[linux.git] / net / sched / cls_cgroup.c

/*
 * net/sched/cls_cgroup.c	Control Group Classifier
 *
 *		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; either version
 *		2 of the License, or (at your option) any later version.
 *
 * Authors:	Thomas Graf <[email protected]>
 */

#include <linux/module.h>
#include <linux/slab.h>
#include <linux/types.h>
#include <linux/string.h>
#include <linux/errno.h>
#include <linux/skbuff.h>
#include <linux/cgroup.h>
#include <linux/rcupdate.h>
#include <net/rtnetlink.h>
#include <net/pkt_cls.h>
#include <net/sock.h>
#include <net/cls_cgroup.h>

static struct cgroup_subsys_state *cgrp_create(struct cgroup_subsys *ss,
					       struct cgroup *cgrp);
static void cgrp_destroy(struct cgroup_subsys *ss, struct cgroup *cgrp);
static int cgrp_populate(struct cgroup_subsys *ss, struct cgroup *cgrp);

struct cgroup_subsys net_cls_subsys = {
	.name		= "net_cls",
	.create		= cgrp_create,
	.destroy	= cgrp_destroy,
	.populate	= cgrp_populate,
#ifdef CONFIG_NET_CLS_CGROUP
	.subsys_id	= net_cls_subsys_id,
#endif
	.module		= THIS_MODULE,
};


static inline struct cgroup_cls_state *cgrp_cls_state(struct cgroup *cgrp)
{
	return container_of(cgroup_subsys_state(cgrp, net_cls_subsys_id),
			    struct cgroup_cls_state, css);
}

static inline struct cgroup_cls_state *task_cls_state(struct task_struct *p)
{
	return container_of(task_subsys_state(p, net_cls_subsys_id),
			    struct cgroup_cls_state, css);
}

static struct cgroup_subsys_state *cgrp_create(struct cgroup_subsys *ss,
						 struct cgroup *cgrp)
{
	struct cgroup_cls_state *cs;

	if (!(cs = kzalloc(sizeof(*cs), GFP_KERNEL)))
		return ERR_PTR(-ENOMEM);

	if (cgrp->parent)
		cs->classid = cgrp_cls_state(cgrp->parent)->classid;

	return &cs->css;
}

static void cgrp_destroy(struct cgroup_subsys *ss, struct cgroup *cgrp)
{
	kfree(cgrp_cls_state(cgrp));
}

static u64 read_classid(struct cgroup *cgrp, struct cftype *cft)
{
	return cgrp_cls_state(cgrp)->classid;
}

static int write_classid(struct cgroup *cgrp, struct cftype *cft, u64 value)
{
	cgrp_cls_state(cgrp)->classid = (u32) value;
	return 0;
}

static struct cftype ss_files[] = {
	{
		.name = "classid",
		.read_u64 = read_classid,
		.write_u64 = write_classid,
	},
};

static int cgrp_populate(struct cgroup_subsys *ss, struct cgroup *cgrp)
{
	return cgroup_add_files(cgrp, ss, ss_files, ARRAY_SIZE(ss_files));
}

struct cls_cgroup_head
{
	u32			handle;
	struct tcf_exts		exts;
	struct tcf_ematch_tree	ematches;
};

static int cls_cgroup_classify(struct sk_buff *skb, struct tcf_proto *tp,
			       struct tcf_result *res)
{
	struct cls_cgroup_head *head = tp->root;
	u32 classid;

	rcu_read_lock();
	classid = task_cls_state(current)->classid;
	rcu_read_unlock();

	/*
	 * Due to the nature of the classifier it is required to ignore all
	 * packets originating from softirq context as accessing `current'
	 * would lead to false results.
	 *
	 * This test assumes that all callers of dev_queue_xmit() explicitely
	 * disable bh. Knowing this, it is possible to detect softirq based
	 * calls by looking at the number of nested bh disable calls because
	 * softirqs always disables bh.
	 */
	if (in_serving_softirq()) {
		/* If there is an sk_classid we'll use that. */
		if (!skb->sk)
			return -1;
		classid = skb->sk->sk_classid;
	}

	if (!classid)
		return -1;

	if (!tcf_em_tree_match(skb, &head->ematches, NULL))
		return -1;

	res->classid = classid;
	res->class = 0;
	return tcf_exts_exec(skb, &head->exts, res);
}

static unsigned long cls_cgroup_get(struct tcf_proto *tp, u32 handle)
{
	return 0UL;
}

static void cls_cgroup_put(struct tcf_proto *tp, unsigned long f)
{
}

static int cls_cgroup_init(struct tcf_proto *tp)
{
	return 0;
}

static const struct tcf_ext_map cgroup_ext_map = {
	.action = TCA_CGROUP_ACT,
	.police = TCA_CGROUP_POLICE,
};

static const struct nla_policy cgroup_policy[TCA_CGROUP_MAX + 1] = {
	[TCA_CGROUP_EMATCHES]	= { .type = NLA_NESTED },
};

static int cls_cgroup_change(struct tcf_proto *tp, unsigned long base,
			     u32 handle, struct nlattr **tca,
			     unsigned long *arg)
{
	struct nlattr *tb[TCA_CGROUP_MAX+1];
	struct cls_cgroup_head *head = tp->root;
	struct tcf_ematch_tree t;
	struct tcf_exts e;
	int err;

	if (!tca[TCA_OPTIONS])
		return -EINVAL;

	if (head == NULL) {
		if (!handle)
			return -EINVAL;

		head = kzalloc(sizeof(*head), GFP_KERNEL);
		if (head == NULL)
			return -ENOBUFS;

		head->handle = handle;

		tcf_tree_lock(tp);
		tp->root = head;
		tcf_tree_unlock(tp);
	}

	if (handle != head->handle)
		return -ENOENT;

	err = nla_parse_nested(tb, TCA_CGROUP_MAX, tca[TCA_OPTIONS],
			       cgroup_policy);
	if (err < 0)
		return err;

	err = tcf_exts_validate(tp, tb, tca[TCA_RATE], &e, &cgroup_ext_map);
	if (err < 0)
		return err;

	err = tcf_em_tree_validate(tp, tb[TCA_CGROUP_EMATCHES], &t);
	if (err < 0)
		return err;

	tcf_exts_change(tp, &head->exts, &e);
	tcf_em_tree_change(tp, &head->ematches, &t);

	return 0;
}

static void cls_cgroup_destroy(struct tcf_proto *tp)
{
	struct cls_cgroup_head *head = tp->root;

	if (head) {
		tcf_exts_destroy(tp, &head->exts);
		tcf_em_tree_destroy(tp, &head->ematches);
		kfree(head);
	}
}

static int cls_cgroup_delete(struct tcf_proto *tp, unsigned long arg)
{
	return -EOPNOTSUPP;
}

static void cls_cgroup_walk(struct tcf_proto *tp, struct tcf_walker *arg)
{
	struct cls_cgroup_head *head = tp->root;

	if (arg->count < arg->skip)
		goto skip;

	if (arg->fn(tp, (unsigned long) head, arg) < 0) {
		arg->stop = 1;
		return;
	}
skip:
	arg->count++;
}

static int cls_cgroup_dump(struct tcf_proto *tp, unsigned long fh,
			   struct sk_buff *skb, struct tcmsg *t)
{
	struct cls_cgroup_head *head = tp->root;
	unsigned char *b = skb_tail_pointer(skb);
	struct nlattr *nest;

	t->tcm_handle = head->handle;

	nest = nla_nest_start(skb, TCA_OPTIONS);
	if (nest == NULL)
		goto nla_put_failure;

	if (tcf_exts_dump(skb, &head->exts, &cgroup_ext_map) < 0 ||
	    tcf_em_tree_dump(skb, &head->ematches, TCA_CGROUP_EMATCHES) < 0)
		goto nla_put_failure;

	nla_nest_end(skb, nest);

	if (tcf_exts_dump_stats(skb, &head->exts, &cgroup_ext_map) < 0)
		goto nla_put_failure;

	return skb->len;

nla_put_failure:
	nlmsg_trim(skb, b);
	return -1;
}

static struct tcf_proto_ops cls_cgroup_ops __read_mostly = {
	.kind		=	"cgroup",
	.init		=	cls_cgroup_init,
	.change		=	cls_cgroup_change,
	.classify	=	cls_cgroup_classify,
	.destroy	=	cls_cgroup_destroy,
	.get		=	cls_cgroup_get,
	.put		=	cls_cgroup_put,
	.delete		=	cls_cgroup_delete,
	.walk		=	cls_cgroup_walk,
	.dump		=	cls_cgroup_dump,
	.owner		=	THIS_MODULE,
};

static int __init init_cgroup_cls(void)
{
	int ret;

	ret = cgroup_load_subsys(&net_cls_subsys);
	if (ret)
		goto out;

#ifndef CONFIG_NET_CLS_CGROUP
	/* We can't use rcu_assign_pointer because this is an int. */
	smp_wmb();
	net_cls_subsys_id = net_cls_subsys.subsys_id;
#endif

	ret = register_tcf_proto_ops(&cls_cgroup_ops);
	if (ret)
		cgroup_unload_subsys(&net_cls_subsys);

out:
	return ret;
}

static void __exit exit_cgroup_cls(void)
{
	unregister_tcf_proto_ops(&cls_cgroup_ops);

#ifndef CONFIG_NET_CLS_CGROUP
	net_cls_subsys_id = -1;
	synchronize_rcu();
#endif

	cgroup_unload_subsys(&net_cls_subsys);
}

module_init(init_cgroup_cls);
module_exit(exit_cgroup_cls);
MODULE_LICENSE("GPL");
Commit	Line	Data
f4009237 TG	1	/*
	2	* net/sched/cls_cgroup.c Control Group Classifier
	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
	6	* as published by the Free Software Foundation; either version
	7	* 2 of the License, or (at your option) any later version.
	8	*
	9	* Authors: Thomas Graf <[email protected]>
	10	*/
	11
	12	#include <linux/module.h>
5a0e3ad6	13	#include <linux/slab.h>
f4009237 TG	14	#include <linux/types.h>
	15	#include <linux/string.h>
	16	#include <linux/errno.h>
	17	#include <linux/skbuff.h>
	18	#include <linux/cgroup.h>
f8451725	19	#include <linux/rcupdate.h>
f4009237 TG	20	#include <net/rtnetlink.h>
f4009237 TG	21	#include <net/pkt_cls.h>
f8451725 HX	22	#include <net/sock.h>
f8451725 HX	23	#include <net/cls_cgroup.h>
f4009237	24
8e039d84 BB	25	static struct cgroup_subsys_state cgrp_create(struct cgroup_subsys ss,
	26	struct cgroup *cgrp);
	27	static void cgrp_destroy(struct cgroup_subsys ss, struct cgroup cgrp);
	28	static int cgrp_populate(struct cgroup_subsys ss, struct cgroup cgrp);
	29
	30	struct cgroup_subsys net_cls_subsys = {
	31	.name = "net_cls",
	32	.create = cgrp_create,
	33	.destroy = cgrp_destroy,
	34	.populate = cgrp_populate,
	35	#ifdef CONFIG_NET_CLS_CGROUP
	36	.subsys_id = net_cls_subsys_id,
8e039d84 BB	37	#endif
	38	.module = THIS_MODULE,
	39	};
	40
	41
8e8ba854	42	static inline struct cgroup_cls_state cgrp_cls_state(struct cgroup cgrp)
f4009237	43	{
8e8ba854 LZ	44	return container_of(cgroup_subsys_state(cgrp, net_cls_subsys_id),
	45	struct cgroup_cls_state, css);
	46	}
	47
	48	static inline struct cgroup_cls_state task_cls_state(struct task_struct p)
	49	{
	50	return container_of(task_subsys_state(p, net_cls_subsys_id),
	51	struct cgroup_cls_state, css);
f4009237 TG	52	}
	53
	54	static struct cgroup_subsys_state cgrp_create(struct cgroup_subsys ss,
	55	struct cgroup *cgrp)
	56	{
	57	struct cgroup_cls_state *cs;
	58
	59	if (!(cs = kzalloc(sizeof(*cs), GFP_KERNEL)))
	60	return ERR_PTR(-ENOMEM);
	61
	62	if (cgrp->parent)
8e8ba854	63	cs->classid = cgrp_cls_state(cgrp->parent)->classid;
f4009237 TG	64
	65	return &cs->css;
	66	}
	67
	68	static void cgrp_destroy(struct cgroup_subsys ss, struct cgroup cgrp)
	69	{
8e8ba854	70	kfree(cgrp_cls_state(cgrp));
f4009237 TG	71	}
	72
	73	static u64 read_classid(struct cgroup cgrp, struct cftype cft)
	74	{
8e8ba854	75	return cgrp_cls_state(cgrp)->classid;
f4009237 TG	76	}
	77
	78	static int write_classid(struct cgroup cgrp, struct cftype cft, u64 value)
	79	{
8e8ba854	80	cgrp_cls_state(cgrp)->classid = (u32) value;
f4009237 TG	81	return 0;
	82	}
	83
	84	static struct cftype ss_files[] = {
	85	{
	86	.name = "classid",
	87	.read_u64 = read_classid,
	88	.write_u64 = write_classid,
	89	},
	90	};
	91
	92	static int cgrp_populate(struct cgroup_subsys ss, struct cgroup cgrp)
	93	{
	94	return cgroup_add_files(cgrp, ss, ss_files, ARRAY_SIZE(ss_files));
	95	}
	96
f4009237 TG	97	struct cls_cgroup_head
	98	{
	99	u32 handle;
	100	struct tcf_exts exts;
	101	struct tcf_ematch_tree ematches;
	102	};
	103
	104	static int cls_cgroup_classify(struct sk_buff skb, struct tcf_proto tp,
	105	struct tcf_result *res)
	106	{
	107	struct cls_cgroup_head *head = tp->root;
e65fcfd6	108	u32 classid;
f4009237	109
f8451725 HX	110	rcu_read_lock();
	111	classid = task_cls_state(current)->classid;
	112	rcu_read_unlock();
	113
f4009237 TG	114	/*
	115	* Due to the nature of the classifier it is required to ignore all
	116	* packets originating from softirq context as accessing `current'
	117	* would lead to false results.
	118	*
	119	* This test assumes that all callers of dev_queue_xmit() explicitely
	120	* disable bh. Knowing this, it is possible to detect softirq based
	121	* calls by looking at the number of nested bh disable calls because
	122	* softirqs always disables bh.
	123	*/
75e1056f	124	if (in_serving_softirq()) {
f8451725 HX	125	/* If there is an sk_classid we'll use that. */
	126	if (!skb->sk)
	127	return -1;
	128	classid = skb->sk->sk_classid;
	129	}
f4009237	130
e65fcfd6 PM	131	if (!classid)
	132	return -1;
	133
	134	if (!tcf_em_tree_match(skb, &head->ematches, NULL))
	135	return -1;
	136
	137	res->classid = classid;
	138	res->class = 0;
	139	return tcf_exts_exec(skb, &head->exts, res);
f4009237 TG	140	}
	141
	142	static unsigned long cls_cgroup_get(struct tcf_proto *tp, u32 handle)
	143	{
	144	return 0UL;
	145	}
	146
	147	static void cls_cgroup_put(struct tcf_proto *tp, unsigned long f)
	148	{
	149	}
	150
	151	static int cls_cgroup_init(struct tcf_proto *tp)
	152	{
	153	return 0;
	154	}
	155
	156	static const struct tcf_ext_map cgroup_ext_map = {
	157	.action = TCA_CGROUP_ACT,
	158	.police = TCA_CGROUP_POLICE,
	159	};
	160
	161	static const struct nla_policy cgroup_policy[TCA_CGROUP_MAX + 1] = {
	162	[TCA_CGROUP_EMATCHES] = { .type = NLA_NESTED },
	163	};
	164
	165	static int cls_cgroup_change(struct tcf_proto *tp, unsigned long base,
	166	u32 handle, struct nlattr **tca,
	167	unsigned long *arg)
	168	{
	169	struct nlattr *tb[TCA_CGROUP_MAX+1];
	170	struct cls_cgroup_head *head = tp->root;
	171	struct tcf_ematch_tree t;
	172	struct tcf_exts e;
	173	int err;
	174
52ea3a56 MU	175	if (!tca[TCA_OPTIONS])
	176	return -EINVAL;
	177
f4009237 TG	178	if (head == NULL) {
	179	if (!handle)
	180	return -EINVAL;
	181
	182	head = kzalloc(sizeof(*head), GFP_KERNEL);
	183	if (head == NULL)
	184	return -ENOBUFS;
	185
	186	head->handle = handle;
	187
	188	tcf_tree_lock(tp);
	189	tp->root = head;
	190	tcf_tree_unlock(tp);
	191	}
	192
	193	if (handle != head->handle)
	194	return -ENOENT;
	195
	196	err = nla_parse_nested(tb, TCA_CGROUP_MAX, tca[TCA_OPTIONS],
	197	cgroup_policy);
	198	if (err < 0)
	199	return err;
	200
	201	err = tcf_exts_validate(tp, tb, tca[TCA_RATE], &e, &cgroup_ext_map);
	202	if (err < 0)
	203	return err;
	204
	205	err = tcf_em_tree_validate(tp, tb[TCA_CGROUP_EMATCHES], &t);
	206	if (err < 0)
	207	return err;
	208
	209	tcf_exts_change(tp, &head->exts, &e);
	210	tcf_em_tree_change(tp, &head->ematches, &t);
	211
	212	return 0;
	213	}
	214
	215	static void cls_cgroup_destroy(struct tcf_proto *tp)
	216	{
47a1a1d4	217	struct cls_cgroup_head *head = tp->root;
f4009237 TG	218
	219	if (head) {
	220	tcf_exts_destroy(tp, &head->exts);
	221	tcf_em_tree_destroy(tp, &head->ematches);
	222	kfree(head);
	223	}
	224	}
	225
	226	static int cls_cgroup_delete(struct tcf_proto *tp, unsigned long arg)
	227	{
	228	return -EOPNOTSUPP;
	229	}
	230
	231	static void cls_cgroup_walk(struct tcf_proto tp, struct tcf_walker arg)
	232	{
	233	struct cls_cgroup_head *head = tp->root;
	234
	235	if (arg->count < arg->skip)
	236	goto skip;
	237
	238	if (arg->fn(tp, (unsigned long) head, arg) < 0) {
	239	arg->stop = 1;
	240	return;
	241	}
	242	skip:
	243	arg->count++;
	244	}
	245
	246	static int cls_cgroup_dump(struct tcf_proto *tp, unsigned long fh,
	247	struct sk_buff skb, struct tcmsg t)
	248	{
	249	struct cls_cgroup_head *head = tp->root;
	250	unsigned char *b = skb_tail_pointer(skb);
	251	struct nlattr *nest;
	252
	253	t->tcm_handle = head->handle;
	254
	255	nest = nla_nest_start(skb, TCA_OPTIONS);
	256	if (nest == NULL)
	257	goto nla_put_failure;
	258
	259	if (tcf_exts_dump(skb, &head->exts, &cgroup_ext_map) < 0 \|\|
	260	tcf_em_tree_dump(skb, &head->ematches, TCA_CGROUP_EMATCHES) < 0)
	261	goto nla_put_failure;
	262
	263	nla_nest_end(skb, nest);
	264
	265	if (tcf_exts_dump_stats(skb, &head->exts, &cgroup_ext_map) < 0)
	266	goto nla_put_failure;
	267
	268	return skb->len;
	269
	270	nla_put_failure:
	271	nlmsg_trim(skb, b);
	272	return -1;
	273	}
	274
	275	static struct tcf_proto_ops cls_cgroup_ops __read_mostly = {
	276	.kind = "cgroup",
	277	.init = cls_cgroup_init,
	278	.change = cls_cgroup_change,
	279	.classify = cls_cgroup_classify,
	280	.destroy = cls_cgroup_destroy,
	281	.get = cls_cgroup_get,
282	.put = cls_cgroup_put,
283	.delete = cls_cgroup_delete,
284	.walk = cls_cgroup_walk,
285	.dump = cls_cgroup_dump,
286	.owner = THIS_MODULE,
287	};
288
289	static int __init init_cgroup_cls(void)
290	{
f8451725 HX	291	int ret;
f8451725 HX	292
8e039d84 BB	293	ret = cgroup_load_subsys(&net_cls_subsys);
8e039d84 BB	294	if (ret)
f8451725 HX	295	goto out;
	296
	297	#ifndef CONFIG_NET_CLS_CGROUP
	298	/* We can't use rcu_assign_pointer because this is an int. */
	299	smp_wmb();
	300	net_cls_subsys_id = net_cls_subsys.subsys_id;
	301	#endif
	302
	303	ret = register_tcf_proto_ops(&cls_cgroup_ops);
	304	if (ret)
	305	cgroup_unload_subsys(&net_cls_subsys);
	306
	307	out:
8e039d84	308	return ret;
f4009237 TG	309	}
	310
	311	static void __exit exit_cgroup_cls(void)
	312	{
	313	unregister_tcf_proto_ops(&cls_cgroup_ops);
f8451725 HX	314
	315	#ifndef CONFIG_NET_CLS_CGROUP
	316	net_cls_subsys_id = -1;
	317	synchronize_rcu();
	318	#endif
	319
8e039d84	320	cgroup_unload_subsys(&net_cls_subsys);
f4009237 TG	321	}
	322
	323	module_init(init_cgroup_cls);
	324	module_exit(exit_cgroup_cls);
	325	MODULE_LICENSE("GPL");