[deliverable/linux.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;

		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);

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