[linux-2.6-block.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 min_sndbuf = SOCK_MIN_SNDBUF;
static int min_rcvbuf = SOCK_MIN_RCVBUF;
static int max_skb_frags = MAX_SKB_FRAGS;

static int net_msg_warn;	/* Unused, but still a sysctl */

#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<<29) {
				/* 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;
				}
				rps_cpu_mask = roundup_pow_of_two(nr_cpu_ids) - 1;
				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 = scnprintf(kbuf, len, "%*pb", cpumask_pr_args(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 int proc_do_rss_key(struct ctl_table *table, int write,
			   void __user *buffer, size_t *lenp, loff_t *ppos)
{
	struct ctl_table fake_table;
	char buf[NETDEV_RSS_KEY_LEN * 3];

	snprintf(buf, sizeof(buf), "%*phC", NETDEV_RSS_KEY_LEN, netdev_rss_key);
	fake_table.data = buf;
	fake_table.maxlen = sizeof(buf);
	return proc_dostring(&fake_table, write, buffer, lenp, ppos);
}

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		= &min_sndbuf,
	},
	{
		.procname	= "rmem_max",
		.data		= &sysctl_rmem_max,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &min_rcvbuf,
	},
	{
		.procname	= "wmem_default",
		.data		= &sysctl_wmem_default,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &min_sndbuf,
	},
	{
		.procname	= "rmem_default",
		.data		= &sysctl_rmem_default,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &min_rcvbuf,
	},
	{
		.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
	},
	{
		.procname	= "netdev_rss_key",
		.data		= &netdev_rss_key,
		.maxlen		= sizeof(int),
		.mode		= 0444,
		.proc_handler	= proc_do_rss_key,
	},
#ifdef CONFIG_BPF_JIT
	{
		.procname	= "bpf_jit_enable",
		.data		= &bpf_jit_enable,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
# ifdef CONFIG_HAVE_EBPF_JIT
	{
		.procname	= "bpf_jit_harden",
		.data		= &bpf_jit_harden,
		.maxlen		= sizeof(int),
		.mode		= 0600,
		.proc_handler	= proc_dointvec,
	},
# endif
#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
	},
	{
		.procname	= "tstamp_allow_data",
		.data		= &sysctl_tstamp_allow_data,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &zero,
		.extra2		= &one
	},
#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
	},
	{
		.procname	= "max_skb_frags",
		.data		= &sysctl_max_skb_frags,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &one,
		.extra2		= &max_skb_frags,
	},
	{ }
};

static struct ctl_table netns_core_table[] = {
	{
		.procname	= "somaxconn",
		.data		= &init_net.core.sysctl_somaxconn,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.extra1		= &zero,
		.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;
b1cb59cf AK	27	static int min_sndbuf = SOCK_MIN_SNDBUF;
b1cb59cf AK	28	static int min_rcvbuf = SOCK_MIN_RCVBUF;
5f74f82e	29	static int max_skb_frags = MAX_SKB_FRAGS;
cdda8891	30
ba7a46f1 JP	31	static int net_msg_warn; /* Unused, but still a sysctl */
ba7a46f1 JP	32
fec5e652	33	#ifdef CONFIG_RPS
fe2c6338	34	static int rps_sock_flow_sysctl(struct ctl_table *table, int write,
fec5e652 TH	35	void __user buffer, size_t lenp, loff_t *ppos)
	36	{
	37	unsigned int orig_size, size;
	38	int ret, i;
fe2c6338	39	struct ctl_table tmp = {
fec5e652 TH	40	.data = &size,
	41	.maxlen = sizeof(size),
	42	.mode = table->mode
	43	};
	44	struct rps_sock_flow_table orig_sock_table, sock_table;
	45	static DEFINE_MUTEX(sock_flow_mutex);
	46
	47	mutex_lock(&sock_flow_mutex);
	48
6e3f7faf ED	49	orig_sock_table = rcu_dereference_protected(rps_sock_flow_table,
6e3f7faf ED	50	lockdep_is_held(&sock_flow_mutex));
fec5e652 TH	51	size = orig_size = orig_sock_table ? orig_sock_table->mask + 1 : 0;
	52
	53	ret = proc_dointvec(&tmp, write, buffer, lenp, ppos);
	54
	55	if (write) {
	56	if (size) {
93c1af6c	57	if (size > 1<<29) {
fec5e652 TH	58	/* Enforce limit to prevent overflow */
	59	mutex_unlock(&sock_flow_mutex);
	60	return -EINVAL;
	61	}
	62	size = roundup_pow_of_two(size);
	63	if (size != orig_size) {
	64	sock_table =
	65	vmalloc(RPS_SOCK_FLOW_TABLE_SIZE(size));
	66	if (!sock_table) {
	67	mutex_unlock(&sock_flow_mutex);
	68	return -ENOMEM;
	69	}
567e4b79	70	rps_cpu_mask = roundup_pow_of_two(nr_cpu_ids) - 1;
fec5e652 TH	71	sock_table->mask = size - 1;
	72	} else
	73	sock_table = orig_sock_table;
	74
	75	for (i = 0; i < size; i++)
	76	sock_table->ents[i] = RPS_NO_CPU;
	77	} else
	78	sock_table = NULL;
	79
	80	if (sock_table != orig_sock_table) {
	81	rcu_assign_pointer(rps_sock_flow_table, sock_table);
adc9300e	82	if (sock_table)
c5905afb	83	static_key_slow_inc(&rps_needed);
adc9300e	84	if (orig_sock_table) {
c5905afb	85	static_key_slow_dec(&rps_needed);
adc9300e ED	86	synchronize_rcu();
	87	vfree(orig_sock_table);
	88	}
fec5e652 TH	89	}
	90	}
	91
	92	mutex_unlock(&sock_flow_mutex);
	93
	94	return ret;
	95	}
	96	#endif /* CONFIG_RPS */
	97
99bbc707 WB	98	#ifdef CONFIG_NET_FLOW_LIMIT
	99	static DEFINE_MUTEX(flow_limit_update_mutex);
	100
fe2c6338	101	static int flow_limit_cpu_sysctl(struct ctl_table *table, int write,
99bbc707 WB	102	void __user buffer, size_t lenp,
	103	loff_t *ppos)
	104	{
	105	struct sd_flow_limit *cur;
	106	struct softnet_data *sd;
	107	cpumask_var_t mask;
	108	int i, len, ret = 0;
	109
	110	if (!alloc_cpumask_var(&mask, GFP_KERNEL))
	111	return -ENOMEM;
	112
	113	if (write) {
	114	ret = cpumask_parse_user(buffer, *lenp, mask);
	115	if (ret)
	116	goto done;
	117
	118	mutex_lock(&flow_limit_update_mutex);
	119	len = sizeof(*cur) + netdev_flow_limit_table_len;
	120	for_each_possible_cpu(i) {
	121	sd = &per_cpu(softnet_data, i);
	122	cur = rcu_dereference_protected(sd->flow_limit,
	123	lockdep_is_held(&flow_limit_update_mutex));
	124	if (cur && !cpumask_test_cpu(i, mask)) {
	125	RCU_INIT_POINTER(sd->flow_limit, NULL);
	126	synchronize_rcu();
	127	kfree(cur);
	128	} else if (!cur && cpumask_test_cpu(i, mask)) {
5b59d467 ED	129	cur = kzalloc_node(len, GFP_KERNEL,
5b59d467 ED	130	cpu_to_node(i));
99bbc707 WB	131	if (!cur) {
	132	/* not unwinding previous changes */
	133	ret = -ENOMEM;
	134	goto write_unlock;
	135	}
	136	cur->num_buckets = netdev_flow_limit_table_len;
	137	rcu_assign_pointer(sd->flow_limit, cur);
	138	}
	139	}
	140	write_unlock:
	141	mutex_unlock(&flow_limit_update_mutex);
	142	} else {
5f121b9a WB	143	char kbuf[128];
5f121b9a WB	144
99bbc707 WB	145	if (ppos \|\| !lenp) {
	146	*lenp = 0;
	147	goto done;
	148	}
	149
	150	cpumask_clear(mask);
	151	rcu_read_lock();
	152	for_each_possible_cpu(i) {
	153	sd = &per_cpu(softnet_data, i);
	154	if (rcu_dereference(sd->flow_limit))
	155	cpumask_set_cpu(i, mask);
	156	}
	157	rcu_read_unlock();
	158
5f121b9a	159	len = min(sizeof(kbuf) - 1, *lenp);
f0906827	160	len = scnprintf(kbuf, len, "%*pb", cpumask_pr_args(mask));
5f121b9a WB	161	if (!len) {
	162	*lenp = 0;
	163	goto done;
	164	}
	165	if (len < *lenp)
	166	kbuf[len++] = '\n';
	167	if (copy_to_user(buffer, kbuf, len)) {
	168	ret = -EFAULT;
	169	goto done;
	170	}
	171	*lenp = len;
	172	*ppos += len;
99bbc707 WB	173	}
	174
	175	done:
	176	free_cpumask_var(mask);
	177	return ret;
	178	}
	179
fe2c6338	180	static int flow_limit_table_len_sysctl(struct ctl_table *table, int write,
99bbc707 WB	181	void __user buffer, size_t lenp,
	182	loff_t *ppos)
	183	{
	184	unsigned int old, *ptr;
	185	int ret;
	186
	187	mutex_lock(&flow_limit_update_mutex);
	188
	189	ptr = table->data;
	190	old = *ptr;
	191	ret = proc_dointvec(table, write, buffer, lenp, ppos);
	192	if (!ret && write && !is_power_of_2(*ptr)) {
	193	*ptr = old;
	194	ret = -EINVAL;
	195	}
	196
	197	mutex_unlock(&flow_limit_update_mutex);
	198	return ret;
	199	}
	200	#endif /* CONFIG_NET_FLOW_LIMIT */
	201
6da7c8fc	202	#ifdef CONFIG_NET_SCHED
	203	static int set_default_qdisc(struct ctl_table *table, int write,
	204	void __user buffer, size_t lenp, loff_t *ppos)
	205	{
	206	char id[IFNAMSIZ];
	207	struct ctl_table tbl = {
	208	.data = id,
	209	.maxlen = IFNAMSIZ,
	210	};
	211	int ret;
	212
	213	qdisc_get_default(id, IFNAMSIZ);
	214
	215	ret = proc_dostring(&tbl, write, buffer, lenp, ppos);
	216	if (write && ret == 0)
	217	ret = qdisc_set_default(id);
	218	return ret;
	219	}
	220	#endif
	221
960fb622 ED	222	static int proc_do_rss_key(struct ctl_table *table, int write,
	223	void __user buffer, size_t lenp, loff_t *ppos)
	224	{
	225	struct ctl_table fake_table;
	226	char buf[NETDEV_RSS_KEY_LEN * 3];
	227
	228	snprintf(buf, sizeof(buf), "%*phC", NETDEV_RSS_KEY_LEN, netdev_rss_key);
	229	fake_table.data = buf;
	230	fake_table.maxlen = sizeof(buf);
	231	return proc_dostring(&fake_table, write, buffer, lenp, ppos);
	232	}
	233
33eb9cfc	234	static struct ctl_table net_core_table[] = {
1da177e4 LT	235	#ifdef CONFIG_NET
1da177e4 LT	236	{
1da177e4 LT	237	.procname = "wmem_max",
	238	.data = &sysctl_wmem_max,
	239	.maxlen = sizeof(int),
	240	.mode = 0644,
cdda8891	241	.proc_handler = proc_dointvec_minmax,
b1cb59cf	242	.extra1 = &min_sndbuf,
1da177e4 LT	243	},
1da177e4 LT	244	{
1da177e4 LT	245	.procname = "rmem_max",
	246	.data = &sysctl_rmem_max,
	247	.maxlen = sizeof(int),
	248	.mode = 0644,
cdda8891	249	.proc_handler = proc_dointvec_minmax,
b1cb59cf	250	.extra1 = &min_rcvbuf,
1da177e4 LT	251	},
1da177e4 LT	252	{
1da177e4 LT	253	.procname = "wmem_default",
	254	.data = &sysctl_wmem_default,
	255	.maxlen = sizeof(int),
	256	.mode = 0644,
cdda8891	257	.proc_handler = proc_dointvec_minmax,
b1cb59cf	258	.extra1 = &min_sndbuf,
1da177e4 LT	259	},
1da177e4 LT	260	{
1da177e4 LT	261	.procname = "rmem_default",
	262	.data = &sysctl_rmem_default,
	263	.maxlen = sizeof(int),
	264	.mode = 0644,
cdda8891	265	.proc_handler = proc_dointvec_minmax,
b1cb59cf	266	.extra1 = &min_rcvbuf,
1da177e4 LT	267	},
1da177e4 LT	268	{
1da177e4 LT	269	.procname = "dev_weight",
	270	.data = &weight_p,
	271	.maxlen = sizeof(int),
	272	.mode = 0644,
6d9f239a	273	.proc_handler = proc_dointvec
1da177e4 LT	274	},
1da177e4 LT	275	{
1da177e4 LT	276	.procname = "netdev_max_backlog",
	277	.data = &netdev_max_backlog,
	278	.maxlen = sizeof(int),
	279	.mode = 0644,
6d9f239a	280	.proc_handler = proc_dointvec
1da177e4	281	},
960fb622 ED	282	{
	283	.procname = "netdev_rss_key",
	284	.data = &netdev_rss_key,
	285	.maxlen = sizeof(int),
	286	.mode = 0444,
	287	.proc_handler = proc_do_rss_key,
	288	},
0a14842f ED	289	#ifdef CONFIG_BPF_JIT
	290	{
	291	.procname = "bpf_jit_enable",
	292	.data = &bpf_jit_enable,
	293	.maxlen = sizeof(int),
	294	.mode = 0644,
	295	.proc_handler = proc_dointvec
	296	},
4f3446bb DB	297	# ifdef CONFIG_HAVE_EBPF_JIT
	298	{
	299	.procname = "bpf_jit_harden",
	300	.data = &bpf_jit_harden,
	301	.maxlen = sizeof(int),
	302	.mode = 0600,
	303	.proc_handler = proc_dointvec,
	304	},
	305	# endif
0a14842f	306	#endif
3b098e2d ED	307	{
	308	.procname = "netdev_tstamp_prequeue",
	309	.data = &netdev_tstamp_prequeue,
	310	.maxlen = sizeof(int),
	311	.mode = 0644,
	312	.proc_handler = proc_dointvec
	313	},
1da177e4	314	{
1da177e4	315	.procname = "message_cost",
717115e1	316	.data = &net_ratelimit_state.interval,
1da177e4 LT	317	.maxlen = sizeof(int),
1da177e4 LT	318	.mode = 0644,
6d9f239a	319	.proc_handler = proc_dointvec_jiffies,
1da177e4 LT	320	},
1da177e4 LT	321	{
1da177e4	322	.procname = "message_burst",
717115e1	323	.data = &net_ratelimit_state.burst,
1da177e4 LT	324	.maxlen = sizeof(int),
1da177e4 LT	325	.mode = 0644,
6d9f239a	326	.proc_handler = proc_dointvec,
1da177e4 LT	327	},
1da177e4 LT	328	{
1da177e4 LT	329	.procname = "optmem_max",
	330	.data = &sysctl_optmem_max,
	331	.maxlen = sizeof(int),
	332	.mode = 0644,
6d9f239a	333	.proc_handler = proc_dointvec
1da177e4	334	},
b245be1f WB	335	{
	336	.procname = "tstamp_allow_data",
	337	.data = &sysctl_tstamp_allow_data,
	338	.maxlen = sizeof(int),
	339	.mode = 0644,
	340	.proc_handler = proc_dointvec_minmax,
	341	.extra1 = &zero,
	342	.extra2 = &one
	343	},
fec5e652 TH	344	#ifdef CONFIG_RPS
	345	{
	346	.procname = "rps_sock_flow_entries",
	347	.maxlen = sizeof(int),
	348	.mode = 0644,
	349	.proc_handler = rps_sock_flow_sysctl
	350	},
	351	#endif
99bbc707 WB	352	#ifdef CONFIG_NET_FLOW_LIMIT
	353	{
	354	.procname = "flow_limit_cpu_bitmap",
	355	.mode = 0644,
	356	.proc_handler = flow_limit_cpu_sysctl
	357	},
	358	{
	359	.procname = "flow_limit_table_len",
	360	.data = &netdev_flow_limit_table_len,
	361	.maxlen = sizeof(int),
	362	.mode = 0644,
	363	.proc_handler = flow_limit_table_len_sysctl
	364	},
	365	#endif /* CONFIG_NET_FLOW_LIMIT */
e0d1095a	366	#ifdef CONFIG_NET_RX_BUSY_POLL
06021292	367	{
64b0dc51 ET	368	.procname = "busy_poll",
64b0dc51 ET	369	.data = &sysctl_net_busy_poll,
eb6db622	370	.maxlen = sizeof(unsigned int),
06021292	371	.mode = 0644,
eb6db622	372	.proc_handler = proc_dointvec
06021292	373	},
2d48d67f	374	{
64b0dc51 ET	375	.procname = "busy_read",
64b0dc51 ET	376	.data = &sysctl_net_busy_read,
2d48d67f ET	377	.maxlen = sizeof(unsigned int),
	378	.mode = 0644,
	379	.proc_handler = proc_dointvec
	380	},
6da7c8fc	381	#endif
	382	#ifdef CONFIG_NET_SCHED
	383	{
	384	.procname = "default_qdisc",
	385	.mode = 0644,
	386	.maxlen = IFNAMSIZ,
	387	.proc_handler = set_default_qdisc
	388	},
06021292	389	#endif
1da177e4	390	#endif /* CONFIG_NET */
51b0bded	391	{
51b0bded SH	392	.procname = "netdev_budget",
	393	.data = &netdev_budget,
	394	.maxlen = sizeof(int),
	395	.mode = 0644,
6d9f239a	396	.proc_handler = proc_dointvec
51b0bded	397	},
a2a316fd	398	{
a2a316fd SH	399	.procname = "warnings",
	400	.data = &net_msg_warn,
	401	.maxlen = sizeof(int),
	402	.mode = 0644,
6d9f239a	403	.proc_handler = proc_dointvec
a2a316fd	404	},
5f74f82e HWR	405	{
	406	.procname = "max_skb_frags",
	407	.data = &sysctl_max_skb_frags,
	408	.maxlen = sizeof(int),
	409	.mode = 0644,
	410	.proc_handler = proc_dointvec_minmax,
	411	.extra1 = &one,
	412	.extra2 = &max_skb_frags,
	413	},
f8572d8f	414	{ }
1da177e4	415	};
33eb9cfc	416
d5a4502e PE	417	static struct ctl_table netns_core_table[] = {
d5a4502e PE	418	{
d5a4502e PE	419	.procname = "somaxconn",
	420	.data = &init_net.core.sysctl_somaxconn,
	421	.maxlen = sizeof(int),
	422	.mode = 0644,
5f671d6b	423	.extra1 = &zero,
5f671d6b	424	.proc_handler = proc_dointvec_minmax
d5a4502e	425	},
f8572d8f	426	{ }
d5a4502e PE	427	};
d5a4502e PE	428
024626e3	429	static __net_init int sysctl_core_net_init(struct net *net)
33eb9cfc	430	{
d5a4502e	431	struct ctl_table *tbl;
024626e3	432
8efa6e93	433	net->core.sysctl_somaxconn = SOMAXCONN;
b8e1f9b5	434
d5a4502e	435	tbl = netns_core_table;
09ad9bc7	436	if (!net_eq(net, &init_net)) {
d5a4502e	437	tbl = kmemdup(tbl, sizeof(netns_core_table), GFP_KERNEL);
024626e3 PE	438	if (tbl == NULL)
	439	goto err_dup;
	440
d5a4502e	441	tbl[0].data = &net->core.sysctl_somaxconn;
464dc801 EB	442
	443	/* Don't export any sysctls to unprivileged users */
	444	if (net->user_ns != &init_user_ns) {
	445	tbl[0].procname = NULL;
	446	}
024626e3 PE	447	}
024626e3 PE	448
ec8f23ce	449	net->core.sysctl_hdr = register_net_sysctl(net, "net/core", tbl);
8efa6e93	450	if (net->core.sysctl_hdr == NULL)
024626e3	451	goto err_reg;
33eb9cfc	452
024626e3 PE	453	return 0;
	454
	455	err_reg:
d5a4502e	456	if (tbl != netns_core_table)
024626e3 PE	457	kfree(tbl);
	458	err_dup:
	459	return -ENOMEM;
	460	}
	461
	462	static __net_exit void sysctl_core_net_exit(struct net *net)
	463	{
	464	struct ctl_table *tbl;
	465
8efa6e93 PE	466	tbl = net->core.sysctl_hdr->ctl_table_arg;
8efa6e93 PE	467	unregister_net_sysctl_table(net->core.sysctl_hdr);
d5a4502e	468	BUG_ON(tbl == netns_core_table);
024626e3 PE	469	kfree(tbl);
	470	}
	471
	472	static __net_initdata struct pernet_operations sysctl_core_ops = {
	473	.init = sysctl_core_net_init,
	474	.exit = sysctl_core_net_exit,
	475	};
	476
	477	static __init int sysctl_core_init(void)
	478	{
43444757	479	register_net_sysctl(&init_net, "net/core", net_core_table);
024626e3	480	return register_pernet_subsys(&sysctl_core_ops);
33eb9cfc PE	481	}
33eb9cfc PE	482
b27aeadb	483	fs_initcall(sysctl_core_init);