[linux-2.6-block.git] / net / netfilter / nfnetlink_acct.c

/*
 * (C) 2011 Pablo Neira Ayuso <pablo@netfilter.org>
 * (C) 2011 Intra2net AG <http://www.intra2net.com>
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation (or any later at your option).
 */
#include <linux/init.h>
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/skbuff.h>
#include <linux/atomic.h>
#include <linux/netlink.h>
#include <linux/rculist.h>
#include <linux/slab.h>
#include <linux/types.h>
#include <linux/errno.h>
#include <net/netlink.h>
#include <net/sock.h>

#include <linux/netfilter.h>
#include <linux/netfilter/nfnetlink.h>
#include <linux/netfilter/nfnetlink_acct.h>

MODULE_LICENSE("GPL");
MODULE_AUTHOR("Pablo Neira Ayuso <pablo@netfilter.org>");
MODULE_DESCRIPTION("nfacct: Extended Netfilter accounting infrastructure");

struct nf_acct {
	atomic64_t		pkts;
	atomic64_t		bytes;
	unsigned long		flags;
	struct list_head	head;
	atomic_t		refcnt;
	char			name[NFACCT_NAME_MAX];
	struct rcu_head		rcu_head;
	char			data[0];
};

struct nfacct_filter {
	u32 value;
	u32 mask;
};

#define NFACCT_F_QUOTA (NFACCT_F_QUOTA_PKTS | NFACCT_F_QUOTA_BYTES)
#define NFACCT_OVERQUOTA_BIT	2	/* NFACCT_F_OVERQUOTA */

static int nfnl_acct_new(struct net *net, struct sock *nfnl,
			 struct sk_buff *skb, const struct nlmsghdr *nlh,
			 const struct nlattr * const tb[])
{
	struct nf_acct *nfacct, *matching = NULL;
	char *acct_name;
	unsigned int size = 0;
	u32 flags = 0;

	if (!tb[NFACCT_NAME])
		return -EINVAL;

	acct_name = nla_data(tb[NFACCT_NAME]);
	if (strlen(acct_name) == 0)
		return -EINVAL;

	list_for_each_entry(nfacct, &net->nfnl_acct_list, head) {
		if (strncmp(nfacct->name, acct_name, NFACCT_NAME_MAX) != 0)
			continue;

                if (nlh->nlmsg_flags & NLM_F_EXCL)
			return -EEXIST;

		matching = nfacct;
		break;
        }

	if (matching) {
		if (nlh->nlmsg_flags & NLM_F_REPLACE) {
			/* reset counters if you request a replacement. */
			atomic64_set(&matching->pkts, 0);
			atomic64_set(&matching->bytes, 0);
			smp_mb__before_atomic();
			/* reset overquota flag if quota is enabled. */
			if ((matching->flags & NFACCT_F_QUOTA))
				clear_bit(NFACCT_OVERQUOTA_BIT,
					  &matching->flags);
			return 0;
		}
		return -EBUSY;
	}

	if (tb[NFACCT_FLAGS]) {
		flags = ntohl(nla_get_be32(tb[NFACCT_FLAGS]));
		if (flags & ~NFACCT_F_QUOTA)
			return -EOPNOTSUPP;
		if ((flags & NFACCT_F_QUOTA) == NFACCT_F_QUOTA)
			return -EINVAL;
		if (flags & NFACCT_F_OVERQUOTA)
			return -EINVAL;
		if ((flags & NFACCT_F_QUOTA) && !tb[NFACCT_QUOTA])
			return -EINVAL;

		size += sizeof(u64);
	}

	nfacct = kzalloc(sizeof(struct nf_acct) + size, GFP_KERNEL);
	if (nfacct == NULL)
		return -ENOMEM;

	if (flags & NFACCT_F_QUOTA) {
		u64 *quota = (u64 *)nfacct->data;

		*quota = be64_to_cpu(nla_get_be64(tb[NFACCT_QUOTA]));
		nfacct->flags = flags;
	}

	strncpy(nfacct->name, nla_data(tb[NFACCT_NAME]), NFACCT_NAME_MAX);

	if (tb[NFACCT_BYTES]) {
		atomic64_set(&nfacct->bytes,
			     be64_to_cpu(nla_get_be64(tb[NFACCT_BYTES])));
	}
	if (tb[NFACCT_PKTS]) {
		atomic64_set(&nfacct->pkts,
			     be64_to_cpu(nla_get_be64(tb[NFACCT_PKTS])));
	}
	atomic_set(&nfacct->refcnt, 1);
	list_add_tail_rcu(&nfacct->head, &net->nfnl_acct_list);
	return 0;
}

static int
nfnl_acct_fill_info(struct sk_buff *skb, u32 portid, u32 seq, u32 type,
		   int event, struct nf_acct *acct)
{
	struct nlmsghdr *nlh;
	struct nfgenmsg *nfmsg;
	unsigned int flags = portid ? NLM_F_MULTI : 0;
	u64 pkts, bytes;
	u32 old_flags;

	event |= NFNL_SUBSYS_ACCT << 8;
	nlh = nlmsg_put(skb, portid, seq, event, sizeof(*nfmsg), flags);
	if (nlh == NULL)
		goto nlmsg_failure;

	nfmsg = nlmsg_data(nlh);
	nfmsg->nfgen_family = AF_UNSPEC;
	nfmsg->version = NFNETLINK_V0;
	nfmsg->res_id = 0;

	if (nla_put_string(skb, NFACCT_NAME, acct->name))
		goto nla_put_failure;

	old_flags = acct->flags;
	if (type == NFNL_MSG_ACCT_GET_CTRZERO) {
		pkts = atomic64_xchg(&acct->pkts, 0);
		bytes = atomic64_xchg(&acct->bytes, 0);
		smp_mb__before_atomic();
		if (acct->flags & NFACCT_F_QUOTA)
			clear_bit(NFACCT_OVERQUOTA_BIT, &acct->flags);
	} else {
		pkts = atomic64_read(&acct->pkts);
		bytes = atomic64_read(&acct->bytes);
	}
	if (nla_put_be64(skb, NFACCT_PKTS, cpu_to_be64(pkts)) ||
	    nla_put_be64(skb, NFACCT_BYTES, cpu_to_be64(bytes)) ||
	    nla_put_be32(skb, NFACCT_USE, htonl(atomic_read(&acct->refcnt))))
		goto nla_put_failure;
	if (acct->flags & NFACCT_F_QUOTA) {
		u64 *quota = (u64 *)acct->data;

		if (nla_put_be32(skb, NFACCT_FLAGS, htonl(old_flags)) ||
		    nla_put_be64(skb, NFACCT_QUOTA, cpu_to_be64(*quota)))
			goto nla_put_failure;
	}
	nlmsg_end(skb, nlh);
	return skb->len;

nlmsg_failure:
nla_put_failure:
	nlmsg_cancel(skb, nlh);
	return -1;
}

static int
nfnl_acct_dump(struct sk_buff *skb, struct netlink_callback *cb)
{
	struct net *net = sock_net(skb->sk);
	struct nf_acct *cur, *last;
	const struct nfacct_filter *filter = cb->data;

	if (cb->args[2])
		return 0;

	last = (struct nf_acct *)cb->args[1];
	if (cb->args[1])
		cb->args[1] = 0;

	rcu_read_lock();
	list_for_each_entry_rcu(cur, &net->nfnl_acct_list, head) {
		if (last) {
			if (cur != last)
				continue;

			last = NULL;
		}

		if (filter && (cur->flags & filter->mask) != filter->value)
			continue;

		if (nfnl_acct_fill_info(skb, NETLINK_CB(cb->skb).portid,
				       cb->nlh->nlmsg_seq,
				       NFNL_MSG_TYPE(cb->nlh->nlmsg_type),
				       NFNL_MSG_ACCT_NEW, cur) < 0) {
			cb->args[1] = (unsigned long)cur;
			break;
		}
	}
	if (!cb->args[1])
		cb->args[2] = 1;
	rcu_read_unlock();
	return skb->len;
}

static int nfnl_acct_done(struct netlink_callback *cb)
{
	kfree(cb->data);
	return 0;
}

static const struct nla_policy filter_policy[NFACCT_FILTER_MAX + 1] = {
	[NFACCT_FILTER_MASK]	= { .type = NLA_U32 },
	[NFACCT_FILTER_VALUE]	= { .type = NLA_U32 },
};

static struct nfacct_filter *
nfacct_filter_alloc(const struct nlattr * const attr)
{
	struct nfacct_filter *filter;
	struct nlattr *tb[NFACCT_FILTER_MAX + 1];
	int err;

	err = nla_parse_nested(tb, NFACCT_FILTER_MAX, attr, filter_policy);
	if (err < 0)
		return ERR_PTR(err);

	if (!tb[NFACCT_FILTER_MASK] || !tb[NFACCT_FILTER_VALUE])
		return ERR_PTR(-EINVAL);

	filter = kzalloc(sizeof(struct nfacct_filter), GFP_KERNEL);
	if (!filter)
		return ERR_PTR(-ENOMEM);

	filter->mask = ntohl(nla_get_be32(tb[NFACCT_FILTER_MASK]));
	filter->value = ntohl(nla_get_be32(tb[NFACCT_FILTER_VALUE]));

	return filter;
}

static int nfnl_acct_get(struct net *net, struct sock *nfnl,
			 struct sk_buff *skb, const struct nlmsghdr *nlh,
			 const struct nlattr * const tb[])
{
	int ret = -ENOENT;
	struct nf_acct *cur;
	char *acct_name;

	if (nlh->nlmsg_flags & NLM_F_DUMP) {
		struct netlink_dump_control c = {
			.dump = nfnl_acct_dump,
			.done = nfnl_acct_done,
		};

		if (tb[NFACCT_FILTER]) {
			struct nfacct_filter *filter;

			filter = nfacct_filter_alloc(tb[NFACCT_FILTER]);
			if (IS_ERR(filter))
				return PTR_ERR(filter);

			c.data = filter;
		}
		return netlink_dump_start(nfnl, skb, nlh, &c);
	}

	if (!tb[NFACCT_NAME])
		return -EINVAL;
	acct_name = nla_data(tb[NFACCT_NAME]);

	list_for_each_entry(cur, &net->nfnl_acct_list, head) {
		struct sk_buff *skb2;

		if (strncmp(cur->name, acct_name, NFACCT_NAME_MAX)!= 0)
			continue;

		skb2 = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
		if (skb2 == NULL) {
			ret = -ENOMEM;
			break;
		}

		ret = nfnl_acct_fill_info(skb2, NETLINK_CB(skb).portid,
					 nlh->nlmsg_seq,
					 NFNL_MSG_TYPE(nlh->nlmsg_type),
					 NFNL_MSG_ACCT_NEW, cur);
		if (ret <= 0) {
			kfree_skb(skb2);
			break;
		}
		ret = netlink_unicast(nfnl, skb2, NETLINK_CB(skb).portid,
					MSG_DONTWAIT);
		if (ret > 0)
			ret = 0;

		/* this avoids a loop in nfnetlink. */
		return ret == -EAGAIN ? -ENOBUFS : ret;
	}
	return ret;
}

/* try to delete object, fail if it is still in use. */
static int nfnl_acct_try_del(struct nf_acct *cur)
{
	int ret = 0;

	/* we want to avoid races with nfnl_acct_find_get. */
	if (atomic_dec_and_test(&cur->refcnt)) {
		/* We are protected by nfnl mutex. */
		list_del_rcu(&cur->head);
		kfree_rcu(cur, rcu_head);
	} else {
		/* still in use, restore reference counter. */
		atomic_inc(&cur->refcnt);
		ret = -EBUSY;
	}
	return ret;
}

static int nfnl_acct_del(struct net *net, struct sock *nfnl,
			 struct sk_buff *skb, const struct nlmsghdr *nlh,
			 const struct nlattr * const tb[])
{
	char *acct_name;
	struct nf_acct *cur;
	int ret = -ENOENT;

	if (!tb[NFACCT_NAME]) {
		list_for_each_entry(cur, &net->nfnl_acct_list, head)
			nfnl_acct_try_del(cur);

		return 0;
	}
	acct_name = nla_data(tb[NFACCT_NAME]);

	list_for_each_entry(cur, &net->nfnl_acct_list, head) {
		if (strncmp(cur->name, acct_name, NFACCT_NAME_MAX) != 0)
			continue;

		ret = nfnl_acct_try_del(cur);
		if (ret < 0)
			return ret;

		break;
	}
	return ret;
}

static const struct nla_policy nfnl_acct_policy[NFACCT_MAX+1] = {
	[NFACCT_NAME] = { .type = NLA_NUL_STRING, .len = NFACCT_NAME_MAX-1 },
	[NFACCT_BYTES] = { .type = NLA_U64 },
	[NFACCT_PKTS] = { .type = NLA_U64 },
	[NFACCT_FLAGS] = { .type = NLA_U32 },
	[NFACCT_QUOTA] = { .type = NLA_U64 },
	[NFACCT_FILTER] = {.type = NLA_NESTED },
};

static const struct nfnl_callback nfnl_acct_cb[NFNL_MSG_ACCT_MAX] = {
	[NFNL_MSG_ACCT_NEW]		= { .call = nfnl_acct_new,
					    .attr_count = NFACCT_MAX,
					    .policy = nfnl_acct_policy },
	[NFNL_MSG_ACCT_GET] 		= { .call = nfnl_acct_get,
					    .attr_count = NFACCT_MAX,
					    .policy = nfnl_acct_policy },
	[NFNL_MSG_ACCT_GET_CTRZERO] 	= { .call = nfnl_acct_get,
					    .attr_count = NFACCT_MAX,
					    .policy = nfnl_acct_policy },
	[NFNL_MSG_ACCT_DEL]		= { .call = nfnl_acct_del,
					    .attr_count = NFACCT_MAX,
					    .policy = nfnl_acct_policy },
};

static const struct nfnetlink_subsystem nfnl_acct_subsys = {
	.name				= "acct",
	.subsys_id			= NFNL_SUBSYS_ACCT,
	.cb_count			= NFNL_MSG_ACCT_MAX,
	.cb				= nfnl_acct_cb,
};

MODULE_ALIAS_NFNL_SUBSYS(NFNL_SUBSYS_ACCT);

struct nf_acct *nfnl_acct_find_get(struct net *net, const char *acct_name)
{
	struct nf_acct *cur, *acct = NULL;

	rcu_read_lock();
	list_for_each_entry_rcu(cur, &net->nfnl_acct_list, head) {
		if (strncmp(cur->name, acct_name, NFACCT_NAME_MAX)!= 0)
			continue;

		if (!try_module_get(THIS_MODULE))
			goto err;

		if (!atomic_inc_not_zero(&cur->refcnt)) {
			module_put(THIS_MODULE);
			goto err;
		}

		acct = cur;
		break;
	}
err:
	rcu_read_unlock();
	return acct;
}
EXPORT_SYMBOL_GPL(nfnl_acct_find_get);

void nfnl_acct_put(struct nf_acct *acct)
{
	if (atomic_dec_and_test(&acct->refcnt))
		kfree_rcu(acct, rcu_head);

	module_put(THIS_MODULE);
}
EXPORT_SYMBOL_GPL(nfnl_acct_put);

void nfnl_acct_update(const struct sk_buff *skb, struct nf_acct *nfacct)
{
	atomic64_inc(&nfacct->pkts);
	atomic64_add(skb->len, &nfacct->bytes);
}
EXPORT_SYMBOL_GPL(nfnl_acct_update);

static void nfnl_overquota_report(struct nf_acct *nfacct)
{
	int ret;
	struct sk_buff *skb;

	skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC);
	if (skb == NULL)
		return;

	ret = nfnl_acct_fill_info(skb, 0, 0, NFNL_MSG_ACCT_OVERQUOTA, 0,
				  nfacct);
	if (ret <= 0) {
		kfree_skb(skb);
		return;
	}
	netlink_broadcast(init_net.nfnl, skb, 0, NFNLGRP_ACCT_QUOTA,
			  GFP_ATOMIC);
}

int nfnl_acct_overquota(const struct sk_buff *skb, struct nf_acct *nfacct)
{
	u64 now;
	u64 *quota;
	int ret = NFACCT_UNDERQUOTA;

	/* no place here if we don't have a quota */
	if (!(nfacct->flags & NFACCT_F_QUOTA))
		return NFACCT_NO_QUOTA;

	quota = (u64 *)nfacct->data;
	now = (nfacct->flags & NFACCT_F_QUOTA_PKTS) ?
	       atomic64_read(&nfacct->pkts) : atomic64_read(&nfacct->bytes);

	ret = now > *quota;

	if (now >= *quota &&
	    !test_and_set_bit(NFACCT_OVERQUOTA_BIT, &nfacct->flags)) {
		nfnl_overquota_report(nfacct);
	}

	return ret;
}
EXPORT_SYMBOL_GPL(nfnl_acct_overquota);

static int __net_init nfnl_acct_net_init(struct net *net)
{
	INIT_LIST_HEAD(&net->nfnl_acct_list);

	return 0;
}

static void __net_exit nfnl_acct_net_exit(struct net *net)
{
	struct nf_acct *cur, *tmp;

	list_for_each_entry_safe(cur, tmp, &net->nfnl_acct_list, head) {
		list_del_rcu(&cur->head);

		if (atomic_dec_and_test(&cur->refcnt))
			kfree_rcu(cur, rcu_head);
	}
}

static struct pernet_operations nfnl_acct_ops = {
        .init   = nfnl_acct_net_init,
        .exit   = nfnl_acct_net_exit,
};

static int __init nfnl_acct_init(void)
{
	int ret;

	ret = register_pernet_subsys(&nfnl_acct_ops);
	if (ret < 0) {
		pr_err("nfnl_acct_init: failed to register pernet ops\n");
		goto err_out;
	}

	pr_info("nfnl_acct: registering with nfnetlink.\n");
	ret = nfnetlink_subsys_register(&nfnl_acct_subsys);
	if (ret < 0) {
		pr_err("nfnl_acct_init: cannot register with nfnetlink.\n");
		goto cleanup_pernet;
	}
	return 0;

cleanup_pernet:
	unregister_pernet_subsys(&nfnl_acct_ops);
err_out:
	return ret;
}

static void __exit nfnl_acct_exit(void)
{
	pr_info("nfnl_acct: unregistering from nfnetlink.\n");
	nfnetlink_subsys_unregister(&nfnl_acct_subsys);
	unregister_pernet_subsys(&nfnl_acct_ops);
}

module_init(nfnl_acct_init);
module_exit(nfnl_acct_exit);
Commit	Line	Data
94139027 PNA	1	/*
	2	* (C) 2011 Pablo Neira Ayuso <pablo@netfilter.org>
	3	* (C) 2011 Intra2net AG <http://www.intra2net.com>
	4	*
	5	* This program is free software; you can redistribute it and/or modify
	6	* it under the terms of the GNU General Public License version 2 as
	7	* published by the Free Software Foundation (or any later at your option).
	8	*/
	9	#include <linux/init.h>
	10	#include <linux/module.h>
	11	#include <linux/kernel.h>
	12	#include <linux/skbuff.h>
6523cf9a	13	#include <linux/atomic.h>
94139027 PNA	14	#include <linux/netlink.h>
	15	#include <linux/rculist.h>
	16	#include <linux/slab.h>
	17	#include <linux/types.h>
	18	#include <linux/errno.h>
	19	#include <net/netlink.h>
	20	#include <net/sock.h>
94139027 PNA	21
	22	#include <linux/netfilter.h>
	23	#include <linux/netfilter/nfnetlink.h>
	24	#include <linux/netfilter/nfnetlink_acct.h>
	25
	26	MODULE_LICENSE("GPL");
	27	MODULE_AUTHOR("Pablo Neira Ayuso <pablo@netfilter.org>");
	28	MODULE_DESCRIPTION("nfacct: Extended Netfilter accounting infrastructure");
	29
94139027 PNA	30	struct nf_acct {
	31	atomic64_t pkts;
	32	atomic64_t bytes;
683399ed	33	unsigned long flags;
94139027 PNA	34	struct list_head head;
	35	atomic_t refcnt;
	36	char name[NFACCT_NAME_MAX];
	37	struct rcu_head rcu_head;
683399ed	38	char data[0];
94139027 PNA	39	};
94139027 PNA	40
f111f780 AP	41	struct nfacct_filter {
	42	u32 value;
	43	u32 mask;
	44	};
	45
683399ed	46	#define NFACCT_F_QUOTA (NFACCT_F_QUOTA_PKTS \| NFACCT_F_QUOTA_BYTES)
b6d04688	47	#define NFACCT_OVERQUOTA_BIT 2 /* NFACCT_F_OVERQUOTA */
683399ed	48
7b8002a1 PNA	49	static int nfnl_acct_new(struct net net, struct sock nfnl,
	50	struct sk_buff skb, const struct nlmsghdr nlh,
	51	const struct nlattr * const tb[])
94139027 PNA	52	{
	53	struct nf_acct nfacct, matching = NULL;
	54	char *acct_name;
683399ed MP	55	unsigned int size = 0;
683399ed MP	56	u32 flags = 0;
94139027 PNA	57
	58	if (!tb[NFACCT_NAME])
	59	return -EINVAL;
	60
	61	acct_name = nla_data(tb[NFACCT_NAME]);
deadcfc3 PNA	62	if (strlen(acct_name) == 0)
deadcfc3 PNA	63	return -EINVAL;
94139027	64
3499abb2	65	list_for_each_entry(nfacct, &net->nfnl_acct_list, head) {
94139027 PNA	66	if (strncmp(nfacct->name, acct_name, NFACCT_NAME_MAX) != 0)
	67	continue;
	68
	69	if (nlh->nlmsg_flags & NLM_F_EXCL)
	70	return -EEXIST;
	71
	72	matching = nfacct;
	73	break;
	74	}
	75
	76	if (matching) {
	77	if (nlh->nlmsg_flags & NLM_F_REPLACE) {
	78	/* reset counters if you request a replacement. */
	79	atomic64_set(&matching->pkts, 0);
	80	atomic64_set(&matching->bytes, 0);
f9da455b	81	smp_mb__before_atomic();
683399ed MP	82	/* reset overquota flag if quota is enabled. */
683399ed MP	83	if ((matching->flags & NFACCT_F_QUOTA))
b6d04688 AP	84	clear_bit(NFACCT_OVERQUOTA_BIT,
b6d04688 AP	85	&matching->flags);
94139027 PNA	86	return 0;
	87	}
	88	return -EBUSY;
	89	}
	90
683399ed MP	91	if (tb[NFACCT_FLAGS]) {
	92	flags = ntohl(nla_get_be32(tb[NFACCT_FLAGS]));
	93	if (flags & ~NFACCT_F_QUOTA)
	94	return -EOPNOTSUPP;
	95	if ((flags & NFACCT_F_QUOTA) == NFACCT_F_QUOTA)
	96	return -EINVAL;
	97	if (flags & NFACCT_F_OVERQUOTA)
	98	return -EINVAL;
eda3fc50 PT	99	if ((flags & NFACCT_F_QUOTA) && !tb[NFACCT_QUOTA])
eda3fc50 PT	100	return -EINVAL;
683399ed MP	101
	102	size += sizeof(u64);
	103	}
	104
	105	nfacct = kzalloc(sizeof(struct nf_acct) + size, GFP_KERNEL);
94139027 PNA	106	if (nfacct == NULL)
	107	return -ENOMEM;
	108
683399ed MP	109	if (flags & NFACCT_F_QUOTA) {
	110	u64 quota = (u64 )nfacct->data;
	111
	112	*quota = be64_to_cpu(nla_get_be64(tb[NFACCT_QUOTA]));
	113	nfacct->flags = flags;
	114	}
	115
94139027 PNA	116	strncpy(nfacct->name, nla_data(tb[NFACCT_NAME]), NFACCT_NAME_MAX);
	117
	118	if (tb[NFACCT_BYTES]) {
	119	atomic64_set(&nfacct->bytes,
fe31d1a8	120	be64_to_cpu(nla_get_be64(tb[NFACCT_BYTES])));
94139027 PNA	121	}
	122	if (tb[NFACCT_PKTS]) {
	123	atomic64_set(&nfacct->pkts,
fe31d1a8	124	be64_to_cpu(nla_get_be64(tb[NFACCT_PKTS])));
94139027 PNA	125	}
94139027 PNA	126	atomic_set(&nfacct->refcnt, 1);
3499abb2	127	list_add_tail_rcu(&nfacct->head, &net->nfnl_acct_list);
94139027 PNA	128	return 0;
	129	}
	130
	131	static int
15e47304	132	nfnl_acct_fill_info(struct sk_buff *skb, u32 portid, u32 seq, u32 type,
94139027 PNA	133	int event, struct nf_acct *acct)
	134	{
	135	struct nlmsghdr *nlh;
	136	struct nfgenmsg *nfmsg;
15e47304	137	unsigned int flags = portid ? NLM_F_MULTI : 0;
94139027	138	u64 pkts, bytes;
d24675cb	139	u32 old_flags;
94139027 PNA	140
94139027 PNA	141	event \|= NFNL_SUBSYS_ACCT << 8;
15e47304	142	nlh = nlmsg_put(skb, portid, seq, event, sizeof(*nfmsg), flags);
94139027 PNA	143	if (nlh == NULL)
	144	goto nlmsg_failure;
	145
	146	nfmsg = nlmsg_data(nlh);
	147	nfmsg->nfgen_family = AF_UNSPEC;
	148	nfmsg->version = NFNETLINK_V0;
	149	nfmsg->res_id = 0;
	150
7c801189 DM	151	if (nla_put_string(skb, NFACCT_NAME, acct->name))
7c801189 DM	152	goto nla_put_failure;
94139027	153
d24675cb	154	old_flags = acct->flags;
94139027 PNA	155	if (type == NFNL_MSG_ACCT_GET_CTRZERO) {
	156	pkts = atomic64_xchg(&acct->pkts, 0);
	157	bytes = atomic64_xchg(&acct->bytes, 0);
f9da455b	158	smp_mb__before_atomic();
683399ed	159	if (acct->flags & NFACCT_F_QUOTA)
b6d04688	160	clear_bit(NFACCT_OVERQUOTA_BIT, &acct->flags);
94139027 PNA	161	} else {
	162	pkts = atomic64_read(&acct->pkts);
	163	bytes = atomic64_read(&acct->bytes);
	164	}
7c801189 DM	165	if (nla_put_be64(skb, NFACCT_PKTS, cpu_to_be64(pkts)) \|\|
	166	nla_put_be64(skb, NFACCT_BYTES, cpu_to_be64(bytes)) \|\|
	167	nla_put_be32(skb, NFACCT_USE, htonl(atomic_read(&acct->refcnt))))
	168	goto nla_put_failure;
683399ed MP	169	if (acct->flags & NFACCT_F_QUOTA) {
683399ed MP	170	u64 quota = (u64 )acct->data;
94139027	171
d24675cb	172	if (nla_put_be32(skb, NFACCT_FLAGS, htonl(old_flags)) \|\|
683399ed MP	173	nla_put_be64(skb, NFACCT_QUOTA, cpu_to_be64(*quota)))
	174	goto nla_put_failure;
	175	}
94139027 PNA	176	nlmsg_end(skb, nlh);
	177	return skb->len;
	178
	179	nlmsg_failure:
	180	nla_put_failure:
	181	nlmsg_cancel(skb, nlh);
	182	return -1;
	183	}
	184
	185	static int
	186	nfnl_acct_dump(struct sk_buff skb, struct netlink_callback cb)
	187	{
3499abb2	188	struct net *net = sock_net(skb->sk);
94139027	189	struct nf_acct cur, last;
f111f780	190	const struct nfacct_filter *filter = cb->data;
94139027 PNA	191
	192	if (cb->args[2])
	193	return 0;
	194
	195	last = (struct nf_acct *)cb->args[1];
	196	if (cb->args[1])
	197	cb->args[1] = 0;
	198
	199	rcu_read_lock();
3499abb2	200	list_for_each_entry_rcu(cur, &net->nfnl_acct_list, head) {
991a6b73 PNA	201	if (last) {
	202	if (cur != last)
	203	continue;
94139027	204
991a6b73 PNA	205	last = NULL;
991a6b73 PNA	206	}
f111f780 AP	207
	208	if (filter && (cur->flags & filter->mask) != filter->value)
	209	continue;
	210
15e47304	211	if (nfnl_acct_fill_info(skb, NETLINK_CB(cb->skb).portid,
94139027 PNA	212	cb->nlh->nlmsg_seq,
	213	NFNL_MSG_TYPE(cb->nlh->nlmsg_type),
	214	NFNL_MSG_ACCT_NEW, cur) < 0) {
	215	cb->args[1] = (unsigned long)cur;
	216	break;
	217	}
	218	}
	219	if (!cb->args[1])
	220	cb->args[2] = 1;
	221	rcu_read_unlock();
	222	return skb->len;
	223	}
	224
f111f780 AP	225	static int nfnl_acct_done(struct netlink_callback *cb)
	226	{
	227	kfree(cb->data);
	228	return 0;
	229	}
	230
	231	static const struct nla_policy filter_policy[NFACCT_FILTER_MAX + 1] = {
	232	[NFACCT_FILTER_MASK] = { .type = NLA_U32 },
	233	[NFACCT_FILTER_VALUE] = { .type = NLA_U32 },
	234	};
	235
	236	static struct nfacct_filter *
	237	nfacct_filter_alloc(const struct nlattr * const attr)
	238	{
	239	struct nfacct_filter *filter;
	240	struct nlattr *tb[NFACCT_FILTER_MAX + 1];
	241	int err;
	242
	243	err = nla_parse_nested(tb, NFACCT_FILTER_MAX, attr, filter_policy);
	244	if (err < 0)
	245	return ERR_PTR(err);
	246
017b1b6d PT	247	if (!tb[NFACCT_FILTER_MASK] \|\| !tb[NFACCT_FILTER_VALUE])
	248	return ERR_PTR(-EINVAL);
	249
f111f780 AP	250	filter = kzalloc(sizeof(struct nfacct_filter), GFP_KERNEL);
	251	if (!filter)
	252	return ERR_PTR(-ENOMEM);
	253
	254	filter->mask = ntohl(nla_get_be32(tb[NFACCT_FILTER_MASK]));
	255	filter->value = ntohl(nla_get_be32(tb[NFACCT_FILTER_VALUE]));
	256
	257	return filter;
	258	}
	259
7b8002a1 PNA	260	static int nfnl_acct_get(struct net net, struct sock nfnl,
	261	struct sk_buff skb, const struct nlmsghdr nlh,
	262	const struct nlattr * const tb[])
94139027	263	{
3ab0b245	264	int ret = -ENOENT;
94139027 PNA	265	struct nf_acct *cur;
	266	char *acct_name;
	267
	268	if (nlh->nlmsg_flags & NLM_F_DUMP) {
80d326fa PNA	269	struct netlink_dump_control c = {
80d326fa PNA	270	.dump = nfnl_acct_dump,
f111f780	271	.done = nfnl_acct_done,
80d326fa	272	};
f111f780 AP	273
	274	if (tb[NFACCT_FILTER]) {
	275	struct nfacct_filter *filter;
	276
	277	filter = nfacct_filter_alloc(tb[NFACCT_FILTER]);
	278	if (IS_ERR(filter))
	279	return PTR_ERR(filter);
	280
	281	c.data = filter;
	282	}
80d326fa	283	return netlink_dump_start(nfnl, skb, nlh, &c);
94139027 PNA	284	}
	285
	286	if (!tb[NFACCT_NAME])
	287	return -EINVAL;
	288	acct_name = nla_data(tb[NFACCT_NAME]);
	289
3499abb2	290	list_for_each_entry(cur, &net->nfnl_acct_list, head) {
94139027 PNA	291	struct sk_buff *skb2;
	292
	293	if (strncmp(cur->name, acct_name, NFACCT_NAME_MAX)!= 0)
	294	continue;
	295
	296	skb2 = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
3ab0b245 PNA	297	if (skb2 == NULL) {
3ab0b245 PNA	298	ret = -ENOMEM;
94139027	299	break;
3ab0b245	300	}
94139027	301
15e47304	302	ret = nfnl_acct_fill_info(skb2, NETLINK_CB(skb).portid,
94139027 PNA	303	nlh->nlmsg_seq,
	304	NFNL_MSG_TYPE(nlh->nlmsg_type),
	305	NFNL_MSG_ACCT_NEW, cur);
3ab0b245	306	if (ret <= 0) {
94139027	307	kfree_skb(skb2);
3ab0b245 PNA	308	break;
3ab0b245 PNA	309	}
15e47304	310	ret = netlink_unicast(nfnl, skb2, NETLINK_CB(skb).portid,
3ab0b245 PNA	311	MSG_DONTWAIT);
	312	if (ret > 0)
	313	ret = 0;
94139027	314
3ab0b245 PNA	315	/* this avoids a loop in nfnetlink. */
3ab0b245 PNA	316	return ret == -EAGAIN ? -ENOBUFS : ret;
94139027 PNA	317	}
	318	return ret;
	319	}
	320
	321	/* try to delete object, fail if it is still in use. */
	322	static int nfnl_acct_try_del(struct nf_acct *cur)
	323	{
	324	int ret = 0;
	325
	326	/* we want to avoid races with nfnl_acct_find_get. */
	327	if (atomic_dec_and_test(&cur->refcnt)) {
	328	/* We are protected by nfnl mutex. */
	329	list_del_rcu(&cur->head);
	330	kfree_rcu(cur, rcu_head);
	331	} else {
	332	/* still in use, restore reference counter. */
	333	atomic_inc(&cur->refcnt);
	334	ret = -EBUSY;
	335	}
	336	return ret;
	337	}
	338
7b8002a1 PNA	339	static int nfnl_acct_del(struct net net, struct sock nfnl,
	340	struct sk_buff skb, const struct nlmsghdr nlh,
	341	const struct nlattr * const tb[])
94139027 PNA	342	{
	343	char *acct_name;
	344	struct nf_acct *cur;
	345	int ret = -ENOENT;
	346
	347	if (!tb[NFACCT_NAME]) {
3499abb2	348	list_for_each_entry(cur, &net->nfnl_acct_list, head)
94139027 PNA	349	nfnl_acct_try_del(cur);
	350
	351	return 0;
	352	}
	353	acct_name = nla_data(tb[NFACCT_NAME]);
	354
3499abb2	355	list_for_each_entry(cur, &net->nfnl_acct_list, head) {
94139027 PNA	356	if (strncmp(cur->name, acct_name, NFACCT_NAME_MAX) != 0)
	357	continue;
	358
	359	ret = nfnl_acct_try_del(cur);
	360	if (ret < 0)
	361	return ret;
	362
	363	break;
	364	}
	365	return ret;
	366	}
	367
	368	static const struct nla_policy nfnl_acct_policy[NFACCT_MAX+1] = {
	369	[NFACCT_NAME] = { .type = NLA_NUL_STRING, .len = NFACCT_NAME_MAX-1 },
	370	[NFACCT_BYTES] = { .type = NLA_U64 },
	371	[NFACCT_PKTS] = { .type = NLA_U64 },
683399ed MP	372	[NFACCT_FLAGS] = { .type = NLA_U32 },
683399ed MP	373	[NFACCT_QUOTA] = { .type = NLA_U64 },
f111f780	374	[NFACCT_FILTER] = {.type = NLA_NESTED },
94139027 PNA	375	};
	376
	377	static const struct nfnl_callback nfnl_acct_cb[NFNL_MSG_ACCT_MAX] = {
	378	[NFNL_MSG_ACCT_NEW] = { .call = nfnl_acct_new,
	379	.attr_count = NFACCT_MAX,
	380	.policy = nfnl_acct_policy },
	381	[NFNL_MSG_ACCT_GET] = { .call = nfnl_acct_get,
	382	.attr_count = NFACCT_MAX,
	383	.policy = nfnl_acct_policy },
	384	[NFNL_MSG_ACCT_GET_CTRZERO] = { .call = nfnl_acct_get,
	385	.attr_count = NFACCT_MAX,
	386	.policy = nfnl_acct_policy },
	387	[NFNL_MSG_ACCT_DEL] = { .call = nfnl_acct_del,
	388	.attr_count = NFACCT_MAX,
	389	.policy = nfnl_acct_policy },
	390	};
	391
	392	static const struct nfnetlink_subsystem nfnl_acct_subsys = {
	393	.name = "acct",
	394	.subsys_id = NFNL_SUBSYS_ACCT,
	395	.cb_count = NFNL_MSG_ACCT_MAX,
	396	.cb = nfnl_acct_cb,
	397	};
	398
	399	MODULE_ALIAS_NFNL_SUBSYS(NFNL_SUBSYS_ACCT);
	400
3499abb2	401	struct nf_acct nfnl_acct_find_get(struct net net, const char *acct_name)
94139027 PNA	402	{
	403	struct nf_acct cur, acct = NULL;
	404
	405	rcu_read_lock();
3499abb2	406	list_for_each_entry_rcu(cur, &net->nfnl_acct_list, head) {
94139027 PNA	407	if (strncmp(cur->name, acct_name, NFACCT_NAME_MAX)!= 0)
	408	continue;
	409
	410	if (!try_module_get(THIS_MODULE))
	411	goto err;
	412
	413	if (!atomic_inc_not_zero(&cur->refcnt)) {
	414	module_put(THIS_MODULE);
	415	goto err;
	416	}
	417
	418	acct = cur;
	419	break;
	420	}
	421	err:
	422	rcu_read_unlock();
	423	return acct;
	424	}
	425	EXPORT_SYMBOL_GPL(nfnl_acct_find_get);
	426
	427	void nfnl_acct_put(struct nf_acct *acct)
	428	{
3499abb2 AS	429	if (atomic_dec_and_test(&acct->refcnt))
	430	kfree_rcu(acct, rcu_head);
	431
94139027 PNA	432	module_put(THIS_MODULE);
	433	}
	434	EXPORT_SYMBOL_GPL(nfnl_acct_put);
	435
	436	void nfnl_acct_update(const struct sk_buff skb, struct nf_acct nfacct)
	437	{
	438	atomic64_inc(&nfacct->pkts);
	439	atomic64_add(skb->len, &nfacct->bytes);
	440	}
	441	EXPORT_SYMBOL_GPL(nfnl_acct_update);
	442
683399ed MP	443	static void nfnl_overquota_report(struct nf_acct *nfacct)
	444	{
	445	int ret;
	446	struct sk_buff *skb;
	447
	448	skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC);
	449	if (skb == NULL)
	450	return;
	451
	452	ret = nfnl_acct_fill_info(skb, 0, 0, NFNL_MSG_ACCT_OVERQUOTA, 0,
	453	nfacct);
	454	if (ret <= 0) {
	455	kfree_skb(skb);
	456	return;
	457	}
	458	netlink_broadcast(init_net.nfnl, skb, 0, NFNLGRP_ACCT_QUOTA,
	459	GFP_ATOMIC);
	460	}
	461
	462	int nfnl_acct_overquota(const struct sk_buff skb, struct nf_acct nfacct)
	463	{
	464	u64 now;
	465	u64 *quota;
	466	int ret = NFACCT_UNDERQUOTA;
	467
	468	/* no place here if we don't have a quota */
	469	if (!(nfacct->flags & NFACCT_F_QUOTA))
	470	return NFACCT_NO_QUOTA;
	471
	472	quota = (u64 *)nfacct->data;
	473	now = (nfacct->flags & NFACCT_F_QUOTA_PKTS) ?
	474	atomic64_read(&nfacct->pkts) : atomic64_read(&nfacct->bytes);
	475
	476	ret = now > *quota;
	477
	478	if (now >= *quota &&
b6d04688	479	!test_and_set_bit(NFACCT_OVERQUOTA_BIT, &nfacct->flags)) {
683399ed MP	480	nfnl_overquota_report(nfacct);
	481	}
	482
	483	return ret;
	484	}
	485	EXPORT_SYMBOL_GPL(nfnl_acct_overquota);
	486
3499abb2 AS	487	static int __net_init nfnl_acct_net_init(struct net *net)
	488	{
	489	INIT_LIST_HEAD(&net->nfnl_acct_list);
	490
	491	return 0;
	492	}
	493
	494	static void __net_exit nfnl_acct_net_exit(struct net *net)
	495	{
	496	struct nf_acct cur, tmp;
	497
	498	list_for_each_entry_safe(cur, tmp, &net->nfnl_acct_list, head) {
	499	list_del_rcu(&cur->head);
	500
	501	if (atomic_dec_and_test(&cur->refcnt))
	502	kfree_rcu(cur, rcu_head);
	503	}
	504	}
	505
	506	static struct pernet_operations nfnl_acct_ops = {
	507	.init = nfnl_acct_net_init,
	508	.exit = nfnl_acct_net_exit,
	509	};
	510
94139027 PNA	511	static int __init nfnl_acct_init(void)
	512	{
	513	int ret;
	514
3499abb2 AS	515	ret = register_pernet_subsys(&nfnl_acct_ops);
	516	if (ret < 0) {
	517	pr_err("nfnl_acct_init: failed to register pernet ops\n");
	518	goto err_out;
	519	}
	520
94139027 PNA	521	pr_info("nfnl_acct: registering with nfnetlink.\n");
	522	ret = nfnetlink_subsys_register(&nfnl_acct_subsys);
	523	if (ret < 0) {
	524	pr_err("nfnl_acct_init: cannot register with nfnetlink.\n");
3499abb2	525	goto cleanup_pernet;
94139027 PNA	526	}
94139027 PNA	527	return 0;
3499abb2 AS	528
	529	cleanup_pernet:
	530	unregister_pernet_subsys(&nfnl_acct_ops);
94139027 PNA	531	err_out:
	532	return ret;
	533	}
	534
	535	static void __exit nfnl_acct_exit(void)
	536	{
94139027 PNA	537	pr_info("nfnl_acct: unregistering from nfnetlink.\n");
94139027 PNA	538	nfnetlink_subsys_unregister(&nfnl_acct_subsys);
3499abb2	539	unregister_pernet_subsys(&nfnl_acct_ops);
94139027 PNA	540	}
	541
	542	module_init(nfnl_acct_init);
	543	module_exit(nfnl_acct_exit);