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.
7 * Robert Olsson <robert.olsson@its.uu.se> Uppsala Universitet
8 * & Swedish University of Agricultural Sciences.
10 * Jens Laas <jens.laas@data.slu.se> Swedish University of
11 * Agricultural Sciences.
13 * Hans Liss <hans.liss@its.uu.se> Uppsala Universitet
15 * This work is based on the LPC-trie which is originally described in:
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/
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
26 * Code from fib_hash has been reused which includes the following header:
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.
33 * IPv4 FIB: lookup engine and maintenance routines.
36 * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
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.
43 * Substantial contributions to this work comes from:
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>
51 #define VERSION "0.409"
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>
59 #include <linux/string.h>
60 #include <linux/socket.h>
61 #include <linux/sockios.h>
62 #include <linux/errno.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 <net/net_namespace.h>
77 #include <net/protocol.h>
78 #include <net/route.h>
81 #include <net/ip_fib.h>
82 #include "fib_lookup.h"
84 #define MAX_STAT_DEPTH 32
86 #define KEYLENGTH (8*sizeof(t_key))
88 typedef unsigned int t_key;
92 #define NODE_TYPE_MASK 0x1UL
93 #define NODE_TYPE(node) ((node)->parent & NODE_TYPE_MASK)
95 #define IS_TNODE(n) (!(n->parent & T_LEAF))
96 #define IS_LEAF(n) (n->parent & T_LEAF)
104 unsigned long parent;
106 struct hlist_head list;
111 struct hlist_node hlist;
114 struct list_head falh;
118 unsigned long parent;
120 unsigned char pos; /* 2log(KEYLENGTH) bits needed */
121 unsigned char bits; /* 2log(KEYLENGTH) bits needed */
122 unsigned int full_children; /* KEYLENGTH bits needed */
123 unsigned int empty_children; /* KEYLENGTH bits needed */
126 struct work_struct work;
127 struct tnode *tnode_free;
129 struct rt_trie_node *child[0];
132 #ifdef CONFIG_IP_FIB_TRIE_STATS
133 struct trie_use_stats {
135 unsigned int backtrack;
136 unsigned int semantic_match_passed;
137 unsigned int semantic_match_miss;
138 unsigned int null_node_hit;
139 unsigned int resize_node_skipped;
144 unsigned int totdepth;
145 unsigned int maxdepth;
148 unsigned int nullpointers;
149 unsigned int prefixes;
150 unsigned int nodesizes[MAX_STAT_DEPTH];
154 struct rt_trie_node *trie;
155 #ifdef CONFIG_IP_FIB_TRIE_STATS
156 struct trie_use_stats stats;
160 static void put_child(struct trie *t, struct tnode *tn, int i, struct rt_trie_node *n);
161 static void tnode_put_child_reorg(struct tnode *tn, int i, struct rt_trie_node *n,
163 static struct rt_trie_node *resize(struct trie *t, struct tnode *tn);
164 static struct tnode *inflate(struct trie *t, struct tnode *tn);
165 static struct tnode *halve(struct trie *t, struct tnode *tn);
166 /* tnodes to free after resize(); protected by RTNL */
167 static struct tnode *tnode_free_head;
168 static size_t tnode_free_size;
171 * synchronize_rcu after call_rcu for that many pages; it should be especially
172 * useful before resizing the root node with PREEMPT_NONE configs; the value was
173 * obtained experimentally, aiming to avoid visible slowdown.
175 static const int sync_pages = 128;
177 static struct kmem_cache *fn_alias_kmem __read_mostly;
178 static struct kmem_cache *trie_leaf_kmem __read_mostly;
180 static inline struct tnode *node_parent(struct rt_trie_node *node)
182 return (struct tnode *)(node->parent & ~NODE_TYPE_MASK);
185 static inline struct tnode *node_parent_rcu(struct rt_trie_node *node)
187 struct tnode *ret = node_parent(node);
189 return rcu_dereference_rtnl(ret);
192 /* Same as rcu_assign_pointer
193 * but that macro() assumes that value is a pointer.
195 static inline void node_set_parent(struct rt_trie_node *node, struct tnode *ptr)
198 node->parent = (unsigned long)ptr | NODE_TYPE(node);
201 static inline struct rt_trie_node *tnode_get_child(struct tnode *tn, unsigned int i)
203 BUG_ON(i >= 1U << tn->bits);
208 static inline struct rt_trie_node *tnode_get_child_rcu(struct tnode *tn, unsigned int i)
210 struct rt_trie_node *ret = tnode_get_child(tn, i);
212 return rcu_dereference_rtnl(ret);
215 static inline int tnode_child_length(const struct tnode *tn)
217 return 1 << tn->bits;
220 static inline t_key mask_pfx(t_key k, unsigned int l)
222 return (l == 0) ? 0 : k >> (KEYLENGTH-l) << (KEYLENGTH-l);
225 static inline t_key tkey_extract_bits(t_key a, unsigned int offset, unsigned int bits)
227 if (offset < KEYLENGTH)
228 return ((t_key)(a << offset)) >> (KEYLENGTH - bits);
233 static inline int tkey_equals(t_key a, t_key b)
238 static inline int tkey_sub_equals(t_key a, int offset, int bits, t_key b)
240 if (bits == 0 || offset >= KEYLENGTH)
242 bits = bits > KEYLENGTH ? KEYLENGTH : bits;
243 return ((a ^ b) << offset) >> (KEYLENGTH - bits) == 0;
246 static inline int tkey_mismatch(t_key a, int offset, t_key b)
253 while ((diff << i) >> (KEYLENGTH-1) == 0)
259 To understand this stuff, an understanding of keys and all their bits is
260 necessary. Every node in the trie has a key associated with it, but not
261 all of the bits in that key are significant.
263 Consider a node 'n' and its parent 'tp'.
265 If n is a leaf, every bit in its key is significant. Its presence is
266 necessitated by path compression, since during a tree traversal (when
267 searching for a leaf - unless we are doing an insertion) we will completely
268 ignore all skipped bits we encounter. Thus we need to verify, at the end of
269 a potentially successful search, that we have indeed been walking the
272 Note that we can never "miss" the correct key in the tree if present by
273 following the wrong path. Path compression ensures that segments of the key
274 that are the same for all keys with a given prefix are skipped, but the
275 skipped part *is* identical for each node in the subtrie below the skipped
276 bit! trie_insert() in this implementation takes care of that - note the
277 call to tkey_sub_equals() in trie_insert().
279 if n is an internal node - a 'tnode' here, the various parts of its key
280 have many different meanings.
283 _________________________________________________________________
284 | i | i | i | i | i | i | i | N | N | N | S | S | S | S | S | C |
285 -----------------------------------------------------------------
286 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
288 _________________________________________________________________
289 | C | C | C | u | u | u | u | u | u | u | u | u | u | u | u | u |
290 -----------------------------------------------------------------
291 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
298 First, let's just ignore the bits that come before the parent tp, that is
299 the bits from 0 to (tp->pos-1). They are *known* but at this point we do
300 not use them for anything.
302 The bits from (tp->pos) to (tp->pos + tp->bits - 1) - "N", above - are the
303 index into the parent's child array. That is, they will be used to find
304 'n' among tp's children.
306 The bits from (tp->pos + tp->bits) to (n->pos - 1) - "S" - are skipped bits
309 All the bits we have seen so far are significant to the node n. The rest
310 of the bits are really not needed or indeed known in n->key.
312 The bits from (n->pos) to (n->pos + n->bits - 1) - "C" - are the index into
313 n's child array, and will of course be different for each child.
316 The rest of the bits, from (n->pos + n->bits) onward, are completely unknown
321 static inline void check_tnode(const struct tnode *tn)
323 WARN_ON(tn && tn->pos+tn->bits > 32);
326 static const int halve_threshold = 25;
327 static const int inflate_threshold = 50;
328 static const int halve_threshold_root = 15;
329 static const int inflate_threshold_root = 30;
331 static void __alias_free_mem(struct rcu_head *head)
333 struct fib_alias *fa = container_of(head, struct fib_alias, rcu);
334 kmem_cache_free(fn_alias_kmem, fa);
337 static inline void alias_free_mem_rcu(struct fib_alias *fa)
339 call_rcu(&fa->rcu, __alias_free_mem);
342 static void __leaf_free_rcu(struct rcu_head *head)
344 struct leaf *l = container_of(head, struct leaf, rcu);
345 kmem_cache_free(trie_leaf_kmem, l);
348 static inline void free_leaf(struct leaf *l)
350 call_rcu_bh(&l->rcu, __leaf_free_rcu);
353 static inline void free_leaf_info(struct leaf_info *leaf)
355 kfree_rcu(leaf, rcu);
358 static struct tnode *tnode_alloc(size_t size)
360 if (size <= PAGE_SIZE)
361 return kzalloc(size, GFP_KERNEL);
363 return vzalloc(size);
366 static void __tnode_vfree(struct work_struct *arg)
368 struct tnode *tn = container_of(arg, struct tnode, work);
372 static void __tnode_free_rcu(struct rcu_head *head)
374 struct tnode *tn = container_of(head, struct tnode, rcu);
375 size_t size = sizeof(struct tnode) +
376 (sizeof(struct rt_trie_node *) << tn->bits);
378 if (size <= PAGE_SIZE)
381 INIT_WORK(&tn->work, __tnode_vfree);
382 schedule_work(&tn->work);
386 static inline void tnode_free(struct tnode *tn)
389 free_leaf((struct leaf *) tn);
391 call_rcu(&tn->rcu, __tnode_free_rcu);
394 static void tnode_free_safe(struct tnode *tn)
397 tn->tnode_free = tnode_free_head;
398 tnode_free_head = tn;
399 tnode_free_size += sizeof(struct tnode) +
400 (sizeof(struct rt_trie_node *) << tn->bits);
403 static void tnode_free_flush(void)
407 while ((tn = tnode_free_head)) {
408 tnode_free_head = tn->tnode_free;
409 tn->tnode_free = NULL;
413 if (tnode_free_size >= PAGE_SIZE * sync_pages) {
419 static struct leaf *leaf_new(void)
421 struct leaf *l = kmem_cache_alloc(trie_leaf_kmem, GFP_KERNEL);
424 INIT_HLIST_HEAD(&l->list);
429 static struct leaf_info *leaf_info_new(int plen)
431 struct leaf_info *li = kmalloc(sizeof(struct leaf_info), GFP_KERNEL);
434 INIT_LIST_HEAD(&li->falh);
439 static struct tnode *tnode_new(t_key key, int pos, int bits)
441 size_t sz = sizeof(struct tnode) + (sizeof(struct rt_trie_node *) << bits);
442 struct tnode *tn = tnode_alloc(sz);
445 tn->parent = T_TNODE;
449 tn->full_children = 0;
450 tn->empty_children = 1<<bits;
453 pr_debug("AT %p s=%zu %zu\n", tn, sizeof(struct tnode),
454 sizeof(struct rt_trie_node) << bits);
459 * Check whether a tnode 'n' is "full", i.e. it is an internal node
460 * and no bits are skipped. See discussion in dyntree paper p. 6
463 static inline int tnode_full(const struct tnode *tn, const struct rt_trie_node *n)
465 if (n == NULL || IS_LEAF(n))
468 return ((struct tnode *) n)->pos == tn->pos + tn->bits;
471 static inline void put_child(struct trie *t, struct tnode *tn, int i,
472 struct rt_trie_node *n)
474 tnode_put_child_reorg(tn, i, n, -1);
478 * Add a child at position i overwriting the old value.
479 * Update the value of full_children and empty_children.
482 static void tnode_put_child_reorg(struct tnode *tn, int i, struct rt_trie_node *n,
485 struct rt_trie_node *chi = tn->child[i];
488 BUG_ON(i >= 1<<tn->bits);
490 /* update emptyChildren */
491 if (n == NULL && chi != NULL)
492 tn->empty_children++;
493 else if (n != NULL && chi == NULL)
494 tn->empty_children--;
496 /* update fullChildren */
498 wasfull = tnode_full(tn, chi);
500 isfull = tnode_full(tn, n);
501 if (wasfull && !isfull)
503 else if (!wasfull && isfull)
507 node_set_parent(n, tn);
509 rcu_assign_pointer(tn->child[i], n);
513 static struct rt_trie_node *resize(struct trie *t, struct tnode *tn)
516 struct tnode *old_tn;
517 int inflate_threshold_use;
518 int halve_threshold_use;
524 pr_debug("In tnode_resize %p inflate_threshold=%d threshold=%d\n",
525 tn, inflate_threshold, halve_threshold);
528 if (tn->empty_children == tnode_child_length(tn)) {
533 if (tn->empty_children == tnode_child_length(tn) - 1)
536 * Double as long as the resulting node has a number of
537 * nonempty nodes that are above the threshold.
541 * From "Implementing a dynamic compressed trie" by Stefan Nilsson of
542 * the Helsinki University of Technology and Matti Tikkanen of Nokia
543 * Telecommunications, page 6:
544 * "A node is doubled if the ratio of non-empty children to all
545 * children in the *doubled* node is at least 'high'."
547 * 'high' in this instance is the variable 'inflate_threshold'. It
548 * is expressed as a percentage, so we multiply it with
549 * tnode_child_length() and instead of multiplying by 2 (since the
550 * child array will be doubled by inflate()) and multiplying
551 * the left-hand side by 100 (to handle the percentage thing) we
552 * multiply the left-hand side by 50.
554 * The left-hand side may look a bit weird: tnode_child_length(tn)
555 * - tn->empty_children is of course the number of non-null children
556 * in the current node. tn->full_children is the number of "full"
557 * children, that is non-null tnodes with a skip value of 0.
558 * All of those will be doubled in the resulting inflated tnode, so
559 * we just count them one extra time here.
561 * A clearer way to write this would be:
563 * to_be_doubled = tn->full_children;
564 * not_to_be_doubled = tnode_child_length(tn) - tn->empty_children -
567 * new_child_length = tnode_child_length(tn) * 2;
569 * new_fill_factor = 100 * (not_to_be_doubled + 2*to_be_doubled) /
571 * if (new_fill_factor >= inflate_threshold)
573 * ...and so on, tho it would mess up the while () loop.
576 * 100 * (not_to_be_doubled + 2*to_be_doubled) / new_child_length >=
580 * 100 * (not_to_be_doubled + 2*to_be_doubled) >=
581 * inflate_threshold * new_child_length
583 * expand not_to_be_doubled and to_be_doubled, and shorten:
584 * 100 * (tnode_child_length(tn) - tn->empty_children +
585 * tn->full_children) >= inflate_threshold * new_child_length
587 * expand new_child_length:
588 * 100 * (tnode_child_length(tn) - tn->empty_children +
589 * tn->full_children) >=
590 * inflate_threshold * tnode_child_length(tn) * 2
593 * 50 * (tn->full_children + tnode_child_length(tn) -
594 * tn->empty_children) >= inflate_threshold *
595 * tnode_child_length(tn)
601 /* Keep root node larger */
603 if (!node_parent((struct rt_trie_node *)tn)) {
604 inflate_threshold_use = inflate_threshold_root;
605 halve_threshold_use = halve_threshold_root;
607 inflate_threshold_use = inflate_threshold;
608 halve_threshold_use = halve_threshold;
612 while ((tn->full_children > 0 && max_work-- &&
613 50 * (tn->full_children + tnode_child_length(tn)
614 - tn->empty_children)
615 >= inflate_threshold_use * tnode_child_length(tn))) {
622 #ifdef CONFIG_IP_FIB_TRIE_STATS
623 t->stats.resize_node_skipped++;
631 /* Return if at least one inflate is run */
632 if (max_work != MAX_WORK)
633 return (struct rt_trie_node *) tn;
636 * Halve as long as the number of empty children in this
637 * node is above threshold.
641 while (tn->bits > 1 && max_work-- &&
642 100 * (tnode_child_length(tn) - tn->empty_children) <
643 halve_threshold_use * tnode_child_length(tn)) {
649 #ifdef CONFIG_IP_FIB_TRIE_STATS
650 t->stats.resize_node_skipped++;
657 /* Only one child remains */
658 if (tn->empty_children == tnode_child_length(tn) - 1) {
660 for (i = 0; i < tnode_child_length(tn); i++) {
661 struct rt_trie_node *n;
667 /* compress one level */
669 node_set_parent(n, NULL);
674 return (struct rt_trie_node *) tn;
677 static struct tnode *inflate(struct trie *t, struct tnode *tn)
679 struct tnode *oldtnode = tn;
680 int olen = tnode_child_length(tn);
683 pr_debug("In inflate\n");
685 tn = tnode_new(oldtnode->key, oldtnode->pos, oldtnode->bits + 1);
688 return ERR_PTR(-ENOMEM);
691 * Preallocate and store tnodes before the actual work so we
692 * don't get into an inconsistent state if memory allocation
693 * fails. In case of failure we return the oldnode and inflate
694 * of tnode is ignored.
697 for (i = 0; i < olen; i++) {
700 inode = (struct tnode *) tnode_get_child(oldtnode, i);
703 inode->pos == oldtnode->pos + oldtnode->bits &&
705 struct tnode *left, *right;
706 t_key m = ~0U << (KEYLENGTH - 1) >> inode->pos;
708 left = tnode_new(inode->key&(~m), inode->pos + 1,
713 right = tnode_new(inode->key|m, inode->pos + 1,
721 put_child(t, tn, 2*i, (struct rt_trie_node *) left);
722 put_child(t, tn, 2*i+1, (struct rt_trie_node *) right);
726 for (i = 0; i < olen; i++) {
728 struct rt_trie_node *node = tnode_get_child(oldtnode, i);
729 struct tnode *left, *right;
736 /* A leaf or an internal node with skipped bits */
738 if (IS_LEAF(node) || ((struct tnode *) node)->pos >
739 tn->pos + tn->bits - 1) {
740 if (tkey_extract_bits(node->key,
741 oldtnode->pos + oldtnode->bits,
743 put_child(t, tn, 2*i, node);
745 put_child(t, tn, 2*i+1, node);
749 /* An internal node with two children */
750 inode = (struct tnode *) node;
752 if (inode->bits == 1) {
753 put_child(t, tn, 2*i, inode->child[0]);
754 put_child(t, tn, 2*i+1, inode->child[1]);
756 tnode_free_safe(inode);
760 /* An internal node with more than two children */
762 /* We will replace this node 'inode' with two new
763 * ones, 'left' and 'right', each with half of the
764 * original children. The two new nodes will have
765 * a position one bit further down the key and this
766 * means that the "significant" part of their keys
767 * (see the discussion near the top of this file)
768 * will differ by one bit, which will be "0" in
769 * left's key and "1" in right's key. Since we are
770 * moving the key position by one step, the bit that
771 * we are moving away from - the bit at position
772 * (inode->pos) - is the one that will differ between
773 * left and right. So... we synthesize that bit in the
775 * The mask 'm' below will be a single "one" bit at
776 * the position (inode->pos)
779 /* Use the old key, but set the new significant
783 left = (struct tnode *) tnode_get_child(tn, 2*i);
784 put_child(t, tn, 2*i, NULL);
788 right = (struct tnode *) tnode_get_child(tn, 2*i+1);
789 put_child(t, tn, 2*i+1, NULL);
793 size = tnode_child_length(left);
794 for (j = 0; j < size; j++) {
795 put_child(t, left, j, inode->child[j]);
796 put_child(t, right, j, inode->child[j + size]);
798 put_child(t, tn, 2*i, resize(t, left));
799 put_child(t, tn, 2*i+1, resize(t, right));
801 tnode_free_safe(inode);
803 tnode_free_safe(oldtnode);
807 int size = tnode_child_length(tn);
810 for (j = 0; j < size; j++)
812 tnode_free((struct tnode *)tn->child[j]);
816 return ERR_PTR(-ENOMEM);
820 static struct tnode *halve(struct trie *t, struct tnode *tn)
822 struct tnode *oldtnode = tn;
823 struct rt_trie_node *left, *right;
825 int olen = tnode_child_length(tn);
827 pr_debug("In halve\n");
829 tn = tnode_new(oldtnode->key, oldtnode->pos, oldtnode->bits - 1);
832 return ERR_PTR(-ENOMEM);
835 * Preallocate and store tnodes before the actual work so we
836 * don't get into an inconsistent state if memory allocation
837 * fails. In case of failure we return the oldnode and halve
838 * of tnode is ignored.
841 for (i = 0; i < olen; i += 2) {
842 left = tnode_get_child(oldtnode, i);
843 right = tnode_get_child(oldtnode, i+1);
845 /* Two nonempty children */
849 newn = tnode_new(left->key, tn->pos + tn->bits, 1);
854 put_child(t, tn, i/2, (struct rt_trie_node *)newn);
859 for (i = 0; i < olen; i += 2) {
860 struct tnode *newBinNode;
862 left = tnode_get_child(oldtnode, i);
863 right = tnode_get_child(oldtnode, i+1);
865 /* At least one of the children is empty */
867 if (right == NULL) /* Both are empty */
869 put_child(t, tn, i/2, right);
874 put_child(t, tn, i/2, left);
878 /* Two nonempty children */
879 newBinNode = (struct tnode *) tnode_get_child(tn, i/2);
880 put_child(t, tn, i/2, NULL);
881 put_child(t, newBinNode, 0, left);
882 put_child(t, newBinNode, 1, right);
883 put_child(t, tn, i/2, resize(t, newBinNode));
885 tnode_free_safe(oldtnode);
889 int size = tnode_child_length(tn);
892 for (j = 0; j < size; j++)
894 tnode_free((struct tnode *)tn->child[j]);
898 return ERR_PTR(-ENOMEM);
902 /* readside must use rcu_read_lock currently dump routines
903 via get_fa_head and dump */
905 static struct leaf_info *find_leaf_info(struct leaf *l, int plen)
907 struct hlist_head *head = &l->list;
908 struct hlist_node *node;
909 struct leaf_info *li;
911 hlist_for_each_entry_rcu(li, node, head, hlist)
912 if (li->plen == plen)
918 static inline struct list_head *get_fa_head(struct leaf *l, int plen)
920 struct leaf_info *li = find_leaf_info(l, plen);
928 static void insert_leaf_info(struct hlist_head *head, struct leaf_info *new)
930 struct leaf_info *li = NULL, *last = NULL;
931 struct hlist_node *node;
933 if (hlist_empty(head)) {
934 hlist_add_head_rcu(&new->hlist, head);
936 hlist_for_each_entry(li, node, head, hlist) {
937 if (new->plen > li->plen)
943 hlist_add_after_rcu(&last->hlist, &new->hlist);
945 hlist_add_before_rcu(&new->hlist, &li->hlist);
949 /* rcu_read_lock needs to be hold by caller from readside */
952 fib_find_node(struct trie *t, u32 key)
956 struct rt_trie_node *n;
959 n = rcu_dereference_rtnl(t->trie);
961 while (n != NULL && NODE_TYPE(n) == T_TNODE) {
962 tn = (struct tnode *) n;
966 if (tkey_sub_equals(tn->key, pos, tn->pos-pos, key)) {
967 pos = tn->pos + tn->bits;
968 n = tnode_get_child_rcu(tn,
969 tkey_extract_bits(key,
975 /* Case we have found a leaf. Compare prefixes */
977 if (n != NULL && IS_LEAF(n) && tkey_equals(key, n->key))
978 return (struct leaf *)n;
983 static void trie_rebalance(struct trie *t, struct tnode *tn)
991 while (tn != NULL && (tp = node_parent((struct rt_trie_node *)tn)) != NULL) {
992 cindex = tkey_extract_bits(key, tp->pos, tp->bits);
993 wasfull = tnode_full(tp, tnode_get_child(tp, cindex));
994 tn = (struct tnode *) resize(t, (struct tnode *)tn);
996 tnode_put_child_reorg((struct tnode *)tp, cindex,
997 (struct rt_trie_node *)tn, wasfull);
999 tp = node_parent((struct rt_trie_node *) tn);
1001 rcu_assign_pointer(t->trie, (struct rt_trie_node *)tn);
1009 /* Handle last (top) tnode */
1011 tn = (struct tnode *)resize(t, (struct tnode *)tn);
1013 rcu_assign_pointer(t->trie, (struct rt_trie_node *)tn);
1017 /* only used from updater-side */
1019 static struct list_head *fib_insert_node(struct trie *t, u32 key, int plen)
1022 struct tnode *tp = NULL, *tn = NULL;
1023 struct rt_trie_node *n;
1026 struct list_head *fa_head = NULL;
1027 struct leaf_info *li;
1033 /* If we point to NULL, stop. Either the tree is empty and we should
1034 * just put a new leaf in if, or we have reached an empty child slot,
1035 * and we should just put our new leaf in that.
1036 * If we point to a T_TNODE, check if it matches our key. Note that
1037 * a T_TNODE might be skipping any number of bits - its 'pos' need
1038 * not be the parent's 'pos'+'bits'!
1040 * If it does match the current key, get pos/bits from it, extract
1041 * the index from our key, push the T_TNODE and walk the tree.
1043 * If it doesn't, we have to replace it with a new T_TNODE.
1045 * If we point to a T_LEAF, it might or might not have the same key
1046 * as we do. If it does, just change the value, update the T_LEAF's
1047 * value, and return it.
1048 * If it doesn't, we need to replace it with a T_TNODE.
1051 while (n != NULL && NODE_TYPE(n) == T_TNODE) {
1052 tn = (struct tnode *) n;
1056 if (tkey_sub_equals(tn->key, pos, tn->pos-pos, key)) {
1058 pos = tn->pos + tn->bits;
1059 n = tnode_get_child(tn,
1060 tkey_extract_bits(key,
1064 BUG_ON(n && node_parent(n) != tn);
1070 * n ----> NULL, LEAF or TNODE
1072 * tp is n's (parent) ----> NULL or TNODE
1075 BUG_ON(tp && IS_LEAF(tp));
1077 /* Case 1: n is a leaf. Compare prefixes */
1079 if (n != NULL && IS_LEAF(n) && tkey_equals(key, n->key)) {
1080 l = (struct leaf *) n;
1081 li = leaf_info_new(plen);
1086 fa_head = &li->falh;
1087 insert_leaf_info(&l->list, li);
1096 li = leaf_info_new(plen);
1103 fa_head = &li->falh;
1104 insert_leaf_info(&l->list, li);
1106 if (t->trie && n == NULL) {
1107 /* Case 2: n is NULL, and will just insert a new leaf */
1109 node_set_parent((struct rt_trie_node *)l, tp);
1111 cindex = tkey_extract_bits(key, tp->pos, tp->bits);
1112 put_child(t, (struct tnode *)tp, cindex, (struct rt_trie_node *)l);
1114 /* Case 3: n is a LEAF or a TNODE and the key doesn't match. */
1116 * Add a new tnode here
1117 * first tnode need some special handling
1121 pos = tp->pos+tp->bits;
1126 newpos = tkey_mismatch(key, pos, n->key);
1127 tn = tnode_new(n->key, newpos, 1);
1130 tn = tnode_new(key, newpos, 1); /* First tnode */
1139 node_set_parent((struct rt_trie_node *)tn, tp);
1141 missbit = tkey_extract_bits(key, newpos, 1);
1142 put_child(t, tn, missbit, (struct rt_trie_node *)l);
1143 put_child(t, tn, 1-missbit, n);
1146 cindex = tkey_extract_bits(key, tp->pos, tp->bits);
1147 put_child(t, (struct tnode *)tp, cindex,
1148 (struct rt_trie_node *)tn);
1150 rcu_assign_pointer(t->trie, (struct rt_trie_node *)tn);
1155 if (tp && tp->pos + tp->bits > 32)
1156 pr_warning("fib_trie"
1157 " tp=%p pos=%d, bits=%d, key=%0x plen=%d\n",
1158 tp, tp->pos, tp->bits, key, plen);
1160 /* Rebalance the trie */
1162 trie_rebalance(t, tp);
1168 * Caller must hold RTNL.
1170 int fib_table_insert(struct fib_table *tb, struct fib_config *cfg)
1172 struct trie *t = (struct trie *) tb->tb_data;
1173 struct fib_alias *fa, *new_fa;
1174 struct list_head *fa_head = NULL;
1175 struct fib_info *fi;
1176 int plen = cfg->fc_dst_len;
1177 u8 tos = cfg->fc_tos;
1185 key = ntohl(cfg->fc_dst);
1187 pr_debug("Insert table=%u %08x/%d\n", tb->tb_id, key, plen);
1189 mask = ntohl(inet_make_mask(plen));
1196 fi = fib_create_info(cfg);
1202 l = fib_find_node(t, key);
1206 fa_head = get_fa_head(l, plen);
1207 fa = fib_find_alias(fa_head, tos, fi->fib_priority);
1210 /* Now fa, if non-NULL, points to the first fib alias
1211 * with the same keys [prefix,tos,priority], if such key already
1212 * exists or to the node before which we will insert new one.
1214 * If fa is NULL, we will need to allocate a new one and
1215 * insert to the head of f.
1217 * If f is NULL, no fib node matched the destination key
1218 * and we need to allocate a new one of those as well.
1221 if (fa && fa->fa_tos == tos &&
1222 fa->fa_info->fib_priority == fi->fib_priority) {
1223 struct fib_alias *fa_first, *fa_match;
1226 if (cfg->fc_nlflags & NLM_F_EXCL)
1230 * 1. Find exact match for type, scope, fib_info to avoid
1232 * 2. Find next 'fa' (or head), NLM_F_APPEND inserts before it
1236 fa = list_entry(fa->fa_list.prev, struct fib_alias, fa_list);
1237 list_for_each_entry_continue(fa, fa_head, fa_list) {
1238 if (fa->fa_tos != tos)
1240 if (fa->fa_info->fib_priority != fi->fib_priority)
1242 if (fa->fa_type == cfg->fc_type &&
1243 fa->fa_info == fi) {
1249 if (cfg->fc_nlflags & NLM_F_REPLACE) {
1250 struct fib_info *fi_drop;
1260 new_fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL);
1264 fi_drop = fa->fa_info;
1265 new_fa->fa_tos = fa->fa_tos;
1266 new_fa->fa_info = fi;
1267 new_fa->fa_type = cfg->fc_type;
1268 state = fa->fa_state;
1269 new_fa->fa_state = state & ~FA_S_ACCESSED;
1271 list_replace_rcu(&fa->fa_list, &new_fa->fa_list);
1272 alias_free_mem_rcu(fa);
1274 fib_release_info(fi_drop);
1275 if (state & FA_S_ACCESSED)
1276 rt_cache_flush(cfg->fc_nlinfo.nl_net, -1);
1277 rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen,
1278 tb->tb_id, &cfg->fc_nlinfo, NLM_F_REPLACE);
1282 /* Error if we find a perfect match which
1283 * uses the same scope, type, and nexthop
1289 if (!(cfg->fc_nlflags & NLM_F_APPEND))
1293 if (!(cfg->fc_nlflags & NLM_F_CREATE))
1297 new_fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL);
1301 new_fa->fa_info = fi;
1302 new_fa->fa_tos = tos;
1303 new_fa->fa_type = cfg->fc_type;
1304 new_fa->fa_state = 0;
1306 * Insert new entry to the list.
1310 fa_head = fib_insert_node(t, key, plen);
1311 if (unlikely(!fa_head)) {
1313 goto out_free_new_fa;
1317 list_add_tail_rcu(&new_fa->fa_list,
1318 (fa ? &fa->fa_list : fa_head));
1320 rt_cache_flush(cfg->fc_nlinfo.nl_net, -1);
1321 rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen, tb->tb_id,
1322 &cfg->fc_nlinfo, 0);
1327 kmem_cache_free(fn_alias_kmem, new_fa);
1329 fib_release_info(fi);
1334 /* should be called with rcu_read_lock */
1335 static int check_leaf(struct fib_table *tb, struct trie *t, struct leaf *l,
1336 t_key key, const struct flowi4 *flp,
1337 struct fib_result *res, int fib_flags)
1339 struct leaf_info *li;
1340 struct hlist_head *hhead = &l->list;
1341 struct hlist_node *node;
1343 hlist_for_each_entry_rcu(li, node, hhead, hlist) {
1344 struct fib_alias *fa;
1345 int plen = li->plen;
1346 __be32 mask = inet_make_mask(plen);
1348 if (l->key != (key & ntohl(mask)))
1351 list_for_each_entry_rcu(fa, &li->falh, fa_list) {
1352 struct fib_info *fi = fa->fa_info;
1355 if (fa->fa_tos && fa->fa_tos != flp->flowi4_tos)
1357 if (fa->fa_info->fib_scope < flp->flowi4_scope)
1359 fib_alias_accessed(fa);
1360 err = fib_props[fa->fa_type].error;
1362 #ifdef CONFIG_IP_FIB_TRIE_STATS
1363 t->stats.semantic_match_passed++;
1367 if (fi->fib_flags & RTNH_F_DEAD)
1369 for (nhsel = 0; nhsel < fi->fib_nhs; nhsel++) {
1370 const struct fib_nh *nh = &fi->fib_nh[nhsel];
1372 if (nh->nh_flags & RTNH_F_DEAD)
1374 if (flp->flowi4_oif && flp->flowi4_oif != nh->nh_oif)
1377 #ifdef CONFIG_IP_FIB_TRIE_STATS
1378 t->stats.semantic_match_passed++;
1380 res->prefixlen = plen;
1381 res->nh_sel = nhsel;
1382 res->type = fa->fa_type;
1383 res->scope = fa->fa_info->fib_scope;
1386 res->fa_head = &li->falh;
1387 if (!(fib_flags & FIB_LOOKUP_NOREF))
1388 atomic_inc(&res->fi->fib_clntref);
1393 #ifdef CONFIG_IP_FIB_TRIE_STATS
1394 t->stats.semantic_match_miss++;
1401 int fib_table_lookup(struct fib_table *tb, const struct flowi4 *flp,
1402 struct fib_result *res, int fib_flags)
1404 struct trie *t = (struct trie *) tb->tb_data;
1406 struct rt_trie_node *n;
1408 unsigned int pos, bits;
1409 t_key key = ntohl(flp->daddr);
1410 unsigned int chopped_off;
1412 unsigned int current_prefix_length = KEYLENGTH;
1414 t_key pref_mismatch;
1418 n = rcu_dereference(t->trie);
1422 #ifdef CONFIG_IP_FIB_TRIE_STATS
1428 ret = check_leaf(tb, t, (struct leaf *)n, key, flp, res, fib_flags);
1432 pn = (struct tnode *) n;
1440 cindex = tkey_extract_bits(mask_pfx(key, current_prefix_length),
1443 n = tnode_get_child_rcu(pn, cindex);
1446 #ifdef CONFIG_IP_FIB_TRIE_STATS
1447 t->stats.null_node_hit++;
1453 ret = check_leaf(tb, t, (struct leaf *)n, key, flp, res, fib_flags);
1459 cn = (struct tnode *)n;
1462 * It's a tnode, and we can do some extra checks here if we
1463 * like, to avoid descending into a dead-end branch.
1464 * This tnode is in the parent's child array at index
1465 * key[p_pos..p_pos+p_bits] but potentially with some bits
1466 * chopped off, so in reality the index may be just a
1467 * subprefix, padded with zero at the end.
1468 * We can also take a look at any skipped bits in this
1469 * tnode - everything up to p_pos is supposed to be ok,
1470 * and the non-chopped bits of the index (se previous
1471 * paragraph) are also guaranteed ok, but the rest is
1472 * considered unknown.
1474 * The skipped bits are key[pos+bits..cn->pos].
1477 /* If current_prefix_length < pos+bits, we are already doing
1478 * actual prefix matching, which means everything from
1479 * pos+(bits-chopped_off) onward must be zero along some
1480 * branch of this subtree - otherwise there is *no* valid
1481 * prefix present. Here we can only check the skipped
1482 * bits. Remember, since we have already indexed into the
1483 * parent's child array, we know that the bits we chopped of
1487 /* NOTA BENE: Checking only skipped bits
1488 for the new node here */
1490 if (current_prefix_length < pos+bits) {
1491 if (tkey_extract_bits(cn->key, current_prefix_length,
1492 cn->pos - current_prefix_length)
1498 * If chopped_off=0, the index is fully validated and we
1499 * only need to look at the skipped bits for this, the new,
1500 * tnode. What we actually want to do is to find out if
1501 * these skipped bits match our key perfectly, or if we will
1502 * have to count on finding a matching prefix further down,
1503 * because if we do, we would like to have some way of
1504 * verifying the existence of such a prefix at this point.
1507 /* The only thing we can do at this point is to verify that
1508 * any such matching prefix can indeed be a prefix to our
1509 * key, and if the bits in the node we are inspecting that
1510 * do not match our key are not ZERO, this cannot be true.
1511 * Thus, find out where there is a mismatch (before cn->pos)
1512 * and verify that all the mismatching bits are zero in the
1517 * Note: We aren't very concerned about the piece of
1518 * the key that precede pn->pos+pn->bits, since these
1519 * have already been checked. The bits after cn->pos
1520 * aren't checked since these are by definition
1521 * "unknown" at this point. Thus, what we want to see
1522 * is if we are about to enter the "prefix matching"
1523 * state, and in that case verify that the skipped
1524 * bits that will prevail throughout this subtree are
1525 * zero, as they have to be if we are to find a
1529 pref_mismatch = mask_pfx(cn->key ^ key, cn->pos);
1532 * In short: If skipped bits in this node do not match
1533 * the search key, enter the "prefix matching"
1536 if (pref_mismatch) {
1537 int mp = KEYLENGTH - fls(pref_mismatch);
1539 if (tkey_extract_bits(cn->key, mp, cn->pos - mp) != 0)
1542 if (current_prefix_length >= cn->pos)
1543 current_prefix_length = mp;
1546 pn = (struct tnode *)n; /* Descend */
1553 /* As zero don't change the child key (cindex) */
1554 while ((chopped_off <= pn->bits)
1555 && !(cindex & (1<<(chopped_off-1))))
1558 /* Decrease current_... with bits chopped off */
1559 if (current_prefix_length > pn->pos + pn->bits - chopped_off)
1560 current_prefix_length = pn->pos + pn->bits
1564 * Either we do the actual chop off according or if we have
1565 * chopped off all bits in this tnode walk up to our parent.
1568 if (chopped_off <= pn->bits) {
1569 cindex &= ~(1 << (chopped_off-1));
1571 struct tnode *parent = node_parent_rcu((struct rt_trie_node *) pn);
1575 /* Get Child's index */
1576 cindex = tkey_extract_bits(pn->key, parent->pos, parent->bits);
1580 #ifdef CONFIG_IP_FIB_TRIE_STATS
1581 t->stats.backtrack++;
1594 * Remove the leaf and return parent.
1596 static void trie_leaf_remove(struct trie *t, struct leaf *l)
1598 struct tnode *tp = node_parent((struct rt_trie_node *) l);
1600 pr_debug("entering trie_leaf_remove(%p)\n", l);
1603 t_key cindex = tkey_extract_bits(l->key, tp->pos, tp->bits);
1604 put_child(t, (struct tnode *)tp, cindex, NULL);
1605 trie_rebalance(t, tp);
1607 rcu_assign_pointer(t->trie, NULL);
1613 * Caller must hold RTNL.
1615 int fib_table_delete(struct fib_table *tb, struct fib_config *cfg)
1617 struct trie *t = (struct trie *) tb->tb_data;
1619 int plen = cfg->fc_dst_len;
1620 u8 tos = cfg->fc_tos;
1621 struct fib_alias *fa, *fa_to_delete;
1622 struct list_head *fa_head;
1624 struct leaf_info *li;
1629 key = ntohl(cfg->fc_dst);
1630 mask = ntohl(inet_make_mask(plen));
1636 l = fib_find_node(t, key);
1641 fa_head = get_fa_head(l, plen);
1642 fa = fib_find_alias(fa_head, tos, 0);
1647 pr_debug("Deleting %08x/%d tos=%d t=%p\n", key, plen, tos, t);
1649 fa_to_delete = NULL;
1650 fa = list_entry(fa->fa_list.prev, struct fib_alias, fa_list);
1651 list_for_each_entry_continue(fa, fa_head, fa_list) {
1652 struct fib_info *fi = fa->fa_info;
1654 if (fa->fa_tos != tos)
1657 if ((!cfg->fc_type || fa->fa_type == cfg->fc_type) &&
1658 (cfg->fc_scope == RT_SCOPE_NOWHERE ||
1659 fa->fa_info->fib_scope == cfg->fc_scope) &&
1660 (!cfg->fc_prefsrc ||
1661 fi->fib_prefsrc == cfg->fc_prefsrc) &&
1662 (!cfg->fc_protocol ||
1663 fi->fib_protocol == cfg->fc_protocol) &&
1664 fib_nh_match(cfg, fi) == 0) {
1674 rtmsg_fib(RTM_DELROUTE, htonl(key), fa, plen, tb->tb_id,
1675 &cfg->fc_nlinfo, 0);
1677 l = fib_find_node(t, key);
1678 li = find_leaf_info(l, plen);
1680 list_del_rcu(&fa->fa_list);
1682 if (list_empty(fa_head)) {
1683 hlist_del_rcu(&li->hlist);
1687 if (hlist_empty(&l->list))
1688 trie_leaf_remove(t, l);
1690 if (fa->fa_state & FA_S_ACCESSED)
1691 rt_cache_flush(cfg->fc_nlinfo.nl_net, -1);
1693 fib_release_info(fa->fa_info);
1694 alias_free_mem_rcu(fa);
1698 static int trie_flush_list(struct list_head *head)
1700 struct fib_alias *fa, *fa_node;
1703 list_for_each_entry_safe(fa, fa_node, head, fa_list) {
1704 struct fib_info *fi = fa->fa_info;
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);
1716 static int trie_flush_leaf(struct leaf *l)
1719 struct hlist_head *lih = &l->list;
1720 struct hlist_node *node, *tmp;
1721 struct leaf_info *li = NULL;
1723 hlist_for_each_entry_safe(li, node, tmp, lih, hlist) {
1724 found += trie_flush_list(&li->falh);
1726 if (list_empty(&li->falh)) {
1727 hlist_del_rcu(&li->hlist);
1735 * Scan for the next right leaf starting at node p->child[idx]
1736 * Since we have back pointer, no recursion necessary.
1738 static struct leaf *leaf_walk_rcu(struct tnode *p, struct rt_trie_node *c)
1744 idx = tkey_extract_bits(c->key, p->pos, p->bits) + 1;
1748 while (idx < 1u << p->bits) {
1749 c = tnode_get_child_rcu(p, idx++);
1754 prefetch(p->child[idx]);
1755 return (struct leaf *) c;
1758 /* Rescan start scanning in new node */
1759 p = (struct tnode *) c;
1763 /* Node empty, walk back up to parent */
1764 c = (struct rt_trie_node *) p;
1765 } while ((p = node_parent_rcu(c)) != NULL);
1767 return NULL; /* Root of trie */
1770 static struct leaf *trie_firstleaf(struct trie *t)
1772 struct tnode *n = (struct tnode *)rcu_dereference_rtnl(t->trie);
1777 if (IS_LEAF(n)) /* trie is just a leaf */
1778 return (struct leaf *) n;
1780 return leaf_walk_rcu(n, NULL);
1783 static struct leaf *trie_nextleaf(struct leaf *l)
1785 struct rt_trie_node *c = (struct rt_trie_node *) l;
1786 struct tnode *p = node_parent_rcu(c);
1789 return NULL; /* trie with just one leaf */
1791 return leaf_walk_rcu(p, c);
1794 static struct leaf *trie_leafindex(struct trie *t, int index)
1796 struct leaf *l = trie_firstleaf(t);
1798 while (l && index-- > 0)
1799 l = trie_nextleaf(l);
1806 * Caller must hold RTNL.
1808 int fib_table_flush(struct fib_table *tb)
1810 struct trie *t = (struct trie *) tb->tb_data;
1811 struct leaf *l, *ll = NULL;
1814 for (l = trie_firstleaf(t); l; l = trie_nextleaf(l)) {
1815 found += trie_flush_leaf(l);
1817 if (ll && hlist_empty(&ll->list))
1818 trie_leaf_remove(t, ll);
1822 if (ll && hlist_empty(&ll->list))
1823 trie_leaf_remove(t, ll);
1825 pr_debug("trie_flush found=%d\n", found);
1829 void fib_free_table(struct fib_table *tb)
1834 static int fn_trie_dump_fa(t_key key, int plen, struct list_head *fah,
1835 struct fib_table *tb,
1836 struct sk_buff *skb, struct netlink_callback *cb)
1839 struct fib_alias *fa;
1840 __be32 xkey = htonl(key);
1845 /* rcu_read_lock is hold by caller */
1847 list_for_each_entry_rcu(fa, fah, fa_list) {
1853 if (fib_dump_info(skb, NETLINK_CB(cb->skb).pid,
1861 fa->fa_info, NLM_F_MULTI) < 0) {
1871 static int fn_trie_dump_leaf(struct leaf *l, struct fib_table *tb,
1872 struct sk_buff *skb, struct netlink_callback *cb)
1874 struct leaf_info *li;
1875 struct hlist_node *node;
1881 /* rcu_read_lock is hold by caller */
1882 hlist_for_each_entry_rcu(li, node, &l->list, hlist) {
1891 if (list_empty(&li->falh))
1894 if (fn_trie_dump_fa(l->key, li->plen, &li->falh, tb, skb, cb) < 0) {
1905 int fib_table_dump(struct fib_table *tb, struct sk_buff *skb,
1906 struct netlink_callback *cb)
1909 struct trie *t = (struct trie *) tb->tb_data;
1910 t_key key = cb->args[2];
1911 int count = cb->args[3];
1914 /* Dump starting at last key.
1915 * Note: 0.0.0.0/0 (ie default) is first key.
1918 l = trie_firstleaf(t);
1920 /* Normally, continue from last key, but if that is missing
1921 * fallback to using slow rescan
1923 l = fib_find_node(t, key);
1925 l = trie_leafindex(t, count);
1929 cb->args[2] = l->key;
1930 if (fn_trie_dump_leaf(l, tb, skb, cb) < 0) {
1931 cb->args[3] = count;
1937 l = trie_nextleaf(l);
1938 memset(&cb->args[4], 0,
1939 sizeof(cb->args) - 4*sizeof(cb->args[0]));
1941 cb->args[3] = count;
1947 void __init fib_trie_init(void)
1949 fn_alias_kmem = kmem_cache_create("ip_fib_alias",
1950 sizeof(struct fib_alias),
1951 0, SLAB_PANIC, NULL);
1953 trie_leaf_kmem = kmem_cache_create("ip_fib_trie",
1954 max(sizeof(struct leaf),
1955 sizeof(struct leaf_info)),
1956 0, SLAB_PANIC, NULL);
1960 struct fib_table *fib_trie_table(u32 id)
1962 struct fib_table *tb;
1965 tb = kmalloc(sizeof(struct fib_table) + sizeof(struct trie),
1971 tb->tb_default = -1;
1973 t = (struct trie *) tb->tb_data;
1974 memset(t, 0, sizeof(*t));
1979 #ifdef CONFIG_PROC_FS
1980 /* Depth first Trie walk iterator */
1981 struct fib_trie_iter {
1982 struct seq_net_private p;
1983 struct fib_table *tb;
1984 struct tnode *tnode;
1989 static struct rt_trie_node *fib_trie_get_next(struct fib_trie_iter *iter)
1991 struct tnode *tn = iter->tnode;
1992 unsigned int cindex = iter->index;
1995 /* A single entry routing table */
1999 pr_debug("get_next iter={node=%p index=%d depth=%d}\n",
2000 iter->tnode, iter->index, iter->depth);
2002 while (cindex < (1<<tn->bits)) {
2003 struct rt_trie_node *n = tnode_get_child_rcu(tn, cindex);
2008 iter->index = cindex + 1;
2010 /* push down one level */
2011 iter->tnode = (struct tnode *) n;
2021 /* Current node exhausted, pop back up */
2022 p = node_parent_rcu((struct rt_trie_node *)tn);
2024 cindex = tkey_extract_bits(tn->key, p->pos, p->bits)+1;
2034 static struct rt_trie_node *fib_trie_get_first(struct fib_trie_iter *iter,
2037 struct rt_trie_node *n;
2042 n = rcu_dereference(t->trie);
2047 iter->tnode = (struct tnode *) n;
2059 static void trie_collect_stats(struct trie *t, struct trie_stat *s)
2061 struct rt_trie_node *n;
2062 struct fib_trie_iter iter;
2064 memset(s, 0, sizeof(*s));
2067 for (n = fib_trie_get_first(&iter, t); n; n = fib_trie_get_next(&iter)) {
2069 struct leaf *l = (struct leaf *)n;
2070 struct leaf_info *li;
2071 struct hlist_node *tmp;
2074 s->totdepth += iter.depth;
2075 if (iter.depth > s->maxdepth)
2076 s->maxdepth = iter.depth;
2078 hlist_for_each_entry_rcu(li, tmp, &l->list, hlist)
2081 const struct tnode *tn = (const struct tnode *) n;
2085 if (tn->bits < MAX_STAT_DEPTH)
2086 s->nodesizes[tn->bits]++;
2088 for (i = 0; i < (1<<tn->bits); i++)
2097 * This outputs /proc/net/fib_triestats
2099 static void trie_show_stats(struct seq_file *seq, struct trie_stat *stat)
2101 unsigned int i, max, pointers, bytes, avdepth;
2104 avdepth = stat->totdepth*100 / stat->leaves;
2108 seq_printf(seq, "\tAver depth: %u.%02d\n",
2109 avdepth / 100, avdepth % 100);
2110 seq_printf(seq, "\tMax depth: %u\n", stat->maxdepth);
2112 seq_printf(seq, "\tLeaves: %u\n", stat->leaves);
2113 bytes = sizeof(struct leaf) * stat->leaves;
2115 seq_printf(seq, "\tPrefixes: %u\n", stat->prefixes);
2116 bytes += sizeof(struct leaf_info) * stat->prefixes;
2118 seq_printf(seq, "\tInternal nodes: %u\n\t", stat->tnodes);
2119 bytes += sizeof(struct tnode) * stat->tnodes;
2121 max = MAX_STAT_DEPTH;
2122 while (max > 0 && stat->nodesizes[max-1] == 0)
2126 for (i = 1; i <= max; i++)
2127 if (stat->nodesizes[i] != 0) {
2128 seq_printf(seq, " %u: %u", i, stat->nodesizes[i]);
2129 pointers += (1<<i) * stat->nodesizes[i];
2131 seq_putc(seq, '\n');
2132 seq_printf(seq, "\tPointers: %u\n", pointers);
2134 bytes += sizeof(struct rt_trie_node *) * pointers;
2135 seq_printf(seq, "Null ptrs: %u\n", stat->nullpointers);
2136 seq_printf(seq, "Total size: %u kB\n", (bytes + 1023) / 1024);
2139 #ifdef CONFIG_IP_FIB_TRIE_STATS
2140 static void trie_show_usage(struct seq_file *seq,
2141 const struct trie_use_stats *stats)
2143 seq_printf(seq, "\nCounters:\n---------\n");
2144 seq_printf(seq, "gets = %u\n", stats->gets);
2145 seq_printf(seq, "backtracks = %u\n", stats->backtrack);
2146 seq_printf(seq, "semantic match passed = %u\n",
2147 stats->semantic_match_passed);
2148 seq_printf(seq, "semantic match miss = %u\n",
2149 stats->semantic_match_miss);
2150 seq_printf(seq, "null node hit= %u\n", stats->null_node_hit);
2151 seq_printf(seq, "skipped node resize = %u\n\n",
2152 stats->resize_node_skipped);
2154 #endif /* CONFIG_IP_FIB_TRIE_STATS */
2156 static void fib_table_print(struct seq_file *seq, struct fib_table *tb)
2158 if (tb->tb_id == RT_TABLE_LOCAL)
2159 seq_puts(seq, "Local:\n");
2160 else if (tb->tb_id == RT_TABLE_MAIN)
2161 seq_puts(seq, "Main:\n");
2163 seq_printf(seq, "Id %d:\n", tb->tb_id);
2167 static int fib_triestat_seq_show(struct seq_file *seq, void *v)
2169 struct net *net = (struct net *)seq->private;
2173 "Basic info: size of leaf:"
2174 " %Zd bytes, size of tnode: %Zd bytes.\n",
2175 sizeof(struct leaf), sizeof(struct tnode));
2177 for (h = 0; h < FIB_TABLE_HASHSZ; h++) {
2178 struct hlist_head *head = &net->ipv4.fib_table_hash[h];
2179 struct hlist_node *node;
2180 struct fib_table *tb;
2182 hlist_for_each_entry_rcu(tb, node, head, tb_hlist) {
2183 struct trie *t = (struct trie *) tb->tb_data;
2184 struct trie_stat stat;
2189 fib_table_print(seq, tb);
2191 trie_collect_stats(t, &stat);
2192 trie_show_stats(seq, &stat);
2193 #ifdef CONFIG_IP_FIB_TRIE_STATS
2194 trie_show_usage(seq, &t->stats);
2202 static int fib_triestat_seq_open(struct inode *inode, struct file *file)
2204 return single_open_net(inode, file, fib_triestat_seq_show);
2207 static const struct file_operations fib_triestat_fops = {
2208 .owner = THIS_MODULE,
2209 .open = fib_triestat_seq_open,
2211 .llseek = seq_lseek,
2212 .release = single_release_net,
2215 static struct rt_trie_node *fib_trie_get_idx(struct seq_file *seq, loff_t pos)
2217 struct fib_trie_iter *iter = seq->private;
2218 struct net *net = seq_file_net(seq);
2222 for (h = 0; h < FIB_TABLE_HASHSZ; h++) {
2223 struct hlist_head *head = &net->ipv4.fib_table_hash[h];
2224 struct hlist_node *node;
2225 struct fib_table *tb;
2227 hlist_for_each_entry_rcu(tb, node, head, tb_hlist) {
2228 struct rt_trie_node *n;
2230 for (n = fib_trie_get_first(iter,
2231 (struct trie *) tb->tb_data);
2232 n; n = fib_trie_get_next(iter))
2243 static void *fib_trie_seq_start(struct seq_file *seq, loff_t *pos)
2247 return fib_trie_get_idx(seq, *pos);
2250 static void *fib_trie_seq_next(struct seq_file *seq, void *v, loff_t *pos)
2252 struct fib_trie_iter *iter = seq->private;
2253 struct net *net = seq_file_net(seq);
2254 struct fib_table *tb = iter->tb;
2255 struct hlist_node *tb_node;
2257 struct rt_trie_node *n;
2260 /* next node in same table */
2261 n = fib_trie_get_next(iter);
2265 /* walk rest of this hash chain */
2266 h = tb->tb_id & (FIB_TABLE_HASHSZ - 1);
2267 while ( (tb_node = rcu_dereference(tb->tb_hlist.next)) ) {
2268 tb = hlist_entry(tb_node, struct fib_table, tb_hlist);
2269 n = fib_trie_get_first(iter, (struct trie *) tb->tb_data);
2274 /* new hash chain */
2275 while (++h < FIB_TABLE_HASHSZ) {
2276 struct hlist_head *head = &net->ipv4.fib_table_hash[h];
2277 hlist_for_each_entry_rcu(tb, tb_node, head, tb_hlist) {
2278 n = fib_trie_get_first(iter, (struct trie *) tb->tb_data);
2290 static void fib_trie_seq_stop(struct seq_file *seq, void *v)
2296 static void seq_indent(struct seq_file *seq, int n)
2302 static inline const char *rtn_scope(char *buf, size_t len, enum rt_scope_t s)
2305 case RT_SCOPE_UNIVERSE: return "universe";
2306 case RT_SCOPE_SITE: return "site";
2307 case RT_SCOPE_LINK: return "link";
2308 case RT_SCOPE_HOST: return "host";
2309 case RT_SCOPE_NOWHERE: return "nowhere";
2311 snprintf(buf, len, "scope=%d", s);
2316 static const char *const rtn_type_names[__RTN_MAX] = {
2317 [RTN_UNSPEC] = "UNSPEC",
2318 [RTN_UNICAST] = "UNICAST",
2319 [RTN_LOCAL] = "LOCAL",
2320 [RTN_BROADCAST] = "BROADCAST",
2321 [RTN_ANYCAST] = "ANYCAST",
2322 [RTN_MULTICAST] = "MULTICAST",
2323 [RTN_BLACKHOLE] = "BLACKHOLE",
2324 [RTN_UNREACHABLE] = "UNREACHABLE",
2325 [RTN_PROHIBIT] = "PROHIBIT",
2326 [RTN_THROW] = "THROW",
2328 [RTN_XRESOLVE] = "XRESOLVE",
2331 static inline const char *rtn_type(char *buf, size_t len, unsigned int t)
2333 if (t < __RTN_MAX && rtn_type_names[t])
2334 return rtn_type_names[t];
2335 snprintf(buf, len, "type %u", t);
2339 /* Pretty print the trie */
2340 static int fib_trie_seq_show(struct seq_file *seq, void *v)
2342 const struct fib_trie_iter *iter = seq->private;
2343 struct rt_trie_node *n = v;
2345 if (!node_parent_rcu(n))
2346 fib_table_print(seq, iter->tb);
2349 struct tnode *tn = (struct tnode *) n;
2350 __be32 prf = htonl(mask_pfx(tn->key, tn->pos));
2352 seq_indent(seq, iter->depth-1);
2353 seq_printf(seq, " +-- %pI4/%d %d %d %d\n",
2354 &prf, tn->pos, tn->bits, tn->full_children,
2355 tn->empty_children);
2358 struct leaf *l = (struct leaf *) n;
2359 struct leaf_info *li;
2360 struct hlist_node *node;
2361 __be32 val = htonl(l->key);
2363 seq_indent(seq, iter->depth);
2364 seq_printf(seq, " |-- %pI4\n", &val);
2366 hlist_for_each_entry_rcu(li, node, &l->list, hlist) {
2367 struct fib_alias *fa;
2369 list_for_each_entry_rcu(fa, &li->falh, fa_list) {
2370 char buf1[32], buf2[32];
2372 seq_indent(seq, iter->depth+1);
2373 seq_printf(seq, " /%d %s %s", li->plen,
2374 rtn_scope(buf1, sizeof(buf1),
2375 fa->fa_info->fib_scope),
2376 rtn_type(buf2, sizeof(buf2),
2379 seq_printf(seq, " tos=%d", fa->fa_tos);
2380 seq_putc(seq, '\n');
2388 static const struct seq_operations fib_trie_seq_ops = {
2389 .start = fib_trie_seq_start,
2390 .next = fib_trie_seq_next,
2391 .stop = fib_trie_seq_stop,
2392 .show = fib_trie_seq_show,
2395 static int fib_trie_seq_open(struct inode *inode, struct file *file)
2397 return seq_open_net(inode, file, &fib_trie_seq_ops,
2398 sizeof(struct fib_trie_iter));
2401 static const struct file_operations fib_trie_fops = {
2402 .owner = THIS_MODULE,
2403 .open = fib_trie_seq_open,
2405 .llseek = seq_lseek,
2406 .release = seq_release_net,
2409 struct fib_route_iter {
2410 struct seq_net_private p;
2411 struct trie *main_trie;
2416 static struct leaf *fib_route_get_idx(struct fib_route_iter *iter, loff_t pos)
2418 struct leaf *l = NULL;
2419 struct trie *t = iter->main_trie;
2421 /* use cache location of last found key */
2422 if (iter->pos > 0 && pos >= iter->pos && (l = fib_find_node(t, iter->key)))
2426 l = trie_firstleaf(t);
2429 while (l && pos-- > 0) {
2431 l = trie_nextleaf(l);
2435 iter->key = pos; /* remember it */
2437 iter->pos = 0; /* forget it */
2442 static void *fib_route_seq_start(struct seq_file *seq, loff_t *pos)
2445 struct fib_route_iter *iter = seq->private;
2446 struct fib_table *tb;
2449 tb = fib_get_table(seq_file_net(seq), RT_TABLE_MAIN);
2453 iter->main_trie = (struct trie *) tb->tb_data;
2455 return SEQ_START_TOKEN;
2457 return fib_route_get_idx(iter, *pos - 1);
2460 static void *fib_route_seq_next(struct seq_file *seq, void *v, loff_t *pos)
2462 struct fib_route_iter *iter = seq->private;
2466 if (v == SEQ_START_TOKEN) {
2468 l = trie_firstleaf(iter->main_trie);
2471 l = trie_nextleaf(l);
2481 static void fib_route_seq_stop(struct seq_file *seq, void *v)
2487 static unsigned int fib_flag_trans(int type, __be32 mask, const struct fib_info *fi)
2489 unsigned int flags = 0;
2491 if (type == RTN_UNREACHABLE || type == RTN_PROHIBIT)
2493 if (fi && fi->fib_nh->nh_gw)
2494 flags |= RTF_GATEWAY;
2495 if (mask == htonl(0xFFFFFFFF))
2502 * This outputs /proc/net/route.
2503 * The format of the file is not supposed to be changed
2504 * and needs to be same as fib_hash output to avoid breaking
2507 static int fib_route_seq_show(struct seq_file *seq, void *v)
2510 struct leaf_info *li;
2511 struct hlist_node *node;
2513 if (v == SEQ_START_TOKEN) {
2514 seq_printf(seq, "%-127s\n", "Iface\tDestination\tGateway "
2515 "\tFlags\tRefCnt\tUse\tMetric\tMask\t\tMTU"
2520 hlist_for_each_entry_rcu(li, node, &l->list, hlist) {
2521 struct fib_alias *fa;
2522 __be32 mask, prefix;
2524 mask = inet_make_mask(li->plen);
2525 prefix = htonl(l->key);
2527 list_for_each_entry_rcu(fa, &li->falh, fa_list) {
2528 const struct fib_info *fi = fa->fa_info;
2529 unsigned int flags = fib_flag_trans(fa->fa_type, mask, fi);
2532 if (fa->fa_type == RTN_BROADCAST
2533 || fa->fa_type == RTN_MULTICAST)
2538 "%s\t%08X\t%08X\t%04X\t%d\t%u\t"
2539 "%d\t%08X\t%d\t%u\t%u%n",
2540 fi->fib_dev ? fi->fib_dev->name : "*",
2542 fi->fib_nh->nh_gw, flags, 0, 0,
2546 fi->fib_advmss + 40 : 0),
2548 fi->fib_rtt >> 3, &len);
2551 "*\t%08X\t%08X\t%04X\t%d\t%u\t"
2552 "%d\t%08X\t%d\t%u\t%u%n",
2553 prefix, 0, flags, 0, 0, 0,
2554 mask, 0, 0, 0, &len);
2556 seq_printf(seq, "%*s\n", 127 - len, "");
2563 static const struct seq_operations fib_route_seq_ops = {
2564 .start = fib_route_seq_start,
2565 .next = fib_route_seq_next,
2566 .stop = fib_route_seq_stop,
2567 .show = fib_route_seq_show,
2570 static int fib_route_seq_open(struct inode *inode, struct file *file)
2572 return seq_open_net(inode, file, &fib_route_seq_ops,
2573 sizeof(struct fib_route_iter));
2576 static const struct file_operations fib_route_fops = {
2577 .owner = THIS_MODULE,
2578 .open = fib_route_seq_open,
2580 .llseek = seq_lseek,
2581 .release = seq_release_net,
2584 int __net_init fib_proc_init(struct net *net)
2586 if (!proc_net_fops_create(net, "fib_trie", S_IRUGO, &fib_trie_fops))
2589 if (!proc_net_fops_create(net, "fib_triestat", S_IRUGO,
2590 &fib_triestat_fops))
2593 if (!proc_net_fops_create(net, "route", S_IRUGO, &fib_route_fops))
2599 proc_net_remove(net, "fib_triestat");
2601 proc_net_remove(net, "fib_trie");
2606 void __net_exit fib_proc_exit(struct net *net)
2608 proc_net_remove(net, "fib_trie");
2609 proc_net_remove(net, "fib_triestat");
2610 proc_net_remove(net, "route");
2613 #endif /* CONFIG_PROC_FS */