[linux.git] / net / core / sysctl_net_core.c

/* -*- linux-c -*-
 * sysctl_net_core.c: sysctl interface to net core subsystem.
 *
 * Begun April 1, 1996, Mike Shaver.
 * Added /proc/sys/net/core directory entry (empty =) ). [MS]
 */

#include <linux/mm.h>
#include <linux/sysctl.h>
#include <linux/module.h>
#include <linux/socket.h>
#include <linux/netdevice.h>
#include <linux/ratelimit.h>
#include <linux/vmalloc.h>
#include <linux/init.h>
#include <linux/slab.h>
#include <linux/kmemleak.h>

#include <net/ip.h>
#include <net/sock.h>
#include <net/net_ratelimit.h>
#include <net/busy_poll.h>
#include <net/pkt_sched.h>

static int zero = 0;
static int one = 1;
static int ushort_max = USHRT_MAX;

#ifdef CONFIG_RPS
static int rps_sock_flow_sysctl(struct ctl_table *table, int write,
				void __user *buffer, size_t *lenp, loff_t *ppos)
{
	unsigned int orig_size, size;
	int ret, i;
	struct ctl_table tmp = {
		.data = &size,
		.maxlen = sizeof(size),
		.mode = table->mode
	};
	struct rps_sock_flow_table *orig_sock_table, *sock_table;
	static DEFINE_MUTEX(sock_flow_mutex);

	mutex_lock(&sock_flow_mutex);

	orig_sock_table = rcu_dereference_protected(rps_sock_flow_table,
					lockdep_is_held(&sock_flow_mutex));
	size = orig_size = orig_sock_table ? orig_sock_table->mask + 1 : 0;

	ret = proc_dointvec(&tmp, write, buffer, lenp, ppos);

	if (write) {
		if (size) {
			if (size > 1<<30) {
				/* Enforce limit to prevent overflow */
				mutex_unlock(&sock_flow_mutex);
				return -EINVAL;
			}
			size = roundup_pow_of_two(size);
			if (size != orig_size) {
				sock_table =
				    vmalloc(RPS_SOCK_FLOW_TABLE_SIZE(size));
				if (!sock_table) {
					mutex_unlock(&sock_flow_mutex);
					return -ENOMEM;
				}

				sock_table->mask = size - 1;
			} else
				sock_table = orig_sock_table;

			for (i = 0; i < size; i++)
				sock_table->ents[i] = RPS_NO_CPU;
		} else
			sock_table = NULL;

		if (sock_table != orig_sock_table) {
			rcu_assign_pointer(rps_sock_flow_table, sock_table);
			if (sock_table)
				static_key_slow_inc(&rps_needed);
			if (orig_sock_table) {
				static_key_slow_dec(&rps_needed);
				synchronize_rcu();
				vfree(orig_sock_table);
			}
		}
	}

	mutex_unlock(&sock_flow_mutex);

	return ret;
}
#endif /* CONFIG_RPS */

#ifdef CONFIG_NET_FLOW_LIMIT
static DEFINE_MUTEX(flow_limit_update_mutex);

static int flow_limit_cpu_sysctl(struct ctl_table *table, int write,
				 void __user *buffer, size_t *lenp,
				 loff_t *ppos)
{
	struct sd_flow_limit *cur;
	struct softnet_data *sd;
	cpumask_var_t mask;
	int i, len, ret = 0;

	if (!alloc_cpumask_var(&mask, GFP_KERNEL))
		return -ENOMEM;

	if (write) {
		ret = cpumask_parse_user(buffer, *lenp, mask);
		if (ret)
			goto done;

		mutex_lock(&flow_limit_update_mutex);
		len = sizeof(*cur) + netdev_flow_limit_table_len;
		for_each_possible_cpu(i) {
			sd = &per_cpu(softnet_data, i);
			cur = rcu_dereference_protected(sd->flow_limit,
				     lockdep_is_held(&flow_limit_update_mutex));
			if (cur && !cpumask_test_cpu(i, mask)) {
				RCU_INIT_POINTER(sd->flow_limit, NULL);
				synchronize_rcu();
				kfree(cur);
			} else if (!cur && cpumask_test_cpu(i, mask)) {
				cur = kzalloc_node(len, GFP_KERNEL,
						   cpu_to_node(i));
				if (!cur) {
					/* not unwinding previous changes */
					ret = -ENOMEM;
					goto write_unlock;
				}
				cur->num_buckets = netdev_flow_limit_table_len;
				rcu_assign_pointer(sd->flow_limit, cur);
			}
		}
write_unlock:
		mutex_unlock(&flow_limit_update_mutex);
	} else {
		char kbuf[128];

		if (*ppos || !*lenp) {
			*lenp = 0;
			goto done;
		}

		cpumask_clear(mask);
		rcu_read_lock();
		for_each_possible_cpu(i) {
			sd = &per_cpu(softnet_data, i);
			if (rcu_dereference(sd->flow_limit))
				cpumask_set_cpu(i, mask);
		}
		rcu_read_unlock();

		len = min(sizeof(kbuf) - 1, *lenp);
		len = cpumask_scnprintf(kbuf, len, mask);
		if (!len) {
			*lenp = 0;
			goto done;
		}
		if (len < *lenp)
			kbuf[len++] = '\n';
		if (copy_to_user(buffer, kbuf, len)) {
			ret = -EFAULT;
			goto done;
		}
		*lenp = len;
		*ppos += len;
	}

done:
	free_cpumask_var(mask);
	return ret;
}

static int flow_limit_table_len_sysctl(struct ctl_table *table, int write,
				       void __user *buffer, size_t *lenp,
				       loff_t *ppos)
{
	unsigned int old, *ptr;
	int ret;

	mutex_lock(&flow_limit_update_mutex);

	ptr = table->data;
	old = *ptr;
	ret = proc_dointvec(table, write, buffer, lenp, ppos);
	if (!ret && write && !is_power_of_2(*ptr)) {
		*ptr = old;
		ret = -EINVAL;
	}

	mutex_unlock(&flow_limit_update_mutex);
	return ret;
}
#endif /* CONFIG_NET_FLOW_LIMIT */

#ifdef CONFIG_NET_SCHED
static int set_default_qdisc(struct ctl_table *table, int write,
			     void __user *buffer, size_t *lenp, loff_t *ppos)
{
	char id[IFNAMSIZ];
	struct ctl_table tbl = {
		.data = id,
		.maxlen = IFNAMSIZ,
	};
	int ret;

	qdisc_get_default(id, IFNAMSIZ);

	ret = proc_dostring(&tbl, write, buffer, lenp, ppos);
	if (write && ret == 0)
		ret = qdisc_set_default(id);
	return ret;
}
#endif

static struct ctl_table net_core_table[] = {
#ifdef CONFIG_NET
	{
		.procname	= "wmem_max",
		.data		= &sysctl_wmem_max,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &one,
	},
	{
		.procname	= "rmem_max",
		.data		= &sysctl_rmem_max,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &one,
	},
	{
		.procname	= "wmem_default",
		.data		= &sysctl_wmem_default,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &one,
	},
	{
		.procname	= "rmem_default",
		.data		= &sysctl_rmem_default,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &one,
	},
	{
		.procname	= "dev_weight",
		.data		= &weight_p,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
	{
		.procname	= "netdev_max_backlog",
		.data		= &netdev_max_backlog,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
#ifdef CONFIG_BPF_JIT
	{
		.procname	= "bpf_jit_enable",
		.data		= &bpf_jit_enable,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
#endif
	{
		.procname	= "netdev_tstamp_prequeue",
		.data		= &netdev_tstamp_prequeue,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
	{
		.procname	= "message_cost",
		.data		= &net_ratelimit_state.interval,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_jiffies,
	},
	{
		.procname	= "message_burst",
		.data		= &net_ratelimit_state.burst,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
	{
		.procname	= "optmem_max",
		.data		= &sysctl_optmem_max,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
#ifdef CONFIG_RPS
	{
		.procname	= "rps_sock_flow_entries",
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= rps_sock_flow_sysctl
	},
#endif
#ifdef CONFIG_NET_FLOW_LIMIT
	{
		.procname	= "flow_limit_cpu_bitmap",
		.mode		= 0644,
		.proc_handler	= flow_limit_cpu_sysctl
	},
	{
		.procname	= "flow_limit_table_len",
		.data		= &netdev_flow_limit_table_len,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= flow_limit_table_len_sysctl
	},
#endif /* CONFIG_NET_FLOW_LIMIT */
#ifdef CONFIG_NET_RX_BUSY_POLL
	{
		.procname	= "busy_poll",
		.data		= &sysctl_net_busy_poll,
		.maxlen		= sizeof(unsigned int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
	{
		.procname	= "busy_read",
		.data		= &sysctl_net_busy_read,
		.maxlen		= sizeof(unsigned int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
#endif
#ifdef CONFIG_NET_SCHED
	{
		.procname	= "default_qdisc",
		.mode		= 0644,
		.maxlen		= IFNAMSIZ,
		.proc_handler	= set_default_qdisc
	},
#endif
#endif /* CONFIG_NET */
	{
		.procname	= "netdev_budget",
		.data		= &netdev_budget,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
	{
		.procname	= "warnings",
		.data		= &net_msg_warn,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
	{ }
};

static struct ctl_table netns_core_table[] = {
	{
		.procname	= "somaxconn",
		.data		= &init_net.core.sysctl_somaxconn,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.extra1		= &zero,
		.extra2		= &ushort_max,
		.proc_handler	= proc_dointvec_minmax
	},
	{ }
};

static __net_init int sysctl_core_net_init(struct net *net)
{
	struct ctl_table *tbl;

	net->core.sysctl_somaxconn = SOMAXCONN;

	tbl = netns_core_table;
	if (!net_eq(net, &init_net)) {
		tbl = kmemdup(tbl, sizeof(netns_core_table), GFP_KERNEL);
		if (tbl == NULL)
			goto err_dup;

		tbl[0].data = &net->core.sysctl_somaxconn;

		/* Don't export any sysctls to unprivileged users */
		if (net->user_ns != &init_user_ns) {
			tbl[0].procname = NULL;
		}
	}

	net->core.sysctl_hdr = register_net_sysctl(net, "net/core", tbl);
	if (net->core.sysctl_hdr == NULL)
		goto err_reg;

	return 0;

err_reg:
	if (tbl != netns_core_table)
		kfree(tbl);
err_dup:
	return -ENOMEM;
}

static __net_exit void sysctl_core_net_exit(struct net *net)
{
	struct ctl_table *tbl;

	tbl = net->core.sysctl_hdr->ctl_table_arg;
	unregister_net_sysctl_table(net->core.sysctl_hdr);
	BUG_ON(tbl == netns_core_table);
	kfree(tbl);
}

static __net_initdata struct pernet_operations sysctl_core_ops = {
	.init = sysctl_core_net_init,
	.exit = sysctl_core_net_exit,
};

static __init int sysctl_core_init(void)
{
	register_net_sysctl(&init_net, "net/core", net_core_table);
	return register_pernet_subsys(&sysctl_core_ops);
}

fs_initcall(sysctl_core_init);
Commit	Line	Data
1da177e4 LT	1	/* -- linux-c --
	2	* sysctl_net_core.c: sysctl interface to net core subsystem.
	3	*
	4	* Begun April 1, 1996, Mike Shaver.
	5	* Added /proc/sys/net/core directory entry (empty =) ). [MS]
	6	*/
	7
	8	#include <linux/mm.h>
	9	#include <linux/sysctl.h>
1da177e4	10	#include <linux/module.h>
20380731	11	#include <linux/socket.h>
a37ae408	12	#include <linux/netdevice.h>
3fff4c42	13	#include <linux/ratelimit.h>
fec5e652	14	#include <linux/vmalloc.h>
33eb9cfc	15	#include <linux/init.h>
5a0e3ad6	16	#include <linux/slab.h>
7426a564	17	#include <linux/kmemleak.h>
3fff4c42	18
63d819ca	19	#include <net/ip.h>
20380731	20	#include <net/sock.h>
c5c177b4	21	#include <net/net_ratelimit.h>
076bb0c8	22	#include <net/busy_poll.h>
6da7c8fc	23	#include <net/pkt_sched.h>
1da177e4	24
5f671d6b	25	static int zero = 0;
cdda8891	26	static int one = 1;
5f671d6b	27	static int ushort_max = USHRT_MAX;
cdda8891	28
fec5e652	29	#ifdef CONFIG_RPS
fe2c6338	30	static int rps_sock_flow_sysctl(struct ctl_table *table, int write,
fec5e652 TH	31	void __user buffer, size_t lenp, loff_t *ppos)
	32	{
	33	unsigned int orig_size, size;
	34	int ret, i;
fe2c6338	35	struct ctl_table tmp = {
fec5e652 TH	36	.data = &size,
	37	.maxlen = sizeof(size),
	38	.mode = table->mode
	39	};
	40	struct rps_sock_flow_table orig_sock_table, sock_table;
	41	static DEFINE_MUTEX(sock_flow_mutex);
	42
	43	mutex_lock(&sock_flow_mutex);
	44
6e3f7faf ED	45	orig_sock_table = rcu_dereference_protected(rps_sock_flow_table,
6e3f7faf ED	46	lockdep_is_held(&sock_flow_mutex));
fec5e652 TH	47	size = orig_size = orig_sock_table ? orig_sock_table->mask + 1 : 0;
	48
	49	ret = proc_dointvec(&tmp, write, buffer, lenp, ppos);
	50
	51	if (write) {
	52	if (size) {
	53	if (size > 1<<30) {
	54	/* Enforce limit to prevent overflow */
	55	mutex_unlock(&sock_flow_mutex);
	56	return -EINVAL;
	57	}
	58	size = roundup_pow_of_two(size);
	59	if (size != orig_size) {
	60	sock_table =
	61	vmalloc(RPS_SOCK_FLOW_TABLE_SIZE(size));
	62	if (!sock_table) {
	63	mutex_unlock(&sock_flow_mutex);
	64	return -ENOMEM;
	65	}
	66
	67	sock_table->mask = size - 1;
	68	} else
	69	sock_table = orig_sock_table;
	70
	71	for (i = 0; i < size; i++)
	72	sock_table->ents[i] = RPS_NO_CPU;
	73	} else
	74	sock_table = NULL;
	75
	76	if (sock_table != orig_sock_table) {
	77	rcu_assign_pointer(rps_sock_flow_table, sock_table);
adc9300e	78	if (sock_table)
c5905afb	79	static_key_slow_inc(&rps_needed);
adc9300e	80	if (orig_sock_table) {
c5905afb	81	static_key_slow_dec(&rps_needed);
adc9300e ED	82	synchronize_rcu();
	83	vfree(orig_sock_table);
	84	}
fec5e652 TH	85	}
	86	}
	87
	88	mutex_unlock(&sock_flow_mutex);
	89
	90	return ret;
	91	}
	92	#endif /* CONFIG_RPS */
	93
99bbc707 WB	94	#ifdef CONFIG_NET_FLOW_LIMIT
	95	static DEFINE_MUTEX(flow_limit_update_mutex);
	96
fe2c6338	97	static int flow_limit_cpu_sysctl(struct ctl_table *table, int write,
99bbc707 WB	98	void __user buffer, size_t lenp,
	99	loff_t *ppos)
	100	{
	101	struct sd_flow_limit *cur;
	102	struct softnet_data *sd;
	103	cpumask_var_t mask;
	104	int i, len, ret = 0;
	105
	106	if (!alloc_cpumask_var(&mask, GFP_KERNEL))
	107	return -ENOMEM;
	108
	109	if (write) {
	110	ret = cpumask_parse_user(buffer, *lenp, mask);
	111	if (ret)
	112	goto done;
	113
	114	mutex_lock(&flow_limit_update_mutex);
	115	len = sizeof(*cur) + netdev_flow_limit_table_len;
	116	for_each_possible_cpu(i) {
	117	sd = &per_cpu(softnet_data, i);
	118	cur = rcu_dereference_protected(sd->flow_limit,
	119	lockdep_is_held(&flow_limit_update_mutex));
	120	if (cur && !cpumask_test_cpu(i, mask)) {
	121	RCU_INIT_POINTER(sd->flow_limit, NULL);
	122	synchronize_rcu();
	123	kfree(cur);
	124	} else if (!cur && cpumask_test_cpu(i, mask)) {
5b59d467 ED	125	cur = kzalloc_node(len, GFP_KERNEL,
5b59d467 ED	126	cpu_to_node(i));
99bbc707 WB	127	if (!cur) {
	128	/* not unwinding previous changes */
	129	ret = -ENOMEM;
	130	goto write_unlock;
	131	}
	132	cur->num_buckets = netdev_flow_limit_table_len;
	133	rcu_assign_pointer(sd->flow_limit, cur);
	134	}
	135	}
	136	write_unlock:
	137	mutex_unlock(&flow_limit_update_mutex);
	138	} else {
5f121b9a WB	139	char kbuf[128];
5f121b9a WB	140
99bbc707 WB	141	if (ppos \|\| !lenp) {
	142	*lenp = 0;
	143	goto done;
	144	}
	145
	146	cpumask_clear(mask);
	147	rcu_read_lock();
	148	for_each_possible_cpu(i) {
	149	sd = &per_cpu(softnet_data, i);
	150	if (rcu_dereference(sd->flow_limit))
	151	cpumask_set_cpu(i, mask);
	152	}
	153	rcu_read_unlock();
	154
5f121b9a WB	155	len = min(sizeof(kbuf) - 1, *lenp);
	156	len = cpumask_scnprintf(kbuf, len, mask);
	157	if (!len) {
	158	*lenp = 0;
	159	goto done;
	160	}
	161	if (len < *lenp)
	162	kbuf[len++] = '\n';
	163	if (copy_to_user(buffer, kbuf, len)) {
	164	ret = -EFAULT;
	165	goto done;
	166	}
	167	*lenp = len;
	168	*ppos += len;
99bbc707 WB	169	}
	170
	171	done:
	172	free_cpumask_var(mask);
	173	return ret;
	174	}
	175
fe2c6338	176	static int flow_limit_table_len_sysctl(struct ctl_table *table, int write,
99bbc707 WB	177	void __user buffer, size_t lenp,
	178	loff_t *ppos)
	179	{
	180	unsigned int old, *ptr;
	181	int ret;
	182
	183	mutex_lock(&flow_limit_update_mutex);
	184
	185	ptr = table->data;
	186	old = *ptr;
	187	ret = proc_dointvec(table, write, buffer, lenp, ppos);
	188	if (!ret && write && !is_power_of_2(*ptr)) {
	189	*ptr = old;
	190	ret = -EINVAL;
	191	}
	192
	193	mutex_unlock(&flow_limit_update_mutex);
	194	return ret;
	195	}
	196	#endif /* CONFIG_NET_FLOW_LIMIT */
	197
6da7c8fc	198	#ifdef CONFIG_NET_SCHED
	199	static int set_default_qdisc(struct ctl_table *table, int write,
	200	void __user buffer, size_t lenp, loff_t *ppos)
	201	{
	202	char id[IFNAMSIZ];
	203	struct ctl_table tbl = {
	204	.data = id,
	205	.maxlen = IFNAMSIZ,
	206	};
	207	int ret;
	208
	209	qdisc_get_default(id, IFNAMSIZ);
	210
	211	ret = proc_dostring(&tbl, write, buffer, lenp, ppos);
	212	if (write && ret == 0)
	213	ret = qdisc_set_default(id);
	214	return ret;
	215	}
	216	#endif
	217
33eb9cfc	218	static struct ctl_table net_core_table[] = {
1da177e4 LT	219	#ifdef CONFIG_NET
1da177e4 LT	220	{
1da177e4 LT	221	.procname = "wmem_max",
	222	.data = &sysctl_wmem_max,
	223	.maxlen = sizeof(int),
	224	.mode = 0644,
cdda8891	225	.proc_handler = proc_dointvec_minmax,
cdda8891	226	.extra1 = &one,
1da177e4 LT	227	},
1da177e4 LT	228	{
1da177e4 LT	229	.procname = "rmem_max",
	230	.data = &sysctl_rmem_max,
	231	.maxlen = sizeof(int),
	232	.mode = 0644,
cdda8891	233	.proc_handler = proc_dointvec_minmax,
cdda8891	234	.extra1 = &one,
1da177e4 LT	235	},
1da177e4 LT	236	{
1da177e4 LT	237	.procname = "wmem_default",
	238	.data = &sysctl_wmem_default,
	239	.maxlen = sizeof(int),
	240	.mode = 0644,
cdda8891	241	.proc_handler = proc_dointvec_minmax,
cdda8891	242	.extra1 = &one,
1da177e4 LT	243	},
1da177e4 LT	244	{
1da177e4 LT	245	.procname = "rmem_default",
	246	.data = &sysctl_rmem_default,
	247	.maxlen = sizeof(int),
	248	.mode = 0644,
cdda8891	249	.proc_handler = proc_dointvec_minmax,
cdda8891	250	.extra1 = &one,
1da177e4 LT	251	},
1da177e4 LT	252	{
1da177e4 LT	253	.procname = "dev_weight",
	254	.data = &weight_p,
	255	.maxlen = sizeof(int),
	256	.mode = 0644,
6d9f239a	257	.proc_handler = proc_dointvec
1da177e4 LT	258	},
1da177e4 LT	259	{
1da177e4 LT	260	.procname = "netdev_max_backlog",
	261	.data = &netdev_max_backlog,
	262	.maxlen = sizeof(int),
	263	.mode = 0644,
6d9f239a	264	.proc_handler = proc_dointvec
1da177e4	265	},
0a14842f ED	266	#ifdef CONFIG_BPF_JIT
	267	{
	268	.procname = "bpf_jit_enable",
	269	.data = &bpf_jit_enable,
	270	.maxlen = sizeof(int),
	271	.mode = 0644,
	272	.proc_handler = proc_dointvec
	273	},
	274	#endif
3b098e2d ED	275	{
	276	.procname = "netdev_tstamp_prequeue",
	277	.data = &netdev_tstamp_prequeue,
	278	.maxlen = sizeof(int),
	279	.mode = 0644,
	280	.proc_handler = proc_dointvec
	281	},
1da177e4	282	{
1da177e4	283	.procname = "message_cost",
717115e1	284	.data = &net_ratelimit_state.interval,
1da177e4 LT	285	.maxlen = sizeof(int),
1da177e4 LT	286	.mode = 0644,
6d9f239a	287	.proc_handler = proc_dointvec_jiffies,
1da177e4 LT	288	},
1da177e4 LT	289	{
1da177e4	290	.procname = "message_burst",
717115e1	291	.data = &net_ratelimit_state.burst,
1da177e4 LT	292	.maxlen = sizeof(int),
1da177e4 LT	293	.mode = 0644,
6d9f239a	294	.proc_handler = proc_dointvec,
1da177e4 LT	295	},
1da177e4 LT	296	{
1da177e4 LT	297	.procname = "optmem_max",
	298	.data = &sysctl_optmem_max,
	299	.maxlen = sizeof(int),
	300	.mode = 0644,
6d9f239a	301	.proc_handler = proc_dointvec
1da177e4	302	},
fec5e652 TH	303	#ifdef CONFIG_RPS
	304	{
	305	.procname = "rps_sock_flow_entries",
	306	.maxlen = sizeof(int),
	307	.mode = 0644,
	308	.proc_handler = rps_sock_flow_sysctl
	309	},
	310	#endif
99bbc707 WB	311	#ifdef CONFIG_NET_FLOW_LIMIT
	312	{
	313	.procname = "flow_limit_cpu_bitmap",
	314	.mode = 0644,
	315	.proc_handler = flow_limit_cpu_sysctl
	316	},
	317	{
	318	.procname = "flow_limit_table_len",
	319	.data = &netdev_flow_limit_table_len,
	320	.maxlen = sizeof(int),
	321	.mode = 0644,
	322	.proc_handler = flow_limit_table_len_sysctl
	323	},
	324	#endif /* CONFIG_NET_FLOW_LIMIT */
e0d1095a	325	#ifdef CONFIG_NET_RX_BUSY_POLL
06021292	326	{
64b0dc51 ET	327	.procname = "busy_poll",
64b0dc51 ET	328	.data = &sysctl_net_busy_poll,
eb6db622	329	.maxlen = sizeof(unsigned int),
06021292	330	.mode = 0644,
eb6db622	331	.proc_handler = proc_dointvec
06021292	332	},
2d48d67f	333	{
64b0dc51 ET	334	.procname = "busy_read",
64b0dc51 ET	335	.data = &sysctl_net_busy_read,
2d48d67f ET	336	.maxlen = sizeof(unsigned int),
	337	.mode = 0644,
	338	.proc_handler = proc_dointvec
	339	},
6da7c8fc	340	#endif
	341	#ifdef CONFIG_NET_SCHED
	342	{
	343	.procname = "default_qdisc",
	344	.mode = 0644,
	345	.maxlen = IFNAMSIZ,
	346	.proc_handler = set_default_qdisc
	347	},
06021292	348	#endif
1da177e4	349	#endif /* CONFIG_NET */
51b0bded	350	{
51b0bded SH	351	.procname = "netdev_budget",
	352	.data = &netdev_budget,
	353	.maxlen = sizeof(int),
	354	.mode = 0644,
6d9f239a	355	.proc_handler = proc_dointvec
51b0bded	356	},
a2a316fd	357	{
a2a316fd SH	358	.procname = "warnings",
	359	.data = &net_msg_warn,
	360	.maxlen = sizeof(int),
	361	.mode = 0644,
6d9f239a	362	.proc_handler = proc_dointvec
a2a316fd	363	},
f8572d8f	364	{ }
1da177e4	365	};
33eb9cfc	366
d5a4502e PE	367	static struct ctl_table netns_core_table[] = {
d5a4502e PE	368	{
d5a4502e PE	369	.procname = "somaxconn",
	370	.data = &init_net.core.sysctl_somaxconn,
	371	.maxlen = sizeof(int),
	372	.mode = 0644,
5f671d6b RG	373	.extra1 = &zero,
	374	.extra2 = &ushort_max,
	375	.proc_handler = proc_dointvec_minmax
d5a4502e	376	},
f8572d8f	377	{ }
d5a4502e PE	378	};
d5a4502e PE	379
024626e3	380	static __net_init int sysctl_core_net_init(struct net *net)
33eb9cfc	381	{
d5a4502e	382	struct ctl_table *tbl;
024626e3	383
8efa6e93	384	net->core.sysctl_somaxconn = SOMAXCONN;
b8e1f9b5	385
d5a4502e	386	tbl = netns_core_table;
09ad9bc7	387	if (!net_eq(net, &init_net)) {
d5a4502e	388	tbl = kmemdup(tbl, sizeof(netns_core_table), GFP_KERNEL);
024626e3 PE	389	if (tbl == NULL)
	390	goto err_dup;
	391
d5a4502e	392	tbl[0].data = &net->core.sysctl_somaxconn;
464dc801 EB	393
	394	/* Don't export any sysctls to unprivileged users */
	395	if (net->user_ns != &init_user_ns) {
	396	tbl[0].procname = NULL;
	397	}
024626e3 PE	398	}
024626e3 PE	399
ec8f23ce	400	net->core.sysctl_hdr = register_net_sysctl(net, "net/core", tbl);
8efa6e93	401	if (net->core.sysctl_hdr == NULL)
024626e3	402	goto err_reg;
33eb9cfc	403
024626e3 PE	404	return 0;
	405
	406	err_reg:
d5a4502e	407	if (tbl != netns_core_table)
024626e3 PE	408	kfree(tbl);
	409	err_dup:
	410	return -ENOMEM;
	411	}
	412
	413	static __net_exit void sysctl_core_net_exit(struct net *net)
	414	{
	415	struct ctl_table *tbl;
	416
8efa6e93 PE	417	tbl = net->core.sysctl_hdr->ctl_table_arg;
8efa6e93 PE	418	unregister_net_sysctl_table(net->core.sysctl_hdr);
d5a4502e	419	BUG_ON(tbl == netns_core_table);
024626e3 PE	420	kfree(tbl);
	421	}
	422
	423	static __net_initdata struct pernet_operations sysctl_core_ops = {
	424	.init = sysctl_core_net_init,
	425	.exit = sysctl_core_net_exit,
	426	};
	427
	428	static __init int sysctl_core_init(void)
	429	{
43444757	430	register_net_sysctl(&init_net, "net/core", net_core_table);
024626e3	431	return register_pernet_subsys(&sysctl_core_ops);
33eb9cfc PE	432	}
33eb9cfc PE	433
b27aeadb	434	fs_initcall(sysctl_core_init);