1.最近学习linux 路由表相关知识现在总结如下：
系统查询路由信息的时候，分为两部先查路由缓存信息，然后查询路由表，route cash信息保存在一个全局的数据结构 rt_hash_table
net/ipv4/ af_inet.c inet_init() -> net/ipv4/ ip_output.c ip_init() -> net/ipv4/ route.c ip_rt_init()
rt_hash_table = (struct rt_hash_bucket *) alloc_large_system_hash("IP route cache", sizeof(struct rt_hash_bucket), rhash_entries, (totalram_pages >= 128 * 1024) ? 15 : 17, 0, &rt_hash_log, &rt_hash_mask, rhash_entries ? 0 : 512 * 1024);
rt_hash_table
| chain |
| chain |
| chain |
------>rt_table---->rt_table------>rt_table
-------->rt_table------->rt_table-------->rt_table
这个是route cache结构数组和一个链表存储
策略路由表结构
net->ipv4.fib_table_hash = kzalloc( sizeof(struct hlist_head)*FIB_TABLE_HASHSZ, GFP_KERNEL); if (net->ipv4.fib_table_hash == NULL) return -ENOMEM; for (i = 0; i < FIB_TABLE_HASHSZ; i++) INIT_HLIST_HEAD(&net->ipv4.fib_table_hash[i]);
注意 根据不同的network namespce和协议都有不同的路由表 比如ipv4和ipv6
查看一个函数fib_get_table
head = &net->ipv4.fib_table_hash[h]; //hash clash hlist_for_each_entry_rcu(tb, node, head, tb_hlist) { if (tb->tb_id == id) { rcu_read_unlock(); return tb; } }
可以看到查找table的时候先是根据hash表来计算，然后在查找冲突
最后谈谈net 这个结构体的
struct net inet_net;
这个结构体定义为network namespace
struct net { atomic_t count; /* To decided when the network * namespace should be freed. */ #ifdef NETNS_REFCNT_DEBUG atomic_t use_count; /* To track references we * destroy on demand */ #endif struct list_head list; /* list of network namespaces */ struct work_struct work; /* work struct for freeing */ struct proc_dir_entry *proc_net; struct proc_dir_entry *proc_net_stat; #ifdef CONFIG_SYSCTL struct ctl_table_set sysctls; #endif struct net_device *loopback_dev; /* The loopback */ struct list_head dev_base_head; struct hlist_head *dev_name_head; struct hlist_head *dev_index_head; /* core fib_rules */ struct list_head rules_ops; spinlock_t rules_mod_lock; struct sock *rtnl; /* rtnetlink socket */ struct sock *genl_sock; struct netns_core core; struct netns_mib mib; struct netns_packet packet; struct netns_unix unx; struct netns_ipv4 ipv4; #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) struct netns_ipv6 ipv6; #endif #if defined(CONFIG_IP_DCCP) || defined(CONFIG_IP_DCCP_MODULE) struct netns_dccp dccp; #endif #ifdef CONFIG_NETFILTER struct netns_xt xt; #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) struct netns_ct ct; #endif #endif #ifdef CONFIG_XFRM struct netns_xfrm xfrm; #endif #ifdef CONFIG_WIRELESS_EXT struct sk_buff_head wext_nlevents; #endif struct net_generic *gen; };
struct net 结构体中和ipv4路由表相关的有
struct list_head rules_ops; //这个是和协议族相关的对策略路由表操作的函数rule_ops 每一个协议族对应一个ops主要用于策略路由表的添加删除匹配等,以及策略路由描述信息
struct netns_ipv4 ipv4; //这个表示ipv4协议族中的路由表相关信息
struct netns ipv4 中有两个重要成员
struct fib_rules_ops *rules_ops; // 策略路由表的描述信息 比如 dev br0 mark 2 的查路由表2
struct hlist_head *fib_table_hash; //256张路由表 指定出口下一调
我们来看下rules_ops这个结构体体
struct fib_rules_ops { int family; struct list_head list; int rule_size; int addr_size; int unresolved_rules; int nr_goto_rules; int (*action)(struct fib_rule *, struct flowi *, int, struct fib_lookup_arg *); int (*match)(struct fib_rule *, struct flowi *, int); int (*configure)(struct fib_rule *, struct sk_buff *, struct fib_rule_hdr *, struct nlattr **); int (*compare)(struct fib_rule *, struct fib_rule_hdr *, struct nlattr **); int (*fill)(struct fib_rule *, struct sk_buff *, struct fib_rule_hdr *); u32 (*default_pref)(struct fib_rules_ops *ops); size_t (*nlmsg_payload)(struct fib_rule *); /* Called after modifications to the rules set, must flush * the route cache if one exists. */ void (*flush_cache)(struct fib_rules_ops *ops); int nlgroup; const struct nla_policy *policy; struct list_head rules_list; struct module *owner; struct net *fro_net; };
其中主要成员说明
int family 表示协议族
struct list_head rules_list; 策略路由表的描述 mark 2查找table 2 ; mark 3 查找table3 策略路由表的描述信息
struct net *fro_net; 指向net 通过net 和table Id 找到所需要的table
现在我们来看看茶表的过程
list_for_each_entry_rcu(rule, &ops->rules_list, list) { jumped: if (!fib_rule_match(rule, ops, fl, flags)) continue; if (rule->action == FR_ACT_GOTO) { struct fib_rule *target; target = rcu_dereference(rule->ctarget); if (target == NULL) { continue; } else { rule = target; goto jumped; } } else if (rule->action == FR_ACT_NOP) continue; else err = ops->action(rule, fl, flags, arg); if (err != -EAGAIN) { fib_rule_get(rule); arg->rule = rule; goto out; } }
fib_rule_match 就是根据策略路由表的信息找到需要查寻哪一张策略路由表
ops->action 就是根据策略路由表决定下一跳信息，这个函数主要分两部根据rule的table id 找到特定的table rule->table rule->net->ipv4->rt_hash_table，这两个来确定的
第二部分就是决定下一调
if ((tbl = fib_get_table(rule->fr_net, rule->table)) == NULL) { if ( rule->table < 254 ) { err = 1; } goto errout; } err = tbl->tb_lookup(tbl, flp, (struct fib_result *) arg->result); if (err > 0) { if ( rule->table < 254 ) { err = 1; } else { err = -EAGAIN; } }