[mirror_ubuntu-artful-kernel.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/refcount.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;
	refcount_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 netlink_ext_ack *extack)
{
	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])));
	}
	refcount_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_msg_type(NFNL_SUBSYS_ACCT, event);
	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),
			 NFACCT_PAD) ||
	    nla_put_be64(skb, NFACCT_BYTES, cpu_to_be64(bytes),
			 NFACCT_PAD) ||
	    nla_put_be32(skb, NFACCT_USE, htonl(refcount_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),
				 NFACCT_PAD))
			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,
			       NULL);
	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[],
			 struct netlink_ext_ack *extack)
{
	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_put. So only when the current
	 * refcnt is 1, we decrease it to 0.
	 */
	if (refcount_dec_if_one(&cur->refcnt)) {
		/* We are protected by nfnl mutex. */
		list_del_rcu(&cur->head);
		kfree_rcu(cur, rcu_head);
	} else {
		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[],
			 struct netlink_ext_ack *extack)
{
	struct nf_acct *cur, *tmp;
	int ret = -ENOENT;
	char *acct_name;

	if (!tb[NFACCT_NAME]) {
		list_for_each_entry_safe(cur, tmp, &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 (!refcount_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 (refcount_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 net *net, 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(net->nfnl, skb, 0, NFNLGRP_ACCT_QUOTA,
			  GFP_ATOMIC);
}

int nfnl_acct_overquota(struct net *net, 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(net, 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 (refcount_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>
b54ab92b	14	#include <linux/refcount.h>
94139027 PNA	15	#include <linux/netlink.h>
	16	#include <linux/rculist.h>
	17	#include <linux/slab.h>
	18	#include <linux/types.h>
	19	#include <linux/errno.h>
	20	#include <net/netlink.h>
	21	#include <net/sock.h>
94139027 PNA	22
	23	#include <linux/netfilter.h>
	24	#include <linux/netfilter/nfnetlink.h>
	25	#include <linux/netfilter/nfnetlink_acct.h>
	26
	27	MODULE_LICENSE("GPL");
	28	MODULE_AUTHOR("Pablo Neira Ayuso <pablo@netfilter.org>");
	29	MODULE_DESCRIPTION("nfacct: Extended Netfilter accounting infrastructure");
	30
94139027 PNA	31	struct nf_acct {
	32	atomic64_t pkts;
	33	atomic64_t bytes;
683399ed	34	unsigned long flags;
94139027	35	struct list_head head;
b54ab92b	36	refcount_t refcnt;
94139027 PNA	37	char name[NFACCT_NAME_MAX];
94139027 PNA	38	struct rcu_head rcu_head;
683399ed	39	char data[0];
94139027 PNA	40	};
94139027 PNA	41
f111f780 AP	42	struct nfacct_filter {
	43	u32 value;
	44	u32 mask;
	45	};
	46
683399ed	47	#define NFACCT_F_QUOTA (NFACCT_F_QUOTA_PKTS \| NFACCT_F_QUOTA_BYTES)
b6d04688	48	#define NFACCT_OVERQUOTA_BIT 2 /* NFACCT_F_OVERQUOTA */
683399ed	49
7b8002a1 PNA	50	static int nfnl_acct_new(struct net net, struct sock nfnl,
7b8002a1 PNA	51	struct sk_buff skb, const struct nlmsghdr nlh,
04ba724b PNA	52	const struct nlattr * const tb[],
04ba724b PNA	53	struct netlink_ext_ack *extack)
94139027 PNA	54	{
	55	struct nf_acct nfacct, matching = NULL;
	56	char *acct_name;
683399ed MP	57	unsigned int size = 0;
683399ed MP	58	u32 flags = 0;
94139027 PNA	59
	60	if (!tb[NFACCT_NAME])
	61	return -EINVAL;
	62
	63	acct_name = nla_data(tb[NFACCT_NAME]);
deadcfc3 PNA	64	if (strlen(acct_name) == 0)
deadcfc3 PNA	65	return -EINVAL;
94139027	66
3499abb2	67	list_for_each_entry(nfacct, &net->nfnl_acct_list, head) {
94139027 PNA	68	if (strncmp(nfacct->name, acct_name, NFACCT_NAME_MAX) != 0)
	69	continue;
	70
	71	if (nlh->nlmsg_flags & NLM_F_EXCL)
	72	return -EEXIST;
	73
	74	matching = nfacct;
	75	break;
	76	}
	77
	78	if (matching) {
	79	if (nlh->nlmsg_flags & NLM_F_REPLACE) {
	80	/* reset counters if you request a replacement. */
	81	atomic64_set(&matching->pkts, 0);
	82	atomic64_set(&matching->bytes, 0);
f9da455b	83	smp_mb__before_atomic();
683399ed MP	84	/* reset overquota flag if quota is enabled. */
683399ed MP	85	if ((matching->flags & NFACCT_F_QUOTA))
b6d04688 AP	86	clear_bit(NFACCT_OVERQUOTA_BIT,
b6d04688 AP	87	&matching->flags);
94139027 PNA	88	return 0;
	89	}
	90	return -EBUSY;
	91	}
	92
683399ed MP	93	if (tb[NFACCT_FLAGS]) {
	94	flags = ntohl(nla_get_be32(tb[NFACCT_FLAGS]));
	95	if (flags & ~NFACCT_F_QUOTA)
	96	return -EOPNOTSUPP;
	97	if ((flags & NFACCT_F_QUOTA) == NFACCT_F_QUOTA)
	98	return -EINVAL;
	99	if (flags & NFACCT_F_OVERQUOTA)
	100	return -EINVAL;
eda3fc50 PT	101	if ((flags & NFACCT_F_QUOTA) && !tb[NFACCT_QUOTA])
eda3fc50 PT	102	return -EINVAL;
683399ed MP	103
	104	size += sizeof(u64);
	105	}
	106
	107	nfacct = kzalloc(sizeof(struct nf_acct) + size, GFP_KERNEL);
94139027 PNA	108	if (nfacct == NULL)
	109	return -ENOMEM;
	110
683399ed MP	111	if (flags & NFACCT_F_QUOTA) {
	112	u64 quota = (u64 )nfacct->data;
	113
	114	*quota = be64_to_cpu(nla_get_be64(tb[NFACCT_QUOTA]));
	115	nfacct->flags = flags;
	116	}
	117
94139027 PNA	118	strncpy(nfacct->name, nla_data(tb[NFACCT_NAME]), NFACCT_NAME_MAX);
	119
	120	if (tb[NFACCT_BYTES]) {
	121	atomic64_set(&nfacct->bytes,
fe31d1a8	122	be64_to_cpu(nla_get_be64(tb[NFACCT_BYTES])));
94139027 PNA	123	}
	124	if (tb[NFACCT_PKTS]) {
	125	atomic64_set(&nfacct->pkts,
fe31d1a8	126	be64_to_cpu(nla_get_be64(tb[NFACCT_PKTS])));
94139027	127	}
b54ab92b	128	refcount_set(&nfacct->refcnt, 1);
3499abb2	129	list_add_tail_rcu(&nfacct->head, &net->nfnl_acct_list);
94139027 PNA	130	return 0;
	131	}
	132
	133	static int
15e47304	134	nfnl_acct_fill_info(struct sk_buff *skb, u32 portid, u32 seq, u32 type,
94139027 PNA	135	int event, struct nf_acct *acct)
	136	{
	137	struct nlmsghdr *nlh;
	138	struct nfgenmsg *nfmsg;
15e47304	139	unsigned int flags = portid ? NLM_F_MULTI : 0;
94139027	140	u64 pkts, bytes;
d24675cb	141	u32 old_flags;
94139027	142
dedb67c4	143	event = nfnl_msg_type(NFNL_SUBSYS_ACCT, event);
15e47304	144	nlh = nlmsg_put(skb, portid, seq, event, sizeof(*nfmsg), flags);
94139027 PNA	145	if (nlh == NULL)
	146	goto nlmsg_failure;
	147
	148	nfmsg = nlmsg_data(nlh);
	149	nfmsg->nfgen_family = AF_UNSPEC;
	150	nfmsg->version = NFNETLINK_V0;
	151	nfmsg->res_id = 0;
	152
7c801189 DM	153	if (nla_put_string(skb, NFACCT_NAME, acct->name))
7c801189 DM	154	goto nla_put_failure;
94139027	155
d24675cb	156	old_flags = acct->flags;
94139027 PNA	157	if (type == NFNL_MSG_ACCT_GET_CTRZERO) {
	158	pkts = atomic64_xchg(&acct->pkts, 0);
	159	bytes = atomic64_xchg(&acct->bytes, 0);
f9da455b	160	smp_mb__before_atomic();
683399ed	161	if (acct->flags & NFACCT_F_QUOTA)
b6d04688	162	clear_bit(NFACCT_OVERQUOTA_BIT, &acct->flags);
94139027 PNA	163	} else {
	164	pkts = atomic64_read(&acct->pkts);
	165	bytes = atomic64_read(&acct->bytes);
	166	}
b46f6ded ND	167	if (nla_put_be64(skb, NFACCT_PKTS, cpu_to_be64(pkts),
	168	NFACCT_PAD) \|\|
	169	nla_put_be64(skb, NFACCT_BYTES, cpu_to_be64(bytes),
	170	NFACCT_PAD) \|\|
b54ab92b	171	nla_put_be32(skb, NFACCT_USE, htonl(refcount_read(&acct->refcnt))))
7c801189	172	goto nla_put_failure;
683399ed MP	173	if (acct->flags & NFACCT_F_QUOTA) {
683399ed MP	174	u64 quota = (u64 )acct->data;
94139027	175
d24675cb	176	if (nla_put_be32(skb, NFACCT_FLAGS, htonl(old_flags)) \|\|
b46f6ded ND	177	nla_put_be64(skb, NFACCT_QUOTA, cpu_to_be64(*quota),
b46f6ded ND	178	NFACCT_PAD))
683399ed MP	179	goto nla_put_failure;
683399ed MP	180	}
94139027 PNA	181	nlmsg_end(skb, nlh);
	182	return skb->len;
	183
	184	nlmsg_failure:
	185	nla_put_failure:
	186	nlmsg_cancel(skb, nlh);
	187	return -1;
	188	}
	189
	190	static int
	191	nfnl_acct_dump(struct sk_buff skb, struct netlink_callback cb)
	192	{
3499abb2	193	struct net *net = sock_net(skb->sk);
94139027	194	struct nf_acct cur, last;
f111f780	195	const struct nfacct_filter *filter = cb->data;
94139027 PNA	196
	197	if (cb->args[2])
	198	return 0;
	199
	200	last = (struct nf_acct *)cb->args[1];
	201	if (cb->args[1])
	202	cb->args[1] = 0;
	203
	204	rcu_read_lock();
3499abb2	205	list_for_each_entry_rcu(cur, &net->nfnl_acct_list, head) {
991a6b73 PNA	206	if (last) {
	207	if (cur != last)
	208	continue;
94139027	209
991a6b73 PNA	210	last = NULL;
991a6b73 PNA	211	}
f111f780 AP	212
	213	if (filter && (cur->flags & filter->mask) != filter->value)
	214	continue;
	215
15e47304	216	if (nfnl_acct_fill_info(skb, NETLINK_CB(cb->skb).portid,
94139027 PNA	217	cb->nlh->nlmsg_seq,
	218	NFNL_MSG_TYPE(cb->nlh->nlmsg_type),
	219	NFNL_MSG_ACCT_NEW, cur) < 0) {
	220	cb->args[1] = (unsigned long)cur;
	221	break;
	222	}
	223	}
	224	if (!cb->args[1])
	225	cb->args[2] = 1;
	226	rcu_read_unlock();
	227	return skb->len;
	228	}
	229
f111f780 AP	230	static int nfnl_acct_done(struct netlink_callback *cb)
	231	{
	232	kfree(cb->data);
	233	return 0;
	234	}
	235
	236	static const struct nla_policy filter_policy[NFACCT_FILTER_MAX + 1] = {
	237	[NFACCT_FILTER_MASK] = { .type = NLA_U32 },
	238	[NFACCT_FILTER_VALUE] = { .type = NLA_U32 },
	239	};
	240
	241	static struct nfacct_filter *
	242	nfacct_filter_alloc(const struct nlattr * const attr)
	243	{
	244	struct nfacct_filter *filter;
	245	struct nlattr *tb[NFACCT_FILTER_MAX + 1];
	246	int err;
	247
fceb6435 JB	248	err = nla_parse_nested(tb, NFACCT_FILTER_MAX, attr, filter_policy,
fceb6435 JB	249	NULL);
f111f780 AP	250	if (err < 0)
	251	return ERR_PTR(err);
	252
017b1b6d PT	253	if (!tb[NFACCT_FILTER_MASK] \|\| !tb[NFACCT_FILTER_VALUE])
	254	return ERR_PTR(-EINVAL);
	255
f111f780 AP	256	filter = kzalloc(sizeof(struct nfacct_filter), GFP_KERNEL);
	257	if (!filter)
	258	return ERR_PTR(-ENOMEM);
	259
	260	filter->mask = ntohl(nla_get_be32(tb[NFACCT_FILTER_MASK]));
	261	filter->value = ntohl(nla_get_be32(tb[NFACCT_FILTER_VALUE]));
	262
	263	return filter;
	264	}
	265
7b8002a1 PNA	266	static int nfnl_acct_get(struct net net, struct sock nfnl,
7b8002a1 PNA	267	struct sk_buff skb, const struct nlmsghdr nlh,
04ba724b PNA	268	const struct nlattr * const tb[],
04ba724b PNA	269	struct netlink_ext_ack *extack)
94139027	270	{
3ab0b245	271	int ret = -ENOENT;
94139027 PNA	272	struct nf_acct *cur;
	273	char *acct_name;
	274
	275	if (nlh->nlmsg_flags & NLM_F_DUMP) {
80d326fa PNA	276	struct netlink_dump_control c = {
80d326fa PNA	277	.dump = nfnl_acct_dump,
f111f780	278	.done = nfnl_acct_done,
80d326fa	279	};
f111f780 AP	280
	281	if (tb[NFACCT_FILTER]) {
	282	struct nfacct_filter *filter;
	283
	284	filter = nfacct_filter_alloc(tb[NFACCT_FILTER]);
	285	if (IS_ERR(filter))
	286	return PTR_ERR(filter);
	287
	288	c.data = filter;
	289	}
80d326fa	290	return netlink_dump_start(nfnl, skb, nlh, &c);
94139027 PNA	291	}
	292
	293	if (!tb[NFACCT_NAME])
	294	return -EINVAL;
	295	acct_name = nla_data(tb[NFACCT_NAME]);
	296
3499abb2	297	list_for_each_entry(cur, &net->nfnl_acct_list, head) {
94139027 PNA	298	struct sk_buff *skb2;
	299
	300	if (strncmp(cur->name, acct_name, NFACCT_NAME_MAX)!= 0)
	301	continue;
	302
	303	skb2 = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
3ab0b245 PNA	304	if (skb2 == NULL) {
3ab0b245 PNA	305	ret = -ENOMEM;
94139027	306	break;
3ab0b245	307	}
94139027	308
15e47304	309	ret = nfnl_acct_fill_info(skb2, NETLINK_CB(skb).portid,
94139027 PNA	310	nlh->nlmsg_seq,
	311	NFNL_MSG_TYPE(nlh->nlmsg_type),
	312	NFNL_MSG_ACCT_NEW, cur);
3ab0b245	313	if (ret <= 0) {
94139027	314	kfree_skb(skb2);
3ab0b245 PNA	315	break;
3ab0b245 PNA	316	}
15e47304	317	ret = netlink_unicast(nfnl, skb2, NETLINK_CB(skb).portid,
3ab0b245 PNA	318	MSG_DONTWAIT);
	319	if (ret > 0)
	320	ret = 0;
94139027	321
3ab0b245 PNA	322	/* this avoids a loop in nfnetlink. */
3ab0b245 PNA	323	return ret == -EAGAIN ? -ENOBUFS : ret;
94139027 PNA	324	}
	325	return ret;
	326	}
	327
	328	/* try to delete object, fail if it is still in use. */
	329	static int nfnl_acct_try_del(struct nf_acct *cur)
	330	{
	331	int ret = 0;
	332
12be15dd LZ	333	/* We want to avoid races with nfnl_acct_put. So only when the current
	334	* refcnt is 1, we decrease it to 0.
	335	*/
b54ab92b	336	if (refcount_dec_if_one(&cur->refcnt)) {
94139027 PNA	337	/* We are protected by nfnl mutex. */
	338	list_del_rcu(&cur->head);
	339	kfree_rcu(cur, rcu_head);
	340	} else {
94139027 PNA	341	ret = -EBUSY;
	342	}
	343	return ret;
	344	}
	345
7b8002a1 PNA	346	static int nfnl_acct_del(struct net net, struct sock nfnl,
7b8002a1 PNA	347	struct sk_buff skb, const struct nlmsghdr nlh,
04ba724b PNA	348	const struct nlattr * const tb[],
04ba724b PNA	349	struct netlink_ext_ack *extack)
94139027	350	{
93fac10b	351	struct nf_acct cur, tmp;
94139027	352	int ret = -ENOENT;
93fac10b	353	char *acct_name;
94139027 PNA	354
94139027 PNA	355	if (!tb[NFACCT_NAME]) {
93fac10b	356	list_for_each_entry_safe(cur, tmp, &net->nfnl_acct_list, head)
94139027 PNA	357	nfnl_acct_try_del(cur);
	358
	359	return 0;
	360	}
	361	acct_name = nla_data(tb[NFACCT_NAME]);
	362
3499abb2	363	list_for_each_entry(cur, &net->nfnl_acct_list, head) {
94139027 PNA	364	if (strncmp(cur->name, acct_name, NFACCT_NAME_MAX) != 0)
	365	continue;
	366
	367	ret = nfnl_acct_try_del(cur);
	368	if (ret < 0)
	369	return ret;
	370
	371	break;
	372	}
	373	return ret;
	374	}
	375
	376	static const struct nla_policy nfnl_acct_policy[NFACCT_MAX+1] = {
	377	[NFACCT_NAME] = { .type = NLA_NUL_STRING, .len = NFACCT_NAME_MAX-1 },
	378	[NFACCT_BYTES] = { .type = NLA_U64 },
	379	[NFACCT_PKTS] = { .type = NLA_U64 },
683399ed MP	380	[NFACCT_FLAGS] = { .type = NLA_U32 },
683399ed MP	381	[NFACCT_QUOTA] = { .type = NLA_U64 },
f111f780	382	[NFACCT_FILTER] = {.type = NLA_NESTED },
94139027 PNA	383	};
	384
	385	static const struct nfnl_callback nfnl_acct_cb[NFNL_MSG_ACCT_MAX] = {
	386	[NFNL_MSG_ACCT_NEW] = { .call = nfnl_acct_new,
	387	.attr_count = NFACCT_MAX,
	388	.policy = nfnl_acct_policy },
	389	[NFNL_MSG_ACCT_GET] = { .call = nfnl_acct_get,
	390	.attr_count = NFACCT_MAX,
	391	.policy = nfnl_acct_policy },
	392	[NFNL_MSG_ACCT_GET_CTRZERO] = { .call = nfnl_acct_get,
	393	.attr_count = NFACCT_MAX,
	394	.policy = nfnl_acct_policy },
	395	[NFNL_MSG_ACCT_DEL] = { .call = nfnl_acct_del,
	396	.attr_count = NFACCT_MAX,
	397	.policy = nfnl_acct_policy },
	398	};
	399
	400	static const struct nfnetlink_subsystem nfnl_acct_subsys = {
	401	.name = "acct",
	402	.subsys_id = NFNL_SUBSYS_ACCT,
	403	.cb_count = NFNL_MSG_ACCT_MAX,
	404	.cb = nfnl_acct_cb,
	405	};
	406
	407	MODULE_ALIAS_NFNL_SUBSYS(NFNL_SUBSYS_ACCT);
	408
3499abb2	409	struct nf_acct nfnl_acct_find_get(struct net net, const char *acct_name)
94139027 PNA	410	{
	411	struct nf_acct cur, acct = NULL;
	412
	413	rcu_read_lock();
3499abb2	414	list_for_each_entry_rcu(cur, &net->nfnl_acct_list, head) {
94139027 PNA	415	if (strncmp(cur->name, acct_name, NFACCT_NAME_MAX)!= 0)
	416	continue;
	417
	418	if (!try_module_get(THIS_MODULE))
	419	goto err;
	420
b54ab92b	421	if (!refcount_inc_not_zero(&cur->refcnt)) {
94139027 PNA	422	module_put(THIS_MODULE);
	423	goto err;
	424	}
	425
	426	acct = cur;
	427	break;
	428	}
	429	err:
	430	rcu_read_unlock();
	431	return acct;
	432	}
	433	EXPORT_SYMBOL_GPL(nfnl_acct_find_get);
	434
	435	void nfnl_acct_put(struct nf_acct *acct)
	436	{
b54ab92b	437	if (refcount_dec_and_test(&acct->refcnt))
3499abb2 AS	438	kfree_rcu(acct, rcu_head);
3499abb2 AS	439
94139027 PNA	440	module_put(THIS_MODULE);
	441	}
	442	EXPORT_SYMBOL_GPL(nfnl_acct_put);
	443
	444	void nfnl_acct_update(const struct sk_buff skb, struct nf_acct nfacct)
	445	{
	446	atomic64_inc(&nfacct->pkts);
	447	atomic64_add(skb->len, &nfacct->bytes);
	448	}
	449	EXPORT_SYMBOL_GPL(nfnl_acct_update);
	450
aca30018	451	static void nfnl_overquota_report(struct net net, struct nf_acct nfacct)
683399ed MP	452	{
	453	int ret;
	454	struct sk_buff *skb;
	455
	456	skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC);
	457	if (skb == NULL)
	458	return;
	459
	460	ret = nfnl_acct_fill_info(skb, 0, 0, NFNL_MSG_ACCT_OVERQUOTA, 0,
	461	nfacct);
	462	if (ret <= 0) {
	463	kfree_skb(skb);
	464	return;
	465	}
aca30018	466	netlink_broadcast(net->nfnl, skb, 0, NFNLGRP_ACCT_QUOTA,
683399ed MP	467	GFP_ATOMIC);
	468	}
	469
aca30018 LZ	470	int nfnl_acct_overquota(struct net net, const struct sk_buff skb,
aca30018 LZ	471	struct nf_acct *nfacct)
683399ed MP	472	{
	473	u64 now;
	474	u64 *quota;
	475	int ret = NFACCT_UNDERQUOTA;
	476
	477	/* no place here if we don't have a quota */
	478	if (!(nfacct->flags & NFACCT_F_QUOTA))
	479	return NFACCT_NO_QUOTA;
	480
	481	quota = (u64 *)nfacct->data;
	482	now = (nfacct->flags & NFACCT_F_QUOTA_PKTS) ?
	483	atomic64_read(&nfacct->pkts) : atomic64_read(&nfacct->bytes);
	484
	485	ret = now > *quota;
	486
	487	if (now >= *quota &&
b6d04688	488	!test_and_set_bit(NFACCT_OVERQUOTA_BIT, &nfacct->flags)) {
aca30018	489	nfnl_overquota_report(net, nfacct);
683399ed MP	490	}
	491
	492	return ret;
	493	}
	494	EXPORT_SYMBOL_GPL(nfnl_acct_overquota);
	495
3499abb2 AS	496	static int __net_init nfnl_acct_net_init(struct net *net)
	497	{
	498	INIT_LIST_HEAD(&net->nfnl_acct_list);
	499
	500	return 0;
	501	}
	502
	503	static void __net_exit nfnl_acct_net_exit(struct net *net)
	504	{
	505	struct nf_acct cur, tmp;
	506
	507	list_for_each_entry_safe(cur, tmp, &net->nfnl_acct_list, head) {
	508	list_del_rcu(&cur->head);
	509
b54ab92b	510	if (refcount_dec_and_test(&cur->refcnt))
3499abb2 AS	511	kfree_rcu(cur, rcu_head);
	512	}
	513	}
	514
	515	static struct pernet_operations nfnl_acct_ops = {
	516	.init = nfnl_acct_net_init,
	517	.exit = nfnl_acct_net_exit,
	518	};
	519
94139027 PNA	520	static int __init nfnl_acct_init(void)
	521	{
	522	int ret;
	523
3499abb2 AS	524	ret = register_pernet_subsys(&nfnl_acct_ops);
	525	if (ret < 0) {
	526	pr_err("nfnl_acct_init: failed to register pernet ops\n");
	527	goto err_out;
	528	}
	529
94139027 PNA	530	pr_info("nfnl_acct: registering with nfnetlink.\n");
	531	ret = nfnetlink_subsys_register(&nfnl_acct_subsys);
	532	if (ret < 0) {
	533	pr_err("nfnl_acct_init: cannot register with nfnetlink.\n");
3499abb2	534	goto cleanup_pernet;
94139027 PNA	535	}
94139027 PNA	536	return 0;
3499abb2 AS	537
	538	cleanup_pernet:
	539	unregister_pernet_subsys(&nfnl_acct_ops);
94139027 PNA	540	err_out:
	541	return ret;
	542	}
	543
	544	static void __exit nfnl_acct_exit(void)
	545	{
94139027 PNA	546	pr_info("nfnl_acct: unregistering from nfnetlink.\n");
94139027 PNA	547	nfnetlink_subsys_unregister(&nfnl_acct_subsys);
3499abb2	548	unregister_pernet_subsys(&nfnl_acct_ops);
94139027 PNA	549	}
	550
	551	module_init(nfnl_acct_init);
	552	module_exit(nfnl_acct_exit);