[net-next-2.6.git] / net / core / fib_rules.c

/*
 * net/core/fib_rules.c		Generic Routing Rules
 *
 *	This program is free software; you can redistribute it and/or
 *	modify it under the terms of the GNU General Public License as
 *	published by the Free Software Foundation, version 2.
 *
 * Authors:	Thomas Graf <tgraf@suug.ch>
 */

#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/list.h>
#include <net/fib_rules.h>

static LIST_HEAD(rules_ops);
static DEFINE_SPINLOCK(rules_mod_lock);

static void notify_rule_change(int event, struct fib_rule *rule,
			       struct fib_rules_ops *ops, struct nlmsghdr *nlh,
			       u32 pid);

static struct fib_rules_ops *lookup_rules_ops(int family)
{
	struct fib_rules_ops *ops;

	rcu_read_lock();
	list_for_each_entry_rcu(ops, &rules_ops, list) {
		if (ops->family == family) {
			if (!try_module_get(ops->owner))
				ops = NULL;
			rcu_read_unlock();
			return ops;
		}
	}
	rcu_read_unlock();

	return NULL;
}

static void rules_ops_put(struct fib_rules_ops *ops)
{
	if (ops)
		module_put(ops->owner);
}

int fib_rules_register(struct fib_rules_ops *ops)
{
	int err = -EEXIST;
	struct fib_rules_ops *o;

	if (ops->rule_size < sizeof(struct fib_rule))
		return -EINVAL;

	if (ops->match == NULL || ops->configure == NULL ||
	    ops->compare == NULL || ops->fill == NULL ||
	    ops->action == NULL)
		return -EINVAL;

	spin_lock(&rules_mod_lock);
	list_for_each_entry(o, &rules_ops, list)
		if (ops->family == o->family)
			goto errout;

	list_add_tail_rcu(&ops->list, &rules_ops);
	err = 0;
errout:
	spin_unlock(&rules_mod_lock);

	return err;
}

EXPORT_SYMBOL_GPL(fib_rules_register);

static void cleanup_ops(struct fib_rules_ops *ops)
{
	struct fib_rule *rule, *tmp;

	list_for_each_entry_safe(rule, tmp, ops->rules_list, list) {
		list_del_rcu(&rule->list);
		fib_rule_put(rule);
	}
}

int fib_rules_unregister(struct fib_rules_ops *ops)
{
	int err = 0;
	struct fib_rules_ops *o;

	spin_lock(&rules_mod_lock);
	list_for_each_entry(o, &rules_ops, list) {
		if (o == ops) {
			list_del_rcu(&o->list);
			cleanup_ops(ops);
			goto out;
		}
	}

	err = -ENOENT;
out:
	spin_unlock(&rules_mod_lock);

	synchronize_rcu();

	return err;
}

EXPORT_SYMBOL_GPL(fib_rules_unregister);

static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops,
			  struct flowi *fl, int flags)
{
	int ret = 0;

	if (rule->ifindex && (rule->ifindex != fl->iif))
		goto out;

	if ((rule->mark ^ fl->mark) & rule->mark_mask)
		goto out;

	ret = ops->match(rule, fl, flags);
out:
	return (rule->flags & FIB_RULE_INVERT) ? !ret : ret;
}

int fib_rules_lookup(struct fib_rules_ops *ops, struct flowi *fl,
		     int flags, struct fib_lookup_arg *arg)
{
	struct fib_rule *rule;
	int err;

	rcu_read_lock();

	list_for_each_entry_rcu(rule, ops->rules_list, list) {
		if (!fib_rule_match(rule, ops, fl, flags))
			continue;

		err = ops->action(rule, fl, flags, arg);
		if (err != -EAGAIN) {
			fib_rule_get(rule);
			arg->rule = rule;
			goto out;
		}
	}

	err = -ENETUNREACH;
out:
	rcu_read_unlock();

	return err;
}

EXPORT_SYMBOL_GPL(fib_rules_lookup);

static int validate_rulemsg(struct fib_rule_hdr *frh, struct nlattr **tb,
			    struct fib_rules_ops *ops)
{
	int err = -EINVAL;

	if (frh->src_len)
		if (tb[FRA_SRC] == NULL ||
		    frh->src_len > (ops->addr_size * 8) ||
		    nla_len(tb[FRA_SRC]) != ops->addr_size)
			goto errout;

	if (frh->dst_len)
		if (tb[FRA_DST] == NULL ||
		    frh->dst_len > (ops->addr_size * 8) ||
		    nla_len(tb[FRA_DST]) != ops->addr_size)
			goto errout;

	err = 0;
errout:
	return err;
}

int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
{
	struct fib_rule_hdr *frh = nlmsg_data(nlh);
	struct fib_rules_ops *ops = NULL;
	struct fib_rule *rule, *r, *last = NULL;
	struct nlattr *tb[FRA_MAX+1];
	int err = -EINVAL;

	if (nlh->nlmsg_len < nlmsg_msg_size(sizeof(*frh)))
		goto errout;

	ops = lookup_rules_ops(frh->family);
	if (ops == NULL) {
		err = EAFNOSUPPORT;
		goto errout;
	}

	err = nlmsg_parse(nlh, sizeof(*frh), tb, FRA_MAX, ops->policy);
	if (err < 0)
		goto errout;

	err = validate_rulemsg(frh, tb, ops);
	if (err < 0)
		goto errout;

	rule = kzalloc(ops->rule_size, GFP_KERNEL);
	if (rule == NULL) {
		err = -ENOMEM;
		goto errout;
	}

	if (tb[FRA_PRIORITY])
		rule->pref = nla_get_u32(tb[FRA_PRIORITY]);

	if (tb[FRA_IFNAME]) {
		struct net_device *dev;

		rule->ifindex = -1;
		nla_strlcpy(rule->ifname, tb[FRA_IFNAME], IFNAMSIZ);
		dev = __dev_get_by_name(rule->ifname);
		if (dev)
			rule->ifindex = dev->ifindex;
	}

	if (tb[FRA_FWMARK]) {
		rule->mark = nla_get_u32(tb[FRA_FWMARK]);
		if (rule->mark)
			/* compatibility: if the mark value is non-zero all bits
			 * are compared unless a mask is explicitly specified.
			 */
			rule->mark_mask = 0xFFFFFFFF;
	}

	if (tb[FRA_FWMASK])
		rule->mark_mask = nla_get_u32(tb[FRA_FWMASK]);

	rule->action = frh->action;
	rule->flags = frh->flags;
	rule->table = frh_get_table(frh, tb);

	if (!rule->pref && ops->default_pref)
		rule->pref = ops->default_pref();

	err = ops->configure(rule, skb, nlh, frh, tb);
	if (err < 0)
		goto errout_free;

	list_for_each_entry(r, ops->rules_list, list) {
		if (r->pref > rule->pref)
			break;
		last = r;
	}

	fib_rule_get(rule);

	if (last)
		list_add_rcu(&rule->list, &last->list);
	else
		list_add_rcu(&rule->list, ops->rules_list);

	notify_rule_change(RTM_NEWRULE, rule, ops, nlh, NETLINK_CB(skb).pid);
	rules_ops_put(ops);
	return 0;

errout_free:
	kfree(rule);
errout:
	rules_ops_put(ops);
	return err;
}

int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
{
	struct fib_rule_hdr *frh = nlmsg_data(nlh);
	struct fib_rules_ops *ops = NULL;
	struct fib_rule *rule;
	struct nlattr *tb[FRA_MAX+1];
	int err = -EINVAL;

	if (nlh->nlmsg_len < nlmsg_msg_size(sizeof(*frh)))
		goto errout;

	ops = lookup_rules_ops(frh->family);
	if (ops == NULL) {
		err = EAFNOSUPPORT;
		goto errout;
	}

	err = nlmsg_parse(nlh, sizeof(*frh), tb, FRA_MAX, ops->policy);
	if (err < 0)
		goto errout;

	err = validate_rulemsg(frh, tb, ops);
	if (err < 0)
		goto errout;

	list_for_each_entry(rule, ops->rules_list, list) {
		if (frh->action && (frh->action != rule->action))
			continue;

		if (frh->table && (frh_get_table(frh, tb) != rule->table))
			continue;

		if (tb[FRA_PRIORITY] &&
		    (rule->pref != nla_get_u32(tb[FRA_PRIORITY])))
			continue;

		if (tb[FRA_IFNAME] &&
		    nla_strcmp(tb[FRA_IFNAME], rule->ifname))
			continue;

		if (tb[FRA_FWMARK] &&
		    (rule->mark != nla_get_u32(tb[FRA_FWMARK])))
			continue;

		if (tb[FRA_FWMASK] &&
		    (rule->mark_mask != nla_get_u32(tb[FRA_FWMASK])))
			continue;

		if (!ops->compare(rule, frh, tb))
			continue;

		if (rule->flags & FIB_RULE_PERMANENT) {
			err = -EPERM;
			goto errout;
		}

		list_del_rcu(&rule->list);
		synchronize_rcu();
		notify_rule_change(RTM_DELRULE, rule, ops, nlh,
				   NETLINK_CB(skb).pid);
		fib_rule_put(rule);
		rules_ops_put(ops);
		return 0;
	}

	err = -ENOENT;
errout:
	rules_ops_put(ops);
	return err;
}

static inline size_t fib_rule_nlmsg_size(struct fib_rules_ops *ops,
					 struct fib_rule *rule)
{
	size_t payload = NLMSG_ALIGN(sizeof(struct fib_rule_hdr))
			 + nla_total_size(IFNAMSIZ) /* FRA_IFNAME */
			 + nla_total_size(4) /* FRA_PRIORITY */
			 + nla_total_size(4) /* FRA_TABLE */
			 + nla_total_size(4) /* FRA_FWMARK */
			 + nla_total_size(4); /* FRA_FWMASK */

	if (ops->nlmsg_payload)
		payload += ops->nlmsg_payload(rule);

	return payload;
}

static int fib_nl_fill_rule(struct sk_buff *skb, struct fib_rule *rule,
			    u32 pid, u32 seq, int type, int flags,
			    struct fib_rules_ops *ops)
{
	struct nlmsghdr *nlh;
	struct fib_rule_hdr *frh;

	nlh = nlmsg_put(skb, pid, seq, type, sizeof(*frh), flags);
	if (nlh == NULL)
		return -EMSGSIZE;

	frh = nlmsg_data(nlh);
	frh->table = rule->table;
	NLA_PUT_U32(skb, FRA_TABLE, rule->table);
	frh->res1 = 0;
	frh->res2 = 0;
	frh->action = rule->action;
	frh->flags = rule->flags;

	if (rule->ifname[0])
		NLA_PUT_STRING(skb, FRA_IFNAME, rule->ifname);

	if (rule->pref)
		NLA_PUT_U32(skb, FRA_PRIORITY, rule->pref);

	if (rule->mark)
		NLA_PUT_U32(skb, FRA_FWMARK, rule->mark);

	if (rule->mark_mask || rule->mark)
		NLA_PUT_U32(skb, FRA_FWMASK, rule->mark_mask);

	if (ops->fill(rule, skb, nlh, frh) < 0)
		goto nla_put_failure;

	return nlmsg_end(skb, nlh);

nla_put_failure:
	nlmsg_cancel(skb, nlh);
	return -EMSGSIZE;
}

int fib_rules_dump(struct sk_buff *skb, struct netlink_callback *cb, int family)
{
	int idx = 0;
	struct fib_rule *rule;
	struct fib_rules_ops *ops;

	ops = lookup_rules_ops(family);
	if (ops == NULL)
		return -EAFNOSUPPORT;

	rcu_read_lock();
	list_for_each_entry_rcu(rule, ops->rules_list, list) {
		if (idx < cb->args[0])
			goto skip;

		if (fib_nl_fill_rule(skb, rule, NETLINK_CB(cb->skb).pid,
				     cb->nlh->nlmsg_seq, RTM_NEWRULE,
				     NLM_F_MULTI, ops) < 0)
			break;
skip:
		idx++;
	}
	rcu_read_unlock();
	cb->args[0] = idx;
	rules_ops_put(ops);

	return skb->len;
}

EXPORT_SYMBOL_GPL(fib_rules_dump);

static void notify_rule_change(int event, struct fib_rule *rule,
			       struct fib_rules_ops *ops, struct nlmsghdr *nlh,
			       u32 pid)
{
	struct sk_buff *skb;
	int err = -ENOBUFS;

	skb = nlmsg_new(fib_rule_nlmsg_size(ops, rule), GFP_KERNEL);
	if (skb == NULL)
		goto errout;

	err = fib_nl_fill_rule(skb, rule, pid, nlh->nlmsg_seq, event, 0, ops);
	if (err < 0) {
		/* -EMSGSIZE implies BUG in fib_rule_nlmsg_size() */
		WARN_ON(err == -EMSGSIZE);
		kfree_skb(skb);
		goto errout;
	}
	err = rtnl_notify(skb, pid, ops->nlgroup, nlh, GFP_KERNEL);
errout:
	if (err < 0)
		rtnl_set_sk_err(ops->nlgroup, err);
}

static void attach_rules(struct list_head *rules, struct net_device *dev)
{
	struct fib_rule *rule;

	list_for_each_entry(rule, rules, list) {
		if (rule->ifindex == -1 &&
		    strcmp(dev->name, rule->ifname) == 0)
			rule->ifindex = dev->ifindex;
	}
}

static void detach_rules(struct list_head *rules, struct net_device *dev)
{
	struct fib_rule *rule;

	list_for_each_entry(rule, rules, list)
		if (rule->ifindex == dev->ifindex)
			rule->ifindex = -1;
}


static int fib_rules_event(struct notifier_block *this, unsigned long event,
			    void *ptr)
{
	struct net_device *dev = ptr;
	struct fib_rules_ops *ops;

	ASSERT_RTNL();
	rcu_read_lock();

	switch (event) {
	case NETDEV_REGISTER:
		list_for_each_entry(ops, &rules_ops, list)
			attach_rules(ops->rules_list, dev);
		break;

	case NETDEV_UNREGISTER:
		list_for_each_entry(ops, &rules_ops, list)
			detach_rules(ops->rules_list, dev);
		break;
	}

	rcu_read_unlock();

	return NOTIFY_DONE;
}

static struct notifier_block fib_rules_notifier = {
	.notifier_call = fib_rules_event,
};

static int __init fib_rules_init(void)
{
	return register_netdevice_notifier(&fib_rules_notifier);
}

subsys_initcall(fib_rules_init);
Commit	Line	Data
14c0b97d TG	1	/*
	2	* net/core/fib_rules.c Generic Routing Rules
	3	*
	4	* This program is free software; you can redistribute it and/or
	5	* modify it under the terms of the GNU General Public License as
	6	* published by the Free Software Foundation, version 2.
	7	*
	8	* Authors: Thomas Graf <tgraf@suug.ch>
	9	*/
	10
14c0b97d TG	11	#include <linux/types.h>
	12	#include <linux/kernel.h>
	13	#include <linux/list.h>
	14	#include <net/fib_rules.h>
	15
	16	static LIST_HEAD(rules_ops);
	17	static DEFINE_SPINLOCK(rules_mod_lock);
	18
	19	static void notify_rule_change(int event, struct fib_rule *rule,
c17084d2 TG	20	struct fib_rules_ops ops, struct nlmsghdr nlh,
c17084d2 TG	21	u32 pid);
14c0b97d TG	22
	23	static struct fib_rules_ops *lookup_rules_ops(int family)
	24	{
	25	struct fib_rules_ops *ops;
	26
	27	rcu_read_lock();
	28	list_for_each_entry_rcu(ops, &rules_ops, list) {
	29	if (ops->family == family) {
	30	if (!try_module_get(ops->owner))
	31	ops = NULL;
	32	rcu_read_unlock();
	33	return ops;
	34	}
	35	}
	36	rcu_read_unlock();
	37
	38	return NULL;
	39	}
	40
	41	static void rules_ops_put(struct fib_rules_ops *ops)
	42	{
	43	if (ops)
	44	module_put(ops->owner);
	45	}
	46
	47	int fib_rules_register(struct fib_rules_ops *ops)
	48	{
	49	int err = -EEXIST;
	50	struct fib_rules_ops *o;
	51
	52	if (ops->rule_size < sizeof(struct fib_rule))
	53	return -EINVAL;
	54
	55	if (ops->match == NULL \|\| ops->configure == NULL \|\|
	56	ops->compare == NULL \|\| ops->fill == NULL \|\|
	57	ops->action == NULL)
	58	return -EINVAL;
	59
	60	spin_lock(&rules_mod_lock);
	61	list_for_each_entry(o, &rules_ops, list)
	62	if (ops->family == o->family)
	63	goto errout;
	64
	65	list_add_tail_rcu(&ops->list, &rules_ops);
	66	err = 0;
	67	errout:
	68	spin_unlock(&rules_mod_lock);
	69
	70	return err;
	71	}
	72
	73	EXPORT_SYMBOL_GPL(fib_rules_register);
	74
	75	static void cleanup_ops(struct fib_rules_ops *ops)
	76	{
	77	struct fib_rule rule, tmp;
	78
	79	list_for_each_entry_safe(rule, tmp, ops->rules_list, list) {
	80	list_del_rcu(&rule->list);
	81	fib_rule_put(rule);
	82	}
	83	}
	84
	85	int fib_rules_unregister(struct fib_rules_ops *ops)
86	{
87	int err = 0;
88	struct fib_rules_ops *o;
89
90	spin_lock(&rules_mod_lock);
91	list_for_each_entry(o, &rules_ops, list) {
92	if (o == ops) {
93	list_del_rcu(&o->list);
94	cleanup_ops(ops);
95	goto out;
96	}
97	}
98
99	err = -ENOENT;
100	out:
101	spin_unlock(&rules_mod_lock);
102
103	synchronize_rcu();
104
105	return err;
106	}
107
108	EXPORT_SYMBOL_GPL(fib_rules_unregister);
109
3dfbcc41 TG	110	static int fib_rule_match(struct fib_rule rule, struct fib_rules_ops ops,
	111	struct flowi *fl, int flags)
	112	{
	113	int ret = 0;
	114
	115	if (rule->ifindex && (rule->ifindex != fl->iif))
	116	goto out;
	117
	118	if ((rule->mark ^ fl->mark) & rule->mark_mask)
	119	goto out;
	120
	121	ret = ops->match(rule, fl, flags);
	122	out:
	123	return (rule->flags & FIB_RULE_INVERT) ? !ret : ret;
	124	}
	125
14c0b97d TG	126	int fib_rules_lookup(struct fib_rules_ops ops, struct flowi fl,
	127	int flags, struct fib_lookup_arg *arg)
	128	{
	129	struct fib_rule *rule;
	130	int err;
	131
	132	rcu_read_lock();
	133
	134	list_for_each_entry_rcu(rule, ops->rules_list, list) {
3dfbcc41	135	if (!fib_rule_match(rule, ops, fl, flags))
14c0b97d TG	136	continue;
	137
	138	err = ops->action(rule, fl, flags, arg);
	139	if (err != -EAGAIN) {
	140	fib_rule_get(rule);
	141	arg->rule = rule;
	142	goto out;
	143	}
	144	}
	145
	146	err = -ENETUNREACH;
	147	out:
	148	rcu_read_unlock();
	149
	150	return err;
	151	}
	152
	153	EXPORT_SYMBOL_GPL(fib_rules_lookup);
	154
e1701c68 TG	155	static int validate_rulemsg(struct fib_rule_hdr frh, struct nlattr *tb,
	156	struct fib_rules_ops *ops)
	157	{
	158	int err = -EINVAL;
	159
	160	if (frh->src_len)
	161	if (tb[FRA_SRC] == NULL \|\|
	162	frh->src_len > (ops->addr_size * 8) \|\|
	163	nla_len(tb[FRA_SRC]) != ops->addr_size)
	164	goto errout;
	165
	166	if (frh->dst_len)
	167	if (tb[FRA_DST] == NULL \|\|
	168	frh->dst_len > (ops->addr_size * 8) \|\|
	169	nla_len(tb[FRA_DST]) != ops->addr_size)
	170	goto errout;
	171
	172	err = 0;
	173	errout:
	174	return err;
	175	}
	176
14c0b97d TG	177	int fib_nl_newrule(struct sk_buff skb, struct nlmsghdr nlh, void *arg)
	178	{
	179	struct fib_rule_hdr *frh = nlmsg_data(nlh);
	180	struct fib_rules_ops *ops = NULL;
	181	struct fib_rule rule, r, *last = NULL;
	182	struct nlattr *tb[FRA_MAX+1];
	183	int err = -EINVAL;
	184
	185	if (nlh->nlmsg_len < nlmsg_msg_size(sizeof(*frh)))
	186	goto errout;
	187
	188	ops = lookup_rules_ops(frh->family);
	189	if (ops == NULL) {
	190	err = EAFNOSUPPORT;
	191	goto errout;
	192	}
	193
	194	err = nlmsg_parse(nlh, sizeof(*frh), tb, FRA_MAX, ops->policy);
	195	if (err < 0)
	196	goto errout;
	197
e1701c68 TG	198	err = validate_rulemsg(frh, tb, ops);
	199	if (err < 0)
	200	goto errout;
	201
14c0b97d TG	202	rule = kzalloc(ops->rule_size, GFP_KERNEL);
	203	if (rule == NULL) {
	204	err = -ENOMEM;
	205	goto errout;
	206	}
	207
	208	if (tb[FRA_PRIORITY])
	209	rule->pref = nla_get_u32(tb[FRA_PRIORITY]);
	210
	211	if (tb[FRA_IFNAME]) {
	212	struct net_device *dev;
	213
	214	rule->ifindex = -1;
5176f91e	215	nla_strlcpy(rule->ifname, tb[FRA_IFNAME], IFNAMSIZ);
14c0b97d TG	216	dev = __dev_get_by_name(rule->ifname);
	217	if (dev)
	218	rule->ifindex = dev->ifindex;
	219	}
	220
b8964ed9 TG	221	if (tb[FRA_FWMARK]) {
	222	rule->mark = nla_get_u32(tb[FRA_FWMARK]);
	223	if (rule->mark)
	224	/* compatibility: if the mark value is non-zero all bits
	225	* are compared unless a mask is explicitly specified.
	226	*/
	227	rule->mark_mask = 0xFFFFFFFF;
	228	}
	229
	230	if (tb[FRA_FWMASK])
	231	rule->mark_mask = nla_get_u32(tb[FRA_FWMASK]);
	232
14c0b97d TG	233	rule->action = frh->action;
14c0b97d TG	234	rule->flags = frh->flags;
9e762a4a	235	rule->table = frh_get_table(frh, tb);
14c0b97d TG	236
	237	if (!rule->pref && ops->default_pref)
	238	rule->pref = ops->default_pref();
	239
	240	err = ops->configure(rule, skb, nlh, frh, tb);
	241	if (err < 0)
	242	goto errout_free;
	243
	244	list_for_each_entry(r, ops->rules_list, list) {
	245	if (r->pref > rule->pref)
	246	break;
	247	last = r;
	248	}
	249
	250	fib_rule_get(rule);
	251
	252	if (last)
	253	list_add_rcu(&rule->list, &last->list);
	254	else
	255	list_add_rcu(&rule->list, ops->rules_list);
	256
c17084d2	257	notify_rule_change(RTM_NEWRULE, rule, ops, nlh, NETLINK_CB(skb).pid);
14c0b97d TG	258	rules_ops_put(ops);
	259	return 0;
	260
	261	errout_free:
	262	kfree(rule);
	263	errout:
	264	rules_ops_put(ops);
	265	return err;
	266	}
	267
	268	int fib_nl_delrule(struct sk_buff skb, struct nlmsghdr nlh, void *arg)
	269	{
	270	struct fib_rule_hdr *frh = nlmsg_data(nlh);
	271	struct fib_rules_ops *ops = NULL;
	272	struct fib_rule *rule;
	273	struct nlattr *tb[FRA_MAX+1];
	274	int err = -EINVAL;
	275
	276	if (nlh->nlmsg_len < nlmsg_msg_size(sizeof(*frh)))
	277	goto errout;
	278
	279	ops = lookup_rules_ops(frh->family);
	280	if (ops == NULL) {
	281	err = EAFNOSUPPORT;
	282	goto errout;
	283	}
	284
	285	err = nlmsg_parse(nlh, sizeof(*frh), tb, FRA_MAX, ops->policy);
	286	if (err < 0)
	287	goto errout;
	288
e1701c68 TG	289	err = validate_rulemsg(frh, tb, ops);
	290	if (err < 0)
	291	goto errout;
	292
14c0b97d TG	293	list_for_each_entry(rule, ops->rules_list, list) {
	294	if (frh->action && (frh->action != rule->action))
	295	continue;
	296
9e762a4a	297	if (frh->table && (frh_get_table(frh, tb) != rule->table))
14c0b97d TG	298	continue;
	299
	300	if (tb[FRA_PRIORITY] &&
	301	(rule->pref != nla_get_u32(tb[FRA_PRIORITY])))
	302	continue;
	303
	304	if (tb[FRA_IFNAME] &&
	305	nla_strcmp(tb[FRA_IFNAME], rule->ifname))
	306	continue;
	307
b8964ed9 TG	308	if (tb[FRA_FWMARK] &&
	309	(rule->mark != nla_get_u32(tb[FRA_FWMARK])))
	310	continue;
	311
	312	if (tb[FRA_FWMASK] &&
	313	(rule->mark_mask != nla_get_u32(tb[FRA_FWMASK])))
	314	continue;
	315
14c0b97d TG	316	if (!ops->compare(rule, frh, tb))
	317	continue;
	318
	319	if (rule->flags & FIB_RULE_PERMANENT) {
	320	err = -EPERM;
	321	goto errout;
	322	}
	323
	324	list_del_rcu(&rule->list);
	325	synchronize_rcu();
c17084d2 TG	326	notify_rule_change(RTM_DELRULE, rule, ops, nlh,
c17084d2 TG	327	NETLINK_CB(skb).pid);
14c0b97d TG	328	fib_rule_put(rule);
	329	rules_ops_put(ops);
	330	return 0;
	331	}
	332
	333	err = -ENOENT;
	334	errout:
	335	rules_ops_put(ops);
	336	return err;
	337	}
	338
339bf98f TG	339	static inline size_t fib_rule_nlmsg_size(struct fib_rules_ops *ops,
	340	struct fib_rule *rule)
	341	{
	342	size_t payload = NLMSG_ALIGN(sizeof(struct fib_rule_hdr))
	343	+ nla_total_size(IFNAMSIZ) /* FRA_IFNAME */
	344	+ nla_total_size(4) /* FRA_PRIORITY */
	345	+ nla_total_size(4) /* FRA_TABLE */
	346	+ nla_total_size(4) /* FRA_FWMARK */
	347	+ nla_total_size(4); /* FRA_FWMASK */
	348
	349	if (ops->nlmsg_payload)
	350	payload += ops->nlmsg_payload(rule);
	351
	352	return payload;
	353	}
	354
14c0b97d TG	355	static int fib_nl_fill_rule(struct sk_buff skb, struct fib_rule rule,
	356	u32 pid, u32 seq, int type, int flags,
	357	struct fib_rules_ops *ops)
	358	{
	359	struct nlmsghdr *nlh;
	360	struct fib_rule_hdr *frh;
	361
	362	nlh = nlmsg_put(skb, pid, seq, type, sizeof(*frh), flags);
	363	if (nlh == NULL)
26932566	364	return -EMSGSIZE;
14c0b97d TG	365
	366	frh = nlmsg_data(nlh);
	367	frh->table = rule->table;
9e762a4a	368	NLA_PUT_U32(skb, FRA_TABLE, rule->table);
14c0b97d TG	369	frh->res1 = 0;
	370	frh->res2 = 0;
	371	frh->action = rule->action;
	372	frh->flags = rule->flags;
	373
	374	if (rule->ifname[0])
	375	NLA_PUT_STRING(skb, FRA_IFNAME, rule->ifname);
	376
	377	if (rule->pref)
	378	NLA_PUT_U32(skb, FRA_PRIORITY, rule->pref);
	379
b8964ed9 TG	380	if (rule->mark)
	381	NLA_PUT_U32(skb, FRA_FWMARK, rule->mark);
	382
	383	if (rule->mark_mask \|\| rule->mark)
	384	NLA_PUT_U32(skb, FRA_FWMASK, rule->mark_mask);
	385
14c0b97d TG	386	if (ops->fill(rule, skb, nlh, frh) < 0)
	387	goto nla_put_failure;
	388
	389	return nlmsg_end(skb, nlh);
	390
	391	nla_put_failure:
26932566 PM	392	nlmsg_cancel(skb, nlh);
26932566 PM	393	return -EMSGSIZE;
14c0b97d TG	394	}
	395
	396	int fib_rules_dump(struct sk_buff skb, struct netlink_callback cb, int family)
	397	{
	398	int idx = 0;
	399	struct fib_rule *rule;
	400	struct fib_rules_ops *ops;
	401
	402	ops = lookup_rules_ops(family);
	403	if (ops == NULL)
	404	return -EAFNOSUPPORT;
	405
	406	rcu_read_lock();
ec25615b	407	list_for_each_entry_rcu(rule, ops->rules_list, list) {
14c0b97d TG	408	if (idx < cb->args[0])
	409	goto skip;
	410
	411	if (fib_nl_fill_rule(skb, rule, NETLINK_CB(cb->skb).pid,
	412	cb->nlh->nlmsg_seq, RTM_NEWRULE,
	413	NLM_F_MULTI, ops) < 0)
	414	break;
	415	skip:
	416	idx++;
	417	}
	418	rcu_read_unlock();
	419	cb->args[0] = idx;
	420	rules_ops_put(ops);
	421
	422	return skb->len;
	423	}
	424
	425	EXPORT_SYMBOL_GPL(fib_rules_dump);
	426
	427	static void notify_rule_change(int event, struct fib_rule *rule,
c17084d2 TG	428	struct fib_rules_ops ops, struct nlmsghdr nlh,
c17084d2 TG	429	u32 pid)
14c0b97d	430	{
c17084d2 TG	431	struct sk_buff *skb;
c17084d2 TG	432	int err = -ENOBUFS;
14c0b97d	433
339bf98f	434	skb = nlmsg_new(fib_rule_nlmsg_size(ops, rule), GFP_KERNEL);
14c0b97d	435	if (skb == NULL)
c17084d2 TG	436	goto errout;
	437
	438	err = fib_nl_fill_rule(skb, rule, pid, nlh->nlmsg_seq, event, 0, ops);
26932566 PM	439	if (err < 0) {
	440	/* -EMSGSIZE implies BUG in fib_rule_nlmsg_size() */
	441	WARN_ON(err == -EMSGSIZE);
	442	kfree_skb(skb);
	443	goto errout;
	444	}
c17084d2 TG	445	err = rtnl_notify(skb, pid, ops->nlgroup, nlh, GFP_KERNEL);
	446	errout:
	447	if (err < 0)
	448	rtnl_set_sk_err(ops->nlgroup, err);
14c0b97d TG	449	}
	450
	451	static void attach_rules(struct list_head rules, struct net_device dev)
	452	{
	453	struct fib_rule *rule;
	454
	455	list_for_each_entry(rule, rules, list) {
	456	if (rule->ifindex == -1 &&
	457	strcmp(dev->name, rule->ifname) == 0)
	458	rule->ifindex = dev->ifindex;
	459	}
	460	}
	461
	462	static void detach_rules(struct list_head rules, struct net_device dev)
	463	{
	464	struct fib_rule *rule;
	465
	466	list_for_each_entry(rule, rules, list)
	467	if (rule->ifindex == dev->ifindex)
	468	rule->ifindex = -1;
	469	}
	470
	471
	472	static int fib_rules_event(struct notifier_block *this, unsigned long event,
	473	void *ptr)
	474	{
	475	struct net_device *dev = ptr;
	476	struct fib_rules_ops *ops;
	477
	478	ASSERT_RTNL();
	479	rcu_read_lock();
	480
	481	switch (event) {
	482	case NETDEV_REGISTER:
	483	list_for_each_entry(ops, &rules_ops, list)
	484	attach_rules(ops->rules_list, dev);
	485	break;
	486
	487	case NETDEV_UNREGISTER:
	488	list_for_each_entry(ops, &rules_ops, list)
	489	detach_rules(ops->rules_list, dev);
	490	break;
	491	}
	492
	493	rcu_read_unlock();
	494
	495	return NOTIFY_DONE;
	496	}
	497
	498	static struct notifier_block fib_rules_notifier = {
	499	.notifier_call = fib_rules_event,
	500	};
	501
	502	static int __init fib_rules_init(void)
	503	{
	504	return register_netdevice_notifier(&fib_rules_notifier);
	505	}
	506
	507	subsys_initcall(fib_rules_init);