diff options
Diffstat (limited to 'net/ipv4/fib_trie.c')
-rw-r--r-- | net/ipv4/fib_trie.c | 272 |
1 files changed, 116 insertions, 156 deletions
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 0f280348e0f..3d28a35c2e1 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -95,7 +95,7 @@ typedef unsigned int t_key; #define IS_TNODE(n) (!(n->parent & T_LEAF)) #define IS_LEAF(n) (n->parent & T_LEAF) -struct node { +struct rt_trie_node { unsigned long parent; t_key key; }; @@ -126,7 +126,7 @@ struct tnode { struct work_struct work; struct tnode *tnode_free; }; - struct node *child[0]; + struct rt_trie_node *child[0]; }; #ifdef CONFIG_IP_FIB_TRIE_STATS @@ -151,16 +151,16 @@ struct trie_stat { }; struct trie { - struct node *trie; + struct rt_trie_node *trie; #ifdef CONFIG_IP_FIB_TRIE_STATS struct trie_use_stats stats; #endif }; -static void put_child(struct trie *t, struct tnode *tn, int i, struct node *n); -static void tnode_put_child_reorg(struct tnode *tn, int i, struct node *n, +static void put_child(struct trie *t, struct tnode *tn, int i, struct rt_trie_node *n); +static void tnode_put_child_reorg(struct tnode *tn, int i, struct rt_trie_node *n, int wasfull); -static struct node *resize(struct trie *t, struct tnode *tn); +static struct rt_trie_node *resize(struct trie *t, struct tnode *tn); static struct tnode *inflate(struct trie *t, struct tnode *tn); static struct tnode *halve(struct trie *t, struct tnode *tn); /* tnodes to free after resize(); protected by RTNL */ @@ -177,12 +177,12 @@ static const int sync_pages = 128; static struct kmem_cache *fn_alias_kmem __read_mostly; static struct kmem_cache *trie_leaf_kmem __read_mostly; -static inline struct tnode *node_parent(struct node *node) +static inline struct tnode *node_parent(struct rt_trie_node *node) { return (struct tnode *)(node->parent & ~NODE_TYPE_MASK); } -static inline struct tnode *node_parent_rcu(struct node *node) +static inline struct tnode *node_parent_rcu(struct rt_trie_node *node) { struct tnode *ret = node_parent(node); @@ -192,22 +192,22 @@ static inline struct tnode *node_parent_rcu(struct node *node) /* Same as rcu_assign_pointer * but that macro() assumes that value is a pointer. */ -static inline void node_set_parent(struct node *node, struct tnode *ptr) +static inline void node_set_parent(struct rt_trie_node *node, struct tnode *ptr) { smp_wmb(); node->parent = (unsigned long)ptr | NODE_TYPE(node); } -static inline struct node *tnode_get_child(struct tnode *tn, unsigned int i) +static inline struct rt_trie_node *tnode_get_child(struct tnode *tn, unsigned int i) { BUG_ON(i >= 1U << tn->bits); return tn->child[i]; } -static inline struct node *tnode_get_child_rcu(struct tnode *tn, unsigned int i) +static inline struct rt_trie_node *tnode_get_child_rcu(struct tnode *tn, unsigned int i) { - struct node *ret = tnode_get_child(tn, i); + struct rt_trie_node *ret = tnode_get_child(tn, i); return rcu_dereference_rtnl(ret); } @@ -217,12 +217,12 @@ static inline int tnode_child_length(const struct tnode *tn) return 1 << tn->bits; } -static inline t_key mask_pfx(t_key k, unsigned short l) +static inline t_key mask_pfx(t_key k, unsigned int l) { return (l == 0) ? 0 : k >> (KEYLENGTH-l) << (KEYLENGTH-l); } -static inline t_key tkey_extract_bits(t_key a, int offset, int bits) +static inline t_key tkey_extract_bits(t_key a, unsigned int offset, unsigned int bits) { if (offset < KEYLENGTH) return ((t_key)(a << offset)) >> (KEYLENGTH - bits); @@ -378,7 +378,7 @@ static void __tnode_free_rcu(struct rcu_head *head) { struct tnode *tn = container_of(head, struct tnode, rcu); size_t size = sizeof(struct tnode) + - (sizeof(struct node *) << tn->bits); + (sizeof(struct rt_trie_node *) << tn->bits); if (size <= PAGE_SIZE) kfree(tn); @@ -402,7 +402,7 @@ static void tnode_free_safe(struct tnode *tn) tn->tnode_free = tnode_free_head; tnode_free_head = tn; tnode_free_size += sizeof(struct tnode) + - (sizeof(struct node *) << tn->bits); + (sizeof(struct rt_trie_node *) << tn->bits); } static void tnode_free_flush(void) @@ -443,7 +443,7 @@ static struct leaf_info *leaf_info_new(int plen) static struct tnode *tnode_new(t_key key, int pos, int bits) { - size_t sz = sizeof(struct tnode) + (sizeof(struct node *) << bits); + size_t sz = sizeof(struct tnode) + (sizeof(struct rt_trie_node *) << bits); struct tnode *tn = tnode_alloc(sz); if (tn) { @@ -456,7 +456,7 @@ static struct tnode *tnode_new(t_key key, int pos, int bits) } pr_debug("AT %p s=%zu %zu\n", tn, sizeof(struct tnode), - sizeof(struct node) << bits); + sizeof(struct rt_trie_node) << bits); return tn; } @@ -465,7 +465,7 @@ static struct tnode *tnode_new(t_key key, int pos, int bits) * and no bits are skipped. See discussion in dyntree paper p. 6 */ -static inline int tnode_full(const struct tnode *tn, const struct node *n) +static inline int tnode_full(const struct tnode *tn, const struct rt_trie_node *n) { if (n == NULL || IS_LEAF(n)) return 0; @@ -474,7 +474,7 @@ static inline int tnode_full(const struct tnode *tn, const struct node *n) } static inline void put_child(struct trie *t, struct tnode *tn, int i, - struct node *n) + struct rt_trie_node *n) { tnode_put_child_reorg(tn, i, n, -1); } @@ -484,10 +484,10 @@ static inline void put_child(struct trie *t, struct tnode *tn, int i, * Update the value of full_children and empty_children. */ -static void tnode_put_child_reorg(struct tnode *tn, int i, struct node *n, +static void tnode_put_child_reorg(struct tnode *tn, int i, struct rt_trie_node *n, int wasfull) { - struct node *chi = tn->child[i]; + struct rt_trie_node *chi = tn->child[i]; int isfull; BUG_ON(i >= 1<<tn->bits); @@ -515,7 +515,7 @@ static void tnode_put_child_reorg(struct tnode *tn, int i, struct node *n, } #define MAX_WORK 10 -static struct node *resize(struct trie *t, struct tnode *tn) +static struct rt_trie_node *resize(struct trie *t, struct tnode *tn) { int i; struct tnode *old_tn; @@ -605,7 +605,7 @@ static struct node *resize(struct trie *t, struct tnode *tn) /* Keep root node larger */ - if (!node_parent((struct node *)tn)) { + if (!node_parent((struct rt_trie_node *)tn)) { inflate_threshold_use = inflate_threshold_root; halve_threshold_use = halve_threshold_root; } else { @@ -635,7 +635,7 @@ static struct node *resize(struct trie *t, struct tnode *tn) /* Return if at least one inflate is run */ if (max_work != MAX_WORK) - return (struct node *) tn; + return (struct rt_trie_node *) tn; /* * Halve as long as the number of empty children in this @@ -663,7 +663,7 @@ static struct node *resize(struct trie *t, struct tnode *tn) if (tn->empty_children == tnode_child_length(tn) - 1) { one_child: for (i = 0; i < tnode_child_length(tn); i++) { - struct node *n; + struct rt_trie_node *n; n = tn->child[i]; if (!n) @@ -676,7 +676,7 @@ one_child: return n; } } - return (struct node *) tn; + return (struct rt_trie_node *) tn; } static struct tnode *inflate(struct trie *t, struct tnode *tn) @@ -723,14 +723,14 @@ static struct tnode *inflate(struct trie *t, struct tnode *tn) goto nomem; } - put_child(t, tn, 2*i, (struct node *) left); - put_child(t, tn, 2*i+1, (struct node *) right); + put_child(t, tn, 2*i, (struct rt_trie_node *) left); + put_child(t, tn, 2*i+1, (struct rt_trie_node *) right); } } for (i = 0; i < olen; i++) { struct tnode *inode; - struct node *node = tnode_get_child(oldtnode, i); + struct rt_trie_node *node = tnode_get_child(oldtnode, i); struct tnode *left, *right; int size, j; @@ -825,7 +825,7 @@ nomem: static struct tnode *halve(struct trie *t, struct tnode *tn) { struct tnode *oldtnode = tn; - struct node *left, *right; + struct rt_trie_node *left, *right; int i; int olen = tnode_child_length(tn); @@ -856,7 +856,7 @@ static struct tnode *halve(struct trie *t, struct tnode *tn) if (!newn) goto nomem; - put_child(t, tn, i/2, (struct node *)newn); + put_child(t, tn, i/2, (struct rt_trie_node *)newn); } } @@ -958,7 +958,7 @@ fib_find_node(struct trie *t, u32 key) { int pos; struct tnode *tn; - struct node *n; + struct rt_trie_node *n; pos = 0; n = rcu_dereference_rtnl(t->trie); @@ -993,17 +993,17 @@ static void trie_rebalance(struct trie *t, struct tnode *tn) key = tn->key; - while (tn != NULL && (tp = node_parent((struct node *)tn)) != NULL) { + while (tn != NULL && (tp = node_parent((struct rt_trie_node *)tn)) != NULL) { cindex = tkey_extract_bits(key, tp->pos, tp->bits); wasfull = tnode_full(tp, tnode_get_child(tp, cindex)); tn = (struct tnode *) resize(t, (struct tnode *)tn); tnode_put_child_reorg((struct tnode *)tp, cindex, - (struct node *)tn, wasfull); + (struct rt_trie_node *)tn, wasfull); - tp = node_parent((struct node *) tn); + tp = node_parent((struct rt_trie_node *) tn); if (!tp) - rcu_assign_pointer(t->trie, (struct node *)tn); + rcu_assign_pointer(t->trie, (struct rt_trie_node *)tn); tnode_free_flush(); if (!tp) @@ -1015,7 +1015,7 @@ static void trie_rebalance(struct trie *t, struct tnode *tn) if (IS_TNODE(tn)) tn = (struct tnode *)resize(t, (struct tnode *)tn); - rcu_assign_pointer(t->trie, (struct node *)tn); + rcu_assign_pointer(t->trie, (struct rt_trie_node *)tn); tnode_free_flush(); } @@ -1025,7 +1025,7 @@ static struct list_head *fib_insert_node(struct trie *t, u32 key, int plen) { int pos, newpos; struct tnode *tp = NULL, *tn = NULL; - struct node *n; + struct rt_trie_node *n; struct leaf *l; int missbit; struct list_head *fa_head = NULL; @@ -1111,10 +1111,10 @@ static struct list_head *fib_insert_node(struct trie *t, u32 key, int plen) if (t->trie && n == NULL) { /* Case 2: n is NULL, and will just insert a new leaf */ - node_set_parent((struct node *)l, tp); + node_set_parent((struct rt_trie_node *)l, tp); cindex = tkey_extract_bits(key, tp->pos, tp->bits); - put_child(t, (struct tnode *)tp, cindex, (struct node *)l); + put_child(t, (struct tnode *)tp, cindex, (struct rt_trie_node *)l); } else { /* Case 3: n is a LEAF or a TNODE and the key doesn't match. */ /* @@ -1141,18 +1141,18 @@ static struct list_head *fib_insert_node(struct trie *t, u32 key, int plen) return NULL; } - node_set_parent((struct node *)tn, tp); + node_set_parent((struct rt_trie_node *)tn, tp); missbit = tkey_extract_bits(key, newpos, 1); - put_child(t, tn, missbit, (struct node *)l); + put_child(t, tn, missbit, (struct rt_trie_node *)l); put_child(t, tn, 1-missbit, n); if (tp) { cindex = tkey_extract_bits(key, tp->pos, tp->bits); put_child(t, (struct tnode *)tp, cindex, - (struct node *)tn); + (struct rt_trie_node *)tn); } else { - rcu_assign_pointer(t->trie, (struct node *)tn); + rcu_assign_pointer(t->trie, (struct rt_trie_node *)tn); tp = tn; } } @@ -1340,8 +1340,8 @@ err: } /* should be called with rcu_read_lock */ -static int check_leaf(struct trie *t, struct leaf *l, - t_key key, const struct flowi *flp, +static int check_leaf(struct fib_table *tb, struct trie *t, struct leaf *l, + t_key key, const struct flowi4 *flp, struct fib_result *res, int fib_flags) { struct leaf_info *li; @@ -1349,40 +1349,75 @@ static int check_leaf(struct trie *t, struct leaf *l, struct hlist_node *node; hlist_for_each_entry_rcu(li, node, hhead, hlist) { - int err; + struct fib_alias *fa; int plen = li->plen; __be32 mask = inet_make_mask(plen); if (l->key != (key & ntohl(mask))) continue; - err = fib_semantic_match(&li->falh, flp, res, plen, fib_flags); + list_for_each_entry_rcu(fa, &li->falh, fa_list) { + struct fib_info *fi = fa->fa_info; + int nhsel, err; + if (fa->fa_tos && fa->fa_tos != flp->flowi4_tos) + continue; + if (fa->fa_scope < flp->flowi4_scope) + continue; + fib_alias_accessed(fa); + err = fib_props[fa->fa_type].error; + if (err) { #ifdef CONFIG_IP_FIB_TRIE_STATS - if (err <= 0) - t->stats.semantic_match_passed++; - else - t->stats.semantic_match_miss++; + t->stats.semantic_match_miss++; +#endif + return 1; + } + if (fi->fib_flags & RTNH_F_DEAD) + continue; + for (nhsel = 0; nhsel < fi->fib_nhs; nhsel++) { + const struct fib_nh *nh = &fi->fib_nh[nhsel]; + + if (nh->nh_flags & RTNH_F_DEAD) + continue; + if (flp->flowi4_oif && flp->flowi4_oif != nh->nh_oif) + continue; + +#ifdef CONFIG_IP_FIB_TRIE_STATS + t->stats.semantic_match_passed++; +#endif + res->prefixlen = plen; + res->nh_sel = nhsel; + res->type = fa->fa_type; + res->scope = fa->fa_scope; + res->fi = fi; + res->table = tb; + res->fa_head = &li->falh; + if (!(fib_flags & FIB_LOOKUP_NOREF)) + atomic_inc(&res->fi->fib_clntref); + return 0; + } + } + +#ifdef CONFIG_IP_FIB_TRIE_STATS + t->stats.semantic_match_miss++; #endif - if (err <= 0) - return err; } return 1; } -int fib_table_lookup(struct fib_table *tb, const struct flowi *flp, +int fib_table_lookup(struct fib_table *tb, const struct flowi4 *flp, struct fib_result *res, int fib_flags) { struct trie *t = (struct trie *) tb->tb_data; int ret; - struct node *n; + struct rt_trie_node *n; struct tnode *pn; - int pos, bits; - t_key key = ntohl(flp->fl4_dst); - int chopped_off; + unsigned int pos, bits; + t_key key = ntohl(flp->daddr); + unsigned int chopped_off; t_key cindex = 0; - int current_prefix_length = KEYLENGTH; + unsigned int current_prefix_length = KEYLENGTH; struct tnode *cn; t_key pref_mismatch; @@ -1398,7 +1433,7 @@ int fib_table_lookup(struct fib_table *tb, const struct flowi *flp, /* Just a leaf? */ if (IS_LEAF(n)) { - ret = check_leaf(t, (struct leaf *)n, key, flp, res, fib_flags); + ret = check_leaf(tb, t, (struct leaf *)n, key, flp, res, fib_flags); goto found; } @@ -1423,7 +1458,7 @@ int fib_table_lookup(struct fib_table *tb, const struct flowi *flp, } if (IS_LEAF(n)) { - ret = check_leaf(t, (struct leaf *)n, key, flp, res, fib_flags); + ret = check_leaf(tb, t, (struct leaf *)n, key, flp, res, fib_flags); if (ret > 0) goto backtrace; goto found; @@ -1541,7 +1576,7 @@ backtrace: if (chopped_off <= pn->bits) { cindex &= ~(1 << (chopped_off-1)); } else { - struct tnode *parent = node_parent_rcu((struct node *) pn); + struct tnode *parent = node_parent_rcu((struct rt_trie_node *) pn); if (!parent) goto failed; @@ -1568,7 +1603,7 @@ found: */ static void trie_leaf_remove(struct trie *t, struct leaf *l) { - struct tnode *tp = node_parent((struct node *) l); + struct tnode *tp = node_parent((struct rt_trie_node *) l); pr_debug("entering trie_leaf_remove(%p)\n", l); @@ -1706,7 +1741,7 @@ static int trie_flush_leaf(struct leaf *l) * Scan for the next right leaf starting at node p->child[idx] * Since we have back pointer, no recursion necessary. */ -static struct leaf *leaf_walk_rcu(struct tnode *p, struct node *c) +static struct leaf *leaf_walk_rcu(struct tnode *p, struct rt_trie_node *c) { do { t_key idx; @@ -1732,7 +1767,7 @@ static struct leaf *leaf_walk_rcu(struct tnode *p, struct node *c) } /* Node empty, walk back up to parent */ - c = (struct node *) p; + c = (struct rt_trie_node *) p; } while ((p = node_parent_rcu(c)) != NULL); return NULL; /* Root of trie */ @@ -1753,7 +1788,7 @@ static struct leaf *trie_firstleaf(struct trie *t) static struct leaf *trie_nextleaf(struct leaf *l) { - struct node *c = (struct node *) l; + struct rt_trie_node *c = (struct rt_trie_node *) l; struct tnode *p = node_parent_rcu(c); if (!p) @@ -1802,80 +1837,6 @@ void fib_free_table(struct fib_table *tb) kfree(tb); } -void fib_table_select_default(struct fib_table *tb, - const struct flowi *flp, - struct fib_result *res) -{ - struct trie *t = (struct trie *) tb->tb_data; - int order, last_idx; - struct fib_info *fi = NULL; - struct fib_info *last_resort; - struct fib_alias *fa = NULL; - struct list_head *fa_head; - struct leaf *l; - - last_idx = -1; - last_resort = NULL; - order = -1; - - rcu_read_lock(); - - l = fib_find_node(t, 0); - if (!l) - goto out; - - fa_head = get_fa_head(l, 0); - if (!fa_head) - goto out; - - if (list_empty(fa_head)) - goto out; - - list_for_each_entry_rcu(fa, fa_head, fa_list) { - struct fib_info *next_fi = fa->fa_info; - - if (fa->fa_scope != res->scope || - fa->fa_type != RTN_UNICAST) - continue; - - if (next_fi->fib_priority > res->fi->fib_priority) - break; - if (!next_fi->fib_nh[0].nh_gw || - next_fi->fib_nh[0].nh_scope != RT_SCOPE_LINK) - continue; - - fib_alias_accessed(fa); - - if (fi == NULL) { - if (next_fi != res->fi) - break; - } else if (!fib_detect_death(fi, order, &last_resort, - &last_idx, tb->tb_default)) { - fib_result_assign(res, fi); - tb->tb_default = order; - goto out; - } - fi = next_fi; - order++; - } - if (order <= 0 || fi == NULL) { - tb->tb_default = -1; - goto out; - } - - if (!fib_detect_death(fi, order, &last_resort, &last_idx, - tb->tb_default)) { - fib_result_assign(res, fi); - tb->tb_default = order; - goto out; - } - if (last_idx >= 0) - fib_result_assign(res, last_resort); - tb->tb_default = last_idx; -out: - rcu_read_unlock(); -} - static int fn_trie_dump_fa(t_key key, int plen, struct list_head *fah, struct fib_table *tb, struct sk_buff *skb, struct netlink_callback *cb) @@ -1990,7 +1951,7 @@ int fib_table_dump(struct fib_table *tb, struct sk_buff *skb, return skb->len; } -void __init fib_hash_init(void) +void __init fib_trie_init(void) { fn_alias_kmem = kmem_cache_create("ip_fib_alias", sizeof(struct fib_alias), @@ -2003,8 +1964,7 @@ void __init fib_hash_init(void) } -/* Fix more generic FIB names for init later */ -struct fib_table *fib_hash_table(u32 id) +struct fib_table *fib_trie_table(u32 id) { struct fib_table *tb; struct trie *t; @@ -2036,7 +1996,7 @@ struct fib_trie_iter { unsigned int depth; }; -static struct node *fib_trie_get_next(struct fib_trie_iter *iter) +static struct rt_trie_node *fib_trie_get_next(struct fib_trie_iter *iter) { struct tnode *tn = iter->tnode; unsigned int cindex = iter->index; @@ -2050,7 +2010,7 @@ static struct node *fib_trie_get_next(struct fib_trie_iter *iter) iter->tnode, iter->index, iter->depth); rescan: while (cindex < (1<<tn->bits)) { - struct node *n = tnode_get_child_rcu(tn, cindex); + struct rt_trie_node *n = tnode_get_child_rcu(tn, cindex); if (n) { if (IS_LEAF(n)) { @@ -2069,7 +2029,7 @@ rescan: } /* Current node exhausted, pop back up */ - p = node_parent_rcu((struct node *)tn); + p = node_parent_rcu((struct rt_trie_node *)tn); if (p) { cindex = tkey_extract_bits(tn->key, p->pos, p->bits)+1; tn = p; @@ -2081,10 +2041,10 @@ rescan: return NULL; } -static struct node *fib_trie_get_first(struct fib_trie_iter *iter, +static struct rt_trie_node *fib_trie_get_first(struct fib_trie_iter *iter, struct trie *t) { - struct node *n; + struct rt_trie_node *n; if (!t) return NULL; @@ -2108,7 +2068,7 @@ static struct node *fib_trie_get_first(struct fib_trie_iter *iter, static void trie_collect_stats(struct trie *t, struct trie_stat *s) { - struct node *n; + struct rt_trie_node *n; struct fib_trie_iter iter; memset(s, 0, sizeof(*s)); @@ -2181,7 +2141,7 @@ static void trie_show_stats(struct seq_file *seq, struct trie_stat *stat) seq_putc(seq, '\n'); seq_printf(seq, "\tPointers: %u\n", pointers); - bytes += sizeof(struct node *) * pointers; + bytes += sizeof(struct rt_trie_node *) * pointers; seq_printf(seq, "Null ptrs: %u\n", stat->nullpointers); seq_printf(seq, "Total size: %u kB\n", (bytes + 1023) / 1024); } @@ -2262,7 +2222,7 @@ static const struct file_operations fib_triestat_fops = { .release = single_release_net, }; -static struct node *fib_trie_get_idx(struct seq_file *seq, loff_t pos) +static struct rt_trie_node *fib_trie_get_idx(struct seq_file *seq, loff_t pos) { struct fib_trie_iter *iter = seq->private; struct net *net = seq_file_net(seq); @@ -2275,7 +2235,7 @@ static struct node *fib_trie_get_idx(struct seq_file *seq, loff_t pos) struct fib_table *tb; hlist_for_each_entry_rcu(tb, node, head, tb_hlist) { - struct node *n; + struct rt_trie_node *n; for (n = fib_trie_get_first(iter, (struct trie *) tb->tb_data); @@ -2304,7 +2264,7 @@ static void *fib_trie_seq_next(struct seq_file *seq, void *v, loff_t *pos) struct fib_table *tb = iter->tb; struct hlist_node *tb_node; unsigned int h; - struct node *n; + struct rt_trie_node *n; ++*pos; /* next node in same table */ @@ -2390,7 +2350,7 @@ static inline const char *rtn_type(char *buf, size_t len, unsigned int t) static int fib_trie_seq_show(struct seq_file *seq, void *v) { const struct fib_trie_iter *iter = seq->private; - struct node *n = v; + struct rt_trie_node *n = v; if (!node_parent_rcu(n)) fib_table_print(seq, iter->tb); |