_list); 1708 fib_release_info(fa->fa_info); 1709 alias_free_mem_rcu(fa); 1710 found++; 1711 } 1712 } 1713 return found; 1714 } 1715 1716 static int trie_flush_leaf(struct leaf *l) 1717 { 1718 int found = 0; 1719 struct hlist_head *lih = &l->list; 1720 struct hlist_node *node, *tmp; 1721 struct leaf_info *li = NULL; 1722 1723 hlist_for_each_entry_safe(li, node, tmp, lih, hlist) { 1724 found += trie_flush_list(&li->falh); 1725 1726 if (list_empty(&li->falh)) { 1727 hlist_del_rcu(&li->hlist); 1728 free_leaf_info(li); 1729 } 1730 } 1731 return found; 1732 } 1733 1734 /* 1735 * Scan for the next right leaf starting at node p->child[idx] 1736 * Since we have back pointer, no recursion necessary. 1737 */ 1738 static struct leaf *leaf_walk_rcu(struct tnode *p, struct node *c) 1739 { 1740 do { 1741 t_key idx; 1742 1743 if (c) 1744 idx = tkey_extract_bits(c->key, p->pos, p->bits) + 1; 1745 else 1746 idx = 0; 1747 1748 while (idx < 1u << p->bits) { 1749 c = tnode_get_child_rcu(p, idx++); 1750 if (!c) 1751 continue; 1752 1753 if (IS_LEAF(c)) { 1754 prefetch(p->child[idx]); 1755 return (struct leaf *) c; 1756 } 1757 1758 /* Rescan start scanning in new node */ 1759 p = (struct tnode *) c; 1760 idx = 0; 1761 } 1762 1763 /* Node empty, walk back up to parent */ 1764 c = (struct node *) p; 1765 } while ( (p = node_parent_rcu(c)) != NULL); 1766 1767 return NULL; /* Root of trie */ 1768 } 1769 1770 static struct leaf *trie_firstleaf(struct trie *t) 1771 { 1772 struct tnode *n = (struct tnode *) rcu_dereference(t->trie); 1773 1774 if (!n) 1775 return NULL; 1776 1777 if (IS_LEAF(n)) /* trie is just a leaf */ 1778 return (struct leaf *) n; 1779 1780 return leaf_walk_rcu(n, NULL); 1781 } 1782 1783 static struct leaf *trie_nextleaf(struct leaf *l) 1784 { 1785 struct node *c = (struct node *) l; 1786 struct tnode *p = node_parent(c); 1787 1788 if (!p) 1789 return NULL; /* trie with just one leaf */ 1790 1791 return leaf_walk_rcu(p, c); 1792 } 1793 1794 static struct leaf *trie_leafindex(struct trie *t, int index) 1795 { 1796 struct leaf *l = trie_firstleaf(t); 1797 1798 while (l && index-- > 0) 1799 l = trie_nextleaf(l); 1800 1801 return l; 1802 } 1803 1804 1805 /* 1806 * Caller must hold RTNL. 1807 */ 1808 static int fn_trie_flush(struct fib_table *tb) 1809 { 1810 struct trie *t = (struct trie *) tb->tb_data; 1811 struct leaf *l, *ll = NULL; 1812 int found = 0; 1813 1814 for (l = trie_firstleaf(t); l; l = trie_nextleaf(l)) { 1815 found += trie_flush_leaf(l); 1816 1817 if (ll && hlist_empty(&ll->list)) 1818 trie_leaf_remove(t, ll); 1819 ll = l; 1820 } 1821 1822 if (ll && hlist_empty(&ll->list)) 1823 trie_leaf_remove(t, ll); 1824 1825 pr_debug("trie_flush found=%d\n", found); 1826 return found; 1827 } 1828 1829 static void fn_trie_select_default(struct fib_table *tb, 1830 const struct flowi *flp, 1831 struct fib_result *res) 1832 { 1833 struct trie *t = (struct trie *) tb->tb_data; 1834 int order, last_idx; 1835 struct fib_info *fi = NULL; 1836 struct fib_info *last_resort; 1837 struct fib_alias *fa = NULL; 1838 struct list_head *fa_head; 1839 struct leaf *l; 1840 1841 last_idx = -1; 1842 last_resort = NULL; 1843 order = -1; 1844 1845 rcu_read_lock(); 1846 1847 l = fib_find_node(t, 0); 1848 if (!l) 1849 goto out; 1850 1851 fa_head = get_fa_head(l, 0); 1852 if (!fa_head) 1853 goto out; 1854 1855 if (list_empty(fa_head)) 1856 goto out; 1857 1858 list_for_each_entry_rcu(fa, fa_head, fa_list) { 1859 struct fib_info *next_fi = fa->fa_info; 1860 1861 if (fa->fa_scope != res->scope || 1862 fa->fa_type != RTN_UNICAST) 1863 continue; 1864 1865 if (next_fi->fib_priority > res->fi->fib_priority) 1866 break; 1867 if (!next_fi->fib_nh[0].nh_gw || 1868 next_fi->fib_nh[0].nh_scope != RT_SCOPE_LINK) 1869 continue; 1870 fa->fa_state |= FA_S_ACCESSED; 1871 1872 if (fi == NULL) { 1873 if (next_fi != res->fi) 1874 break; 1875 } else if (!fib_detect_death(fi, order, &last_resort, 1876 &last_idx, tb->tb_default)) { 1877 fib_result_assign(res, fi); 1878 tb->tb_default = order; 1879 goto out; 1880 } 1881 fi = next_fi; 1882 order++; 1883 } 1884 if (order <= 0 || fi == NULL) { 1885 tb->tb_default = -1; 1886 goto out; 1887 } 1888 1889 if (!fib_detect_death(fi, order, &last_resort, &last_idx, 1890 tb->tb_default)) { 1891 fib_result_assign(res, fi); 1892 tb->tb_default = order; 1893 goto out; 1894 } 1895 if (last_idx >= 0) 1896 fib_result_assign(res, last_resort); 1897 tb->tb_default = last_idx; 1898 out: 1899 rcu_read_unlock(); 1900 } 1901 1902 static int fn_trie_dump_fa(t_key key, int plen, struct list_head *fah, 1903 struct fib_table *tb, 1904 struct sk_buff *skb, struct netlink_callback *cb) 1905 { 1906 int i, s_i; 1907 struct fib_alias *fa; 1908 __be32 xkey = htonl(key); 1909 1910 s_i = cb->args[5]; 1911 i = 0; 1912 1913 /* rcu_read_lock is hold by caller */ 1914 1915 list_for_each_entry_rcu(fa, fah, fa_list) { 1916 if (i < s_i) { 1917 i++; 1918 continue; 1919 } 1920 1921 if (fib_dump_info(skb, NETLINK_CB(cb->skb).pid, 1922 cb->nlh->nlmsg_seq, 1923 RTM_NEWROUTE, 1924 tb->tb_id, 1925 fa->fa_type, 1926 fa->fa_scope, 1927 xkey, 1928 plen, 1929 fa->fa_tos, 1930 fa->fa_info, NLM_F_MULTI) < 0) { 1931 cb->args[5] = i; 1932 return -1; 1933 } 1934 i++; 1935 } 1936 cb->args[5] = i; 1937 return skb->len; 1938 } 1939 1940 static int fn_trie_dump_leaf(struct leaf *l, struct fib_table *tb, 1941 struct sk_buff *skb, struct netlink_callback *cb) 1942 { 1943 struct leaf_info *li; 1944 struct hlist_node *node; 1945 int i, s_i; 1946 1947 s_i = cb->args[4]; 1948 i = 0; 1949 1950 /* rcu_read_lock is hold by caller */ 1951 hlist_for_each_entry_rcu(li, node, &l->list, hlist) { 1952 if (i < s_i) { 1953 i++; 1954 continue; 1955 } 1956 1957 if (i > s_i) 1958 cb->args[5] = 0; 1959 1960 if (list_empty(&li->falh)) 1961 continue; 1962 1963 if (fn_trie_dump_fa(l->key, li->plen, &li->falh, tb, skb, cb) < 0) { 1964 cb->args[4] = i; 1965 return -1; 1966 } 1967 i++; 1968 } 1969 1970 cb->args[4] = i; 1971 return skb->len; 1972 } 1973 1974 static int fn_trie_dump(struct fib_table *tb, struct sk_buff *skb, 1975 struct netlink_callback *cb) 1976 { 1977 struct leaf *l; 1978 struct trie *t = (struct trie *) tb->tb_data; 1979 t_key key = cb->args[2]; 1980 int count = cb->args[3]; 1981 1982 rcu_read_lock(); 1983 /* Dump starting at last key. 1984 * Note: 0.0.0.0/0 (ie default) is first key. 1985 */ 1986 if (count == 0) 1987 l = trie_firstleaf(t); 1988 else { 1989 /* Normally, continue from last key, but if that is missing 1990 * fallback to using slow rescan 1991 */ 1992 l = fib_find_node(t, key); 1993 if (!l) 1994 l = trie_leafindex(t, count); 1995 } 1996 1997 while (l) { 1998 cb->args[2] = l->key; 1999 if (fn_trie_dump_leaf(l, tb, skb, cb) < 0) { 2000 cb->args[3] = count; 2001 rcu_read_unlock(); 2002 return -1; 2003 } 2004 2005 ++count; 2006 l = trie_nextleaf(l); 2007 memset(&cb->args[4], 0, 2008 sizeof(cb->args) - 4*sizeof(cb->args[0])); 2009 } 2010 cb->args[3] = count; 2011 rcu_read_unlock(); 2012 2013 return skb->len; 2014 } 2015 2016 void __init fib_hash_init(void) 2017 { 2018 fn_alias_kmem = kmem_cache_create("ip_fib_alias", 2019 sizeof(struct fib_alias), 2020 0, SLAB_PANIC, NULL); 2021 2022 trie_leaf_kmem = kmem_cache_create("ip_fib_trie", 2023 max(sizeof(struct leaf), 2024 sizeof(struct leaf_info)), 2025 0, SLAB_PANIC, NULL); 2026 } 2027 2028 2029 /* Fix more generic FIB names for init later */ 2030 struct fib_table *fib_hash_table(u32 id) 2031 { 2032 struct fib_table *tb; 2033 struct trie *t; 2034 2035 tb = kmalloc(sizeof(struct fib_table) + sizeof(struct trie), 2036 GFP_KERNEL); 2037 if (tb == NULL) 2038 return NULL; 2039 2040 tb->tb_id = id; 2041 tb->tb_default = -1; 2042 tb->tb_lookup = fn_trie_lookup; 2043 tb->tb_insert = fn_trie_insert; 2044 tb->tb_delete = fn_trie_delete; 2045 tb->tb_flush = fn_trie_flush; 2046 tb->tb_select_default = fn_trie_select_default; 2047 tb->tb_dump = fn_trie_dump; 2048 2049 t = (struct trie *) tb->tb_data; 2050 memset(t, 0, sizeof(*t)); 2051 2052 if (id == RT_TABLE_LOCAL) 2053 pr_info("IPv4 FIB: Using LC-trie version %s\n", VERSION); 2054 2055 return tb; 2056 } 2057 2058 #ifdef CONFIG_PROC_FS 2059 /* Depth first Trie walk iterator */ 2060 struct fib_trie_iter { 2061 struct seq_net_private p; 2062 struct fib_table *tb; 2063 struct tnode *tnode; 2064 unsigned index; 2065 unsigned depth; 2066 }; 2067 2068 static struct node *fib_trie_get_next(struct fib_trie_iter *iter) 2069 { 2070 struct tnode *tn = iter->tnode; 2071 unsigned cindex = iter->index; 2072 struct tnode *p; 2073 2074 /* A single entry routing table */ 2075 if (!tn) 2076 return NULL; 2077 2078 pr_debug("get_next iter={node=%p index=%d depth=%d}\n", 2079 iter->tnode, iter->index, iter->depth); 2080 rescan: 2081 while (cindex < (1<<tn->bits)) { 2082 struct node *n = tnode_get_child_rcu(tn, cindex); 2083 2084 if (n) { 2085 if (IS_LEAF(n)) { 2086 iter->tnode = tn; 2087 iter->index = cindex + 1; 2088 } else { 2089 /* push down one level */ 2090 iter->tnode = (struct tnode *) n; 2091 iter->index = 0; 2092 ++iter->depth; 2093 } 2094 return n; 2095 } 2096 2097 ++cindex; 2098 } 2099 2100 /* Current node exhausted, pop back up */ 2101 p = node_parent_rcu((struct node *)tn); 2102 if (p) { 2103 cindex = tkey_extract_bits(tn->key, p->pos, p->bits)+1; 2104 tn = p; 2105 --iter->depth; 2106 goto rescan; 2107 } 2108 2109 /* got root? */ 2110 return NULL; 2111 } 2112 2113 static struct node *fib_trie_get_first(struct fib_trie_iter *iter, 2114 struct trie *t) 2115 { 2116 struct node *n; 2117 2118 if (!t) 2119 return NULL; 2120 2121 n = rcu_dereference(t->trie); 2122 if (!n) 2123 return NULL; 2124 2125 if (IS_TNODE(n)) { 2126 iter->tnode = (struct tnode *) n; 2127 iter->index = 0; 2128 iter->depth = 1; 2129 } else { 2130 iter->tnode = NULL; 2131 iter->index = 0; 2132 iter->depth = 0; 2133 } 2134 2135 return n; 2136 } 2137 2138 static void trie_collect_stats(struct trie *t, struct trie_stat *s) 2139 { 2140 struct node *n; 2141 struct fib_trie_iter iter; 2142 2143 memset(s, 0, sizeof(*s)); 2144 2145 rcu_read_lock(); 2146 for (n = fib_trie_get_first(&iter, t); n; n = fib_trie_get_next(&iter)) { 2147 if (IS_LEAF(n)) { 2148 struct leaf *l = (struct leaf *)n; 2149 struct leaf_info *li; 2150 struct hlist_node *tmp; 2151 2152 s->leaves++; 2153 s->totdepth += iter.depth; 2154 if (iter.depth > s->maxdepth) 2155 s->maxdepth = iter.depth; 2156 2157 hlist_for_each_entry_rcu(li, tmp, &l->list, hlist) 2158 ++s->prefixes; 2159 } else { 2160 const struct tnode *tn = (const struct tnode *) n; 2161 int i; 2162 2163 s->tnodes++; 2164 if (tn->bits < MAX_STAT_DEPTH) 2165 s->nodesizes[tn->bits]++; 2166 2167 for (i = 0; i < (1<<tn->bits); i++) 2168 if (!tn->child[i]) 2169 s->nullpointers++; 2170 } 2171 } 2172 rcu_read_unlock(); 2173 } 2174 2175 /* 2176 * This outputs /proc/net/fib_triestats 2177 */ 2178 static void trie_show_stats(struct seq_file *seq, struct trie_stat *stat) 2179 { 2180 unsigned i, max, pointers, bytes, avdepth; 2181 2182 if (stat->leaves) 2183 avdepth = stat->totdepth*100 / stat->leaves; 2184 else 2185 avdepth = 0; 2186 2187 seq_printf(seq, "\tAver depth: %u.%02d\n", 2188 avdepth / 100, avdepth % 100); 2189 seq_printf(seq, "\tMax depth: %u\n", stat->maxdepth); 2190 2191 seq_printf(seq, "\tLeaves: %u\n", stat->leaves); 2192 bytes = sizeof(struct leaf) * stat->leaves; 2193 2194 seq_printf(seq, "\tPrefixes: %u\n", stat->prefixes); 2195 bytes += sizeof(struct leaf_info) * stat->prefixes; 2196 2197 seq_printf(seq, "\tInternal nodes: %u\n\t", stat->tnodes); 2198 bytes += sizeof(struct tnode) * stat->tnodes; 2199 2200 max = MAX_STAT_DEPTH; 2201 while (max > 0 && stat->nodesizes[max-1] == 0) 2202 max--; 2203 2204 pointers = 0; 2205 for (i = 1; i <= max; i++) 2206 if (stat->nodesizes[i] != 0) { 2207 seq_printf(seq, " %u: %u", i, stat->nodesizes[i]); 2208 pointers += (1<<i) * stat->nodesizes[i]; 2209 } 2210 seq_putc(seq, '\n'); 2211 seq_printf(seq, "\tPointers: %u\n", pointers); 2212 2213 bytes += sizeof(struct node *) * pointers; 2214 seq_printf(seq, "Null ptrs: %u\n", stat->nullpointers); 2215 seq_printf(seq, "Total size: %u kB\n", (bytes + 1023) / 1024); 2216 } 2217 2218 #ifdef CONFIG_IP_FIB_TRIE_STATS 2219 static void trie_show_usage(struct seq_file *seq, 2220 const struct trie_use_stats *stats) 2221 { 2222 seq_printf(seq, "\nCounters:\n---------\n"); 2223 seq_printf(seq, "gets = %u\n", stats->gets); 2224 seq_printf(seq, "backtracks = %u\n", stats->backtrack); 2225 seq_printf(seq, "semantic match passed = %u\n", 2226 stats->semantic_match_passed); 2227 seq_printf(seq, "semantic match miss = %u\n", 2228 stats->semantic_match_miss); 2229 seq_printf(seq, "null node hit= %u\n", stats->null_node_hit); 2230 seq_printf(seq, "skipped node resize = %u\n\n", 2231 stats->resize_node_skipped); 2232 } 2233 #endif /* CONFIG_IP_FIB_TRIE_STATS */ 2234 2235 static void fib_table_print(struct seq_file *seq, struct fib_table *tb) 2236 { 2237 if (tb->tb_id == RT_TABLE_LOCAL) 2238 seq_puts(seq, "Local:\n"); 2239 else if (tb->tb_id == RT_TABLE_MAIN) 2240 seq_puts(seq, "Main:\n"); 2241 else 2242 seq_printf(seq, "Id %d:\n", tb->tb_id); 2243 } 2244 2245 2246 static int fib_triestat_seq_show(struct seq_file *seq, void *v) 2247 { 2248 struct net *net = (struct net *)seq->private; 2249 unsigned int h; 2250 2251 seq_printf(seq, 2252 "Basic info: size of leaf:" 2253 " %Zd bytes, size of tnode: %Zd bytes.\n", 2254 sizeof(struct leaf), sizeof(struct tnode)); 2255 2256 for (h = 0; h < FIB_TABLE_HASHSZ; h++) { 2257 struct hlist_head *head = &net->ipv4.fib_table_hash[h]; 2258 struct hlist_node *node; 2259 struct fib_table *tb; 2260 2261 hlist_for_each_entry_rcu(tb, node, head, tb_hlist) { 2262 struct trie *t = (struct trie *) tb->tb_data; 2263 struct trie_stat stat; 2264 2265 if (!t) 2266 continue; 2267 2268 fib_table_print(seq, tb); 2269 2270 trie_collect_stats(t, &stat); 2271 trie_show_stats(seq, &stat); 2272 #ifdef CONFIG_IP_FIB_TRIE_STATS 2273 trie_show_usage(seq, &t->stats); 2274 #endif 2275 } 2276 } 2277 2278 return 0; 2279 } 2280 2281 static int fib_triestat_seq_open(struct inode *inode, struct file *file) 2282 { 2283 return single_open_net(inode, file, fib_triestat_seq_show); 2284 } 2285 2286 static const struct file_operations fib_triestat_fops = { 2287 .owner = THIS_MODULE, 2288 .open = fib_triestat_seq_open, 2289 .read = seq_read, 2290 .llseek = seq_lseek, 2291 .release = single_release_net, 2292 }; 2293 2294 static struct node *fib_trie_get_idx(struct seq_file *seq, loff_t pos) 2295 { 2296 struct fib_trie_iter *iter = seq->private; 2297 struct net *net = seq_file_net(seq); 2298 loff_t idx = 0; 2299 unsigned int h; 2300 2301 for (h = 0; h < FIB_TABLE_HASHSZ; h++) { 2302 struct hlist_head *head = &net->ipv4.fib_table_hash[h]; 2303 struct hlist_node *node; 2304 struct fib_table *tb; 2305 2306 hlist_for_each_entry_rcu(tb, node, head, tb_hlist) { 2307 struct node *n; 2308 2309 for (n = fib_trie_get_first(iter, 2310 (struct trie *) tb->tb_data); 2311 n; n = fib_trie_get_next(iter)) 2312 if (pos == idx++) { 2313 iter->tb = tb; 2314 return n; 2315 } 2316 } 2317 } 2318 2319 return NULL; 2320 } 2321 2322 static void *fib_trie_seq_start(struct seq_file *seq, loff_t *pos) 2323 __acquires(RCU) 2324 { 2325 rcu_read_lock(); 2326 return fib_trie_get_idx(seq, *pos); 2327 } 2328 2329 static void *fib_trie_seq_next(struct seq_file *seq, void *v, loff_t *pos) 2330 { 2331 struct fib_trie_iter *iter = seq->private; 2332 struct net *net = seq_file_net(seq); 2333 struct fib_table *tb = iter->tb; 2334 struct hlist_node *tb_node; 2335 unsigned int h; 2336 struct node *n; 2337 2338 ++*pos; 2339 /* next node in same table */ 2340 n = fib_trie_get_next(iter); 2341 if (n) 2342 return n; 2343 2344 /* walk rest of this hash chain */ 2345 h = tb->tb_id & (FIB_TABLE_HASHSZ - 1); 2346 while ( (tb_node = rcu_dereference(tb->tb_hlist.next)) ) { 2347 tb = hlist_entry(tb_node, struct fib_table, tb_hlist); 2348 n = fib_trie_get_first(iter, (struct trie *) tb->tb_data); 2349 if (n) 2350 goto found; 2351 } 2352 2353 /* new hash chain */ 2354 while (++h < FIB_TABLE_HASHSZ) { 2355 struct hlist_head *head = &net->ipv4.fib_table_hash[h]; 2356 hlist_for_each_entry_rcu(tb, tb_node, head, tb_hlist) { 2357 n = fib_trie_get_first(iter, (struct trie *) tb->tb_data); 2358 if (n) 2359 goto found; 2360 } 2361 } 2362 return NULL; 2363 2364 found: 2365 iter->tb = tb; 2366 return n; 2367 } 2368 2369 static void fib_trie_seq_stop(struct seq_file *seq, void *v) 2370 __releases(RCU) 2371 { 2372 rcu_read_unlock(); 2373 } 2374 2375 static void seq_indent(struct seq_file *seq, int n) 2376 { 2377 while (n-- > 0) seq_puts(seq, " "); 2378 } 2379 2380 static inline const char *rtn_scope(char *buf, size_t len, enum rt_scope_t s) 2381 { 2382 switch (s) { 2383 case RT_SCOPE_UNIVERSE: return "universe"; 2384 case RT_SCOPE_SITE: return "site"; 2385 case RT_SCOPE_LINK: return "link"; 2386 case RT_SCOPE_HOST: return "host"; 2387 case RT_SCOPE_NOWHERE: return "nowhere"; 2388 default: 2389 snprintf(buf, len, "scope=%d", s); 2390 return buf; 2391 } 2392 } 2393 2394 static const char *rtn_type_names[__RTN_MAX] = { 2395 [RTN_UNSPEC] = "UNSPEC", 2396 [RTN_UNICAST] = "UNICAST", 2397 [RTN_LOCAL] = "LOCAL", 2398 [RTN_BROADCAST] = "BROADCAST", 2399 [RTN_ANYCAST] = "ANYCAST", 2400 [RTN_MULTICAST] = "MULTICAST", 2401 [RTN_BLACKHOLE] = "BLACKHOLE", 2402 [RTN_UNREACHABLE] = "UNREACHABLE", 2403 [RTN_PROHIBIT] = "PROHIBIT", 2404 [RTN_THROW] = "THROW", 2405 [RTN_NAT] = "NAT", 2406 [RTN_XRESOLVE] = "XRESOLVE", 2407 }; 2408 2409 static inline const char *rtn_type(char *buf, size_t len, unsigned t) 2410 { 2411 if (t < __RTN_MAX && rtn_type_names[t]) 2412 return rtn_type_names[t]; 2413 snprintf(buf, len, "type %u", t); 2414 return buf; 2415 } 2416 2417 /* Pretty print the trie */ 2418 static int fib_trie_seq_show(struct seq_file *seq, void *v) 2419 { 2420 const struct fib_trie_iter *iter = seq->private; 2421 struct node *n = v; 2422 2423 if (!node_parent_rcu(n)) 2424 fib_table_print(seq, iter->tb); 2425 2426 if (IS_TNODE(n)) { 2427 struct tnode *tn = (struct tnode *) n; 2428 __be32 prf = htonl(mask_pfx(tn->key, tn->pos)); 2429 2430 seq_indent(seq, iter->depth-1); 2431 seq_printf(seq, " +-- %pI4/%d %d %d %d\n", 2432 &prf, tn->pos, tn->bits, tn->full_children, 2433 tn->empty_children); 2434 2435 } else { 2436 struct leaf *l = (struct leaf *) n; 2437 struct leaf_info *li; 2438 struct hlist_node *node; 2439 __be32 val = htonl(l->key); 2440 2441 seq_indent(seq, iter->depth); 2442 seq_printf(seq, " |-- %pI4\n", &val); 2443 2444 hlist_for_each_entry_rcu(li, node, &l->list, hlist) { 2445 struct fib_alias *fa; 2446 2447 list_for_each_entry_rcu(fa, &li->falh, fa_list) { 2448 char buf1[32], buf2[32]; 2449 2450 seq_indent(seq, iter->depth+1); 2451 seq_printf(seq, " /%d %s %s", li->plen, 2452 rtn_scope(buf1, sizeof(buf1), 2453 fa->fa_scope), 2454 rtn_type(buf2, sizeof(buf2), 2455 fa->fa_type)); 2456 if (fa->fa_tos) 2457 seq_printf(seq, " tos=%d", fa->fa_tos); 2458 seq_putc(seq, '\n'); 2459 } 2460 } 2461 } 2462 2463 return 0; 2464 } 2465 2466 static const struct seq_operations fib_trie_seq_ops = { 2467 .start = fib_trie_seq_start, 2468 .next = fib_trie_seq_next, 2469 .stop = fib_trie_seq_stop, 2470 .show = fib_trie_seq_show, 2471 }; 2472 2473 static int fib_trie_seq_open(struct inode *inode, struct file *file) 2474 { 2475 return seq_open_net(inode, file, &fib_trie_seq_ops, 2476 sizeof(struct fib_trie_iter)); 2477 } 2478 2479 static const struct file_operations fib_trie_fops = { 2480 .owner = THIS_MODULE, 2481 .open = fib_trie_seq_open, 2482 .read = seq_read, 2483 .llseek = seq_lseek, 2484 .release = seq_release_net, 2485 }; 2486 2487 struct fib_route_iter { 2488 struct seq_net_private p; 2489 struct trie *main_trie; 2490 loff_t pos; 2491 t_key key; 2492 }; 2493 2494 static struct leaf *fib_route_get_idx(struct fib_route_iter *iter, loff_t pos) 2495 { 2496 struct leaf *l = NULL; 2497 struct trie *t = iter->main_trie; 2498 2499 /* use cache location of last found key */ 2500 if (iter->pos > 0 && pos >= iter->pos && (l = fib_find_node(t, iter->key))) 2501 pos -= iter->pos; 2502 else { 2503 iter->pos = 0; 2504 l = trie_firstleaf(t); 2505 } 2506 2507 while (l && pos-- > 0) { 2508 iter->pos++; 2509 l = trie_nextleaf(l); 2510 } 2511 2512 if (l) 2513 iter->key = pos; /* remember it */ 2514 else 2515 iter->pos = 0; /* forget it */ 2516 2517 return l; 2518 } 2519 2520 static void *fib_route_seq_start(struct seq_file *seq, loff_t *pos) 2521 __acquires(RCU) 2522 { 2523 struct fib_route_iter *iter = seq->private; 2524 struct fib_table *tb; 2525 2526 rcu_read_lock(); 2527 tb = fib_get_table(seq_file_net(seq), RT_TABLE_MAIN); 2528 if (!tb) 2529 return NULL; 2530 2531 iter->main_trie = (struct trie *) tb->tb_data; 2532 if (*pos == 0) 2533 return SEQ_START_TOKEN; 2534 else 2535 return fib_route_get_idx(iter, *pos - 1); 2536 } 2537 2538 static void *fib_route_seq_next(struct seq_file *seq, void *v, loff_t *pos) 2539 { 2540 struct fib_route_iter *iter = seq->private; 2541 struct leaf *l = v; 2542 2543 ++*pos; 2544 if (v == SEQ_START_TOKEN) { 2545 iter->pos = 0; 2546 l = trie_firstleaf(iter->main_trie); 2547 } else { 2548 iter->pos++; 2549 l = trie_nextleaf(l); 2550 } 2551 2552 if (l) 2553 iter->key = l->key; 2554 else 2555 iter->pos = 0; 2556 return l; 2557 } 2558 2559 static void fib_route_seq_stop(struct seq_file *seq, void *v) 2560 __releases(RCU) 2561 { 2562 rcu_read_unlock(); 2563 } 2564 2565 static unsigned fib_flag_trans(int type, __be32 mask, const struct fib_info *fi) 2566 { 2567 static unsigned type2flags[RTN_MAX + 1] = { 2568 [7] = RTF_REJECT, [8] = RTF_REJECT, 2569 }; 2570 unsigned flags = type2flags[type]; 2571 2572 if (fi && fi->fib_nh->nh_gw) 2573 flags |= RTF_GATEWAY; 2574 if (mask == htonl(0xFFFFFFFF)) 2575 flags |= RTF_HOST; 2576 flags |= RTF_UP; 2577 return flags; 2578 } 2579 2580 /* 2581 * This outputs /proc/net/route. 2582 * The format of the file is not supposed to be changed 2583 * and needs to be same as fib_hash output to avoid breaking 2584 * legacy utilities 2585 */ 2586 static int fib_route_seq_show(struct seq_file *seq, void *v) 2587 { 2588 struct leaf *l = v; 2589 struct leaf_info *li; 2590 struct hlist_node *node; 2591 2592 if (v == SEQ_START_TOKEN) { 2593 seq_printf(seq, "%-127s\n", "Iface\tDestination\tGateway " 2594 "\tFlags\tRefCnt\tUse\tMetric\tMask\t\tMTU" 2595 "\tWindow\tIRTT"); 2596 return 0; 2597 } 2598 2599 hlist_for_each_entry_rcu(li, node, &l->list, hlist) { 2600 struct fib_alias *fa; 2601 __be32 mask, prefix; 2602 2603 mask = inet_make_mask(li->plen); 2604 prefix = htonl(l->key); 2605 2606 list_for_each_entry_rcu(fa, &li->falh, fa_list) { 2607 const struct fib_info *fi = fa->fa_info; 2608 unsigned flags = fib_flag_trans(fa->fa_type, mask, fi); 2609 int len; 2610 2611 if (fa->fa_type == RTN_BROADCAST 2612 || fa->fa_type == RTN_MULTICAST) 2613 continue; 2614 2615 if (fi) 2616 seq_printf(seq, 2617 "%s\t%08X\t%08X\t%04X\t%d\t%u\t" 2618 "%d\t%08X\t%d\t%u\t%u%n", 2619 fi->fib_dev ? fi->fib_dev->name : "*", 2620 prefix, 2621 fi->fib_nh->nh_gw, flags, 0, 0, 2622 fi->fib_priority, 2623 mask, 2624 (fi->fib_advmss ? 2625 fi->fib_advmss + 40 : 0), 2626 fi->fib_window, 2627 fi->fib_rtt >> 3, &len); 2628 else 2629 seq_printf(seq, 2630 "*\t%08X\t%08X\t%04X\t%d\t%u\t" 2631 "%d\t%08X\t%d\t%u\t%u%n", 2632 prefix, 0, flags, 0, 0, 0, 2633 mask, 0, 0, 0, &len); 2634 2635 seq_printf(seq, "%*s\n", 127 - len, ""); 2636 } 2637 } 2638 2639 return 0; 2640 } 2641 2642 static const struct seq_operations fib_route_seq_ops = { 2643 .start = fib_route_seq_start, 2644 .next = fib_route_seq_next, 2645 .stop = fib_route_seq_stop, 2646 .show = fib_route_seq_show, 2647 }; 2648 2649 static int fib_route_seq_open(struct inode *inode, struct file *file) 2650 { 2651 return seq_open_net(inode, file, &fib_route_seq_ops, 2652 sizeof(struct fib_route_iter)); 2653 } 2654 2655 static const struct file_operations fib_route_fops = { 2656 .owner = THIS_MODULE, 2657 .open = fib_route_seq_open, 2658 .read = seq_read, 2659 .llseek = seq_lseek, 2660 .release = seq_release_net, 2661 }; 2662 2663 int __net_init fib_proc_init(struct net *net) 2664 { 2665 if (!proc_net_fops_create(net, "fib_trie", S_IRUGO, &fib_trie_fops)) 2666 goto out1; 2667 2668 if (!proc_net_fops_create(net, "fib_triestat", S_IRUGO, 2669 &fib_triestat_fops)) 2670 goto out2; 2671 2672 if (!proc_net_fops_create(net, "route", S_IRUGO, &fib_route_fops)) 2673 goto out3; 2674 2675 return 0; 2676 2677 out3: 2678 proc_net_remove(net, "fib_triestat"); 2679 out2: 2680 proc_net_remove(net, "fib_trie"); 2681 out1: 2682 return -ENOMEM; 2683 } 2684 2685 void __net_exit fib_proc_exit(struct net *net) 2686 { 2687 proc_net_remove(net, "fib_trie"); 2688 proc_net_remove(net, "fib_triestat"); 2689 proc_net_remove(net, "route"); 2690 } 2691 2692 #endif /* CONFIG_PROC_FS */ 2693