[linux-2.6-block.git] / net / core / sysctl_net_core.c

// SPDX-License-Identifier: GPL-2.0
/* -*- 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);
				static_key_slow_inc(&rfs_needed);
			}
			if (orig_sock_table) {
				static_key_slow_dec(&rps_needed);
				static_key_slow_dec(&rfs_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_dev_weight(struct ctl_table *table, int write,
			   void __user *buffer, size_t *lenp, loff_t *ppos)
{
	int ret;

	ret = proc_dointvec(table, write, buffer, lenp, ppos);
	if (ret != 0)
		return ret;

	dev_rx_weight = weight_p * dev_weight_rx_bias;
	dev_tx_weight = weight_p * dev_weight_tx_bias;

	return ret;
}

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_do_dev_weight,
	},
	{
		.procname	= "dev_weight_rx_bias",
		.data		= &dev_weight_rx_bias,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_do_dev_weight,
	},
	{
		.procname	= "dev_weight_tx_bias",
		.data		= &dev_weight_tx_bias,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_do_dev_weight,
	},
	{
		.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,
#ifndef CONFIG_BPF_JIT_ALWAYS_ON
		.proc_handler	= proc_dointvec
#else
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &one,
		.extra2		= &one,
#endif
	},
# ifdef CONFIG_HAVE_EBPF_JIT
	{
		.procname	= "bpf_jit_harden",
		.data		= &bpf_jit_harden,
		.maxlen		= sizeof(int),
		.mode		= 0600,
		.proc_handler	= proc_dointvec,
	},
	{
		.procname	= "bpf_jit_kallsyms",
		.data		= &bpf_jit_kallsyms,
		.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_minmax,
		.extra1		= &zero,
	},
	{
		.procname	= "busy_read",
		.data		= &sysctl_net_busy_read,
		.maxlen		= sizeof(unsigned int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &zero,
	},
#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,
	},
	{
		.procname	= "netdev_budget_usecs",
		.data		= &netdev_budget_usecs,
		.maxlen		= sizeof(unsigned int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &zero,
	},
	{ }
};

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;

	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
b2441318	1	// SPDX-License-Identifier: GPL-2.0
1da177e4 LT	2	/* -- linux-c --
	3	* sysctl_net_core.c: sysctl interface to net core subsystem.
	4	*
	5	* Begun April 1, 1996, Mike Shaver.
	6	* Added /proc/sys/net/core directory entry (empty =) ). [MS]
	7	*/
	8
	9	#include <linux/mm.h>
	10	#include <linux/sysctl.h>
1da177e4	11	#include <linux/module.h>
20380731	12	#include <linux/socket.h>
a37ae408	13	#include <linux/netdevice.h>
3fff4c42	14	#include <linux/ratelimit.h>
fec5e652	15	#include <linux/vmalloc.h>
33eb9cfc	16	#include <linux/init.h>
5a0e3ad6	17	#include <linux/slab.h>
7426a564	18	#include <linux/kmemleak.h>
3fff4c42	19
63d819ca	20	#include <net/ip.h>
20380731	21	#include <net/sock.h>
c5c177b4	22	#include <net/net_ratelimit.h>
076bb0c8	23	#include <net/busy_poll.h>
6da7c8fc	24	#include <net/pkt_sched.h>
1da177e4	25
5f671d6b	26	static int zero = 0;
cdda8891	27	static int one = 1;
b1cb59cf AK	28	static int min_sndbuf = SOCK_MIN_SNDBUF;
b1cb59cf AK	29	static int min_rcvbuf = SOCK_MIN_RCVBUF;
5f74f82e	30	static int max_skb_frags = MAX_SKB_FRAGS;
cdda8891	31
ba7a46f1 JP	32	static int net_msg_warn; /* Unused, but still a sysctl */
ba7a46f1 JP	33
fec5e652	34	#ifdef CONFIG_RPS
fe2c6338	35	static int rps_sock_flow_sysctl(struct ctl_table *table, int write,
fec5e652 TH	36	void __user buffer, size_t lenp, loff_t *ppos)
	37	{
	38	unsigned int orig_size, size;
	39	int ret, i;
fe2c6338	40	struct ctl_table tmp = {
fec5e652 TH	41	.data = &size,
	42	.maxlen = sizeof(size),
	43	.mode = table->mode
	44	};
	45	struct rps_sock_flow_table orig_sock_table, sock_table;
	46	static DEFINE_MUTEX(sock_flow_mutex);
	47
	48	mutex_lock(&sock_flow_mutex);
	49
6e3f7faf ED	50	orig_sock_table = rcu_dereference_protected(rps_sock_flow_table,
6e3f7faf ED	51	lockdep_is_held(&sock_flow_mutex));
fec5e652 TH	52	size = orig_size = orig_sock_table ? orig_sock_table->mask + 1 : 0;
	53
	54	ret = proc_dointvec(&tmp, write, buffer, lenp, ppos);
	55
	56	if (write) {
	57	if (size) {
93c1af6c	58	if (size > 1<<29) {
fec5e652 TH	59	/* Enforce limit to prevent overflow */
	60	mutex_unlock(&sock_flow_mutex);
	61	return -EINVAL;
	62	}
	63	size = roundup_pow_of_two(size);
	64	if (size != orig_size) {
	65	sock_table =
	66	vmalloc(RPS_SOCK_FLOW_TABLE_SIZE(size));
	67	if (!sock_table) {
	68	mutex_unlock(&sock_flow_mutex);
	69	return -ENOMEM;
	70	}
567e4b79	71	rps_cpu_mask = roundup_pow_of_two(nr_cpu_ids) - 1;
fec5e652 TH	72	sock_table->mask = size - 1;
	73	} else
	74	sock_table = orig_sock_table;
	75
	76	for (i = 0; i < size; i++)
	77	sock_table->ents[i] = RPS_NO_CPU;
	78	} else
	79	sock_table = NULL;
	80
	81	if (sock_table != orig_sock_table) {
	82	rcu_assign_pointer(rps_sock_flow_table, sock_table);
13bfff25	83	if (sock_table) {
c5905afb	84	static_key_slow_inc(&rps_needed);
13bfff25 ED	85	static_key_slow_inc(&rfs_needed);
13bfff25 ED	86	}
adc9300e	87	if (orig_sock_table) {
c5905afb	88	static_key_slow_dec(&rps_needed);
13bfff25	89	static_key_slow_dec(&rfs_needed);
adc9300e ED	90	synchronize_rcu();
	91	vfree(orig_sock_table);
	92	}
fec5e652 TH	93	}
	94	}
	95
	96	mutex_unlock(&sock_flow_mutex);
	97
	98	return ret;
	99	}
	100	#endif /* CONFIG_RPS */
	101
99bbc707 WB	102	#ifdef CONFIG_NET_FLOW_LIMIT
	103	static DEFINE_MUTEX(flow_limit_update_mutex);
	104
fe2c6338	105	static int flow_limit_cpu_sysctl(struct ctl_table *table, int write,
99bbc707 WB	106	void __user buffer, size_t lenp,
	107	loff_t *ppos)
	108	{
	109	struct sd_flow_limit *cur;
	110	struct softnet_data *sd;
	111	cpumask_var_t mask;
	112	int i, len, ret = 0;
	113
	114	if (!alloc_cpumask_var(&mask, GFP_KERNEL))
	115	return -ENOMEM;
	116
	117	if (write) {
	118	ret = cpumask_parse_user(buffer, *lenp, mask);
	119	if (ret)
	120	goto done;
	121
	122	mutex_lock(&flow_limit_update_mutex);
	123	len = sizeof(*cur) + netdev_flow_limit_table_len;
	124	for_each_possible_cpu(i) {
	125	sd = &per_cpu(softnet_data, i);
	126	cur = rcu_dereference_protected(sd->flow_limit,
	127	lockdep_is_held(&flow_limit_update_mutex));
	128	if (cur && !cpumask_test_cpu(i, mask)) {
	129	RCU_INIT_POINTER(sd->flow_limit, NULL);
	130	synchronize_rcu();
	131	kfree(cur);
	132	} else if (!cur && cpumask_test_cpu(i, mask)) {
5b59d467 ED	133	cur = kzalloc_node(len, GFP_KERNEL,
5b59d467 ED	134	cpu_to_node(i));
99bbc707 WB	135	if (!cur) {
	136	/* not unwinding previous changes */
	137	ret = -ENOMEM;
	138	goto write_unlock;
	139	}
	140	cur->num_buckets = netdev_flow_limit_table_len;
	141	rcu_assign_pointer(sd->flow_limit, cur);
	142	}
	143	}
	144	write_unlock:
	145	mutex_unlock(&flow_limit_update_mutex);
	146	} else {
5f121b9a WB	147	char kbuf[128];
5f121b9a WB	148
99bbc707 WB	149	if (ppos \|\| !lenp) {
	150	*lenp = 0;
	151	goto done;
	152	}
	153
	154	cpumask_clear(mask);
	155	rcu_read_lock();
	156	for_each_possible_cpu(i) {
	157	sd = &per_cpu(softnet_data, i);
	158	if (rcu_dereference(sd->flow_limit))
	159	cpumask_set_cpu(i, mask);
	160	}
	161	rcu_read_unlock();
	162
5f121b9a	163	len = min(sizeof(kbuf) - 1, *lenp);
f0906827	164	len = scnprintf(kbuf, len, "%*pb", cpumask_pr_args(mask));
5f121b9a WB	165	if (!len) {
	166	*lenp = 0;
	167	goto done;
	168	}
	169	if (len < *lenp)
	170	kbuf[len++] = '\n';
	171	if (copy_to_user(buffer, kbuf, len)) {
	172	ret = -EFAULT;
	173	goto done;
	174	}
	175	*lenp = len;
	176	*ppos += len;
99bbc707 WB	177	}
	178
	179	done:
	180	free_cpumask_var(mask);
	181	return ret;
	182	}
	183
fe2c6338	184	static int flow_limit_table_len_sysctl(struct ctl_table *table, int write,
99bbc707 WB	185	void __user buffer, size_t lenp,
	186	loff_t *ppos)
	187	{
	188	unsigned int old, *ptr;
	189	int ret;
	190
	191	mutex_lock(&flow_limit_update_mutex);
	192
	193	ptr = table->data;
	194	old = *ptr;
	195	ret = proc_dointvec(table, write, buffer, lenp, ppos);
	196	if (!ret && write && !is_power_of_2(*ptr)) {
	197	*ptr = old;
	198	ret = -EINVAL;
	199	}
	200
	201	mutex_unlock(&flow_limit_update_mutex);
	202	return ret;
	203	}
	204	#endif /* CONFIG_NET_FLOW_LIMIT */
	205
6da7c8fc	206	#ifdef CONFIG_NET_SCHED
	207	static int set_default_qdisc(struct ctl_table *table, int write,
	208	void __user buffer, size_t lenp, loff_t *ppos)
	209	{
	210	char id[IFNAMSIZ];
	211	struct ctl_table tbl = {
	212	.data = id,
	213	.maxlen = IFNAMSIZ,
	214	};
	215	int ret;
	216
	217	qdisc_get_default(id, IFNAMSIZ);
	218
	219	ret = proc_dostring(&tbl, write, buffer, lenp, ppos);
	220	if (write && ret == 0)
	221	ret = qdisc_set_default(id);
	222	return ret;
	223	}
	224	#endif
	225
3d48b53f MT	226	static int proc_do_dev_weight(struct ctl_table *table, int write,
	227	void __user buffer, size_t lenp, loff_t *ppos)
	228	{
	229	int ret;
	230
	231	ret = proc_dointvec(table, write, buffer, lenp, ppos);
	232	if (ret != 0)
	233	return ret;
	234
	235	dev_rx_weight = weight_p * dev_weight_rx_bias;
	236	dev_tx_weight = weight_p * dev_weight_tx_bias;
	237
	238	return ret;
	239	}
	240
960fb622 ED	241	static int proc_do_rss_key(struct ctl_table *table, int write,
	242	void __user buffer, size_t lenp, loff_t *ppos)
	243	{
	244	struct ctl_table fake_table;
	245	char buf[NETDEV_RSS_KEY_LEN * 3];
	246
	247	snprintf(buf, sizeof(buf), "%*phC", NETDEV_RSS_KEY_LEN, netdev_rss_key);
	248	fake_table.data = buf;
	249	fake_table.maxlen = sizeof(buf);
	250	return proc_dostring(&fake_table, write, buffer, lenp, ppos);
	251	}
	252
33eb9cfc	253	static struct ctl_table net_core_table[] = {
1da177e4 LT	254	#ifdef CONFIG_NET
1da177e4 LT	255	{
1da177e4 LT	256	.procname = "wmem_max",
	257	.data = &sysctl_wmem_max,
	258	.maxlen = sizeof(int),
	259	.mode = 0644,
cdda8891	260	.proc_handler = proc_dointvec_minmax,
b1cb59cf	261	.extra1 = &min_sndbuf,
1da177e4 LT	262	},
1da177e4 LT	263	{
1da177e4 LT	264	.procname = "rmem_max",
	265	.data = &sysctl_rmem_max,
	266	.maxlen = sizeof(int),
	267	.mode = 0644,
cdda8891	268	.proc_handler = proc_dointvec_minmax,
b1cb59cf	269	.extra1 = &min_rcvbuf,
1da177e4 LT	270	},
1da177e4 LT	271	{
1da177e4 LT	272	.procname = "wmem_default",
	273	.data = &sysctl_wmem_default,
	274	.maxlen = sizeof(int),
	275	.mode = 0644,
cdda8891	276	.proc_handler = proc_dointvec_minmax,
b1cb59cf	277	.extra1 = &min_sndbuf,
1da177e4 LT	278	},
1da177e4 LT	279	{
1da177e4 LT	280	.procname = "rmem_default",
	281	.data = &sysctl_rmem_default,
	282	.maxlen = sizeof(int),
	283	.mode = 0644,
cdda8891	284	.proc_handler = proc_dointvec_minmax,
b1cb59cf	285	.extra1 = &min_rcvbuf,
1da177e4 LT	286	},
1da177e4 LT	287	{
1da177e4 LT	288	.procname = "dev_weight",
	289	.data = &weight_p,
	290	.maxlen = sizeof(int),
	291	.mode = 0644,
3d48b53f MT	292	.proc_handler = proc_do_dev_weight,
	293	},
	294	{
	295	.procname = "dev_weight_rx_bias",
	296	.data = &dev_weight_rx_bias,
	297	.maxlen = sizeof(int),
	298	.mode = 0644,
	299	.proc_handler = proc_do_dev_weight,
	300	},
	301	{
	302	.procname = "dev_weight_tx_bias",
	303	.data = &dev_weight_tx_bias,
	304	.maxlen = sizeof(int),
	305	.mode = 0644,
	306	.proc_handler = proc_do_dev_weight,
1da177e4 LT	307	},
1da177e4 LT	308	{
1da177e4 LT	309	.procname = "netdev_max_backlog",
	310	.data = &netdev_max_backlog,
	311	.maxlen = sizeof(int),
	312	.mode = 0644,
6d9f239a	313	.proc_handler = proc_dointvec
1da177e4	314	},
960fb622 ED	315	{
	316	.procname = "netdev_rss_key",
	317	.data = &netdev_rss_key,
	318	.maxlen = sizeof(int),
	319	.mode = 0444,
	320	.proc_handler = proc_do_rss_key,
	321	},
0a14842f ED	322	#ifdef CONFIG_BPF_JIT
	323	{
	324	.procname = "bpf_jit_enable",
	325	.data = &bpf_jit_enable,
	326	.maxlen = sizeof(int),
	327	.mode = 0644,
290af866	328	#ifndef CONFIG_BPF_JIT_ALWAYS_ON
0a14842f	329	.proc_handler = proc_dointvec
290af866 AS	330	#else
	331	.proc_handler = proc_dointvec_minmax,
	332	.extra1 = &one,
	333	.extra2 = &one,
	334	#endif
0a14842f	335	},
4f3446bb DB	336	# ifdef CONFIG_HAVE_EBPF_JIT
	337	{
	338	.procname = "bpf_jit_harden",
	339	.data = &bpf_jit_harden,
	340	.maxlen = sizeof(int),
	341	.mode = 0600,
	342	.proc_handler = proc_dointvec,
	343	},
74451e66 DB	344	{
	345	.procname = "bpf_jit_kallsyms",
	346	.data = &bpf_jit_kallsyms,
	347	.maxlen = sizeof(int),
	348	.mode = 0600,
	349	.proc_handler = proc_dointvec,
	350	},
4f3446bb	351	# endif
0a14842f	352	#endif
3b098e2d ED	353	{
	354	.procname = "netdev_tstamp_prequeue",
	355	.data = &netdev_tstamp_prequeue,
	356	.maxlen = sizeof(int),
	357	.mode = 0644,
	358	.proc_handler = proc_dointvec
	359	},
1da177e4	360	{
1da177e4	361	.procname = "message_cost",
717115e1	362	.data = &net_ratelimit_state.interval,
1da177e4 LT	363	.maxlen = sizeof(int),
1da177e4 LT	364	.mode = 0644,
6d9f239a	365	.proc_handler = proc_dointvec_jiffies,
1da177e4 LT	366	},
1da177e4 LT	367	{
1da177e4	368	.procname = "message_burst",
717115e1	369	.data = &net_ratelimit_state.burst,
1da177e4 LT	370	.maxlen = sizeof(int),
1da177e4 LT	371	.mode = 0644,
6d9f239a	372	.proc_handler = proc_dointvec,
1da177e4 LT	373	},
1da177e4 LT	374	{
1da177e4 LT	375	.procname = "optmem_max",
	376	.data = &sysctl_optmem_max,
	377	.maxlen = sizeof(int),
	378	.mode = 0644,
6d9f239a	379	.proc_handler = proc_dointvec
1da177e4	380	},
b245be1f WB	381	{
	382	.procname = "tstamp_allow_data",
	383	.data = &sysctl_tstamp_allow_data,
	384	.maxlen = sizeof(int),
	385	.mode = 0644,
	386	.proc_handler = proc_dointvec_minmax,
	387	.extra1 = &zero,
	388	.extra2 = &one
	389	},
fec5e652 TH	390	#ifdef CONFIG_RPS
	391	{
	392	.procname = "rps_sock_flow_entries",
	393	.maxlen = sizeof(int),
	394	.mode = 0644,
	395	.proc_handler = rps_sock_flow_sysctl
	396	},
	397	#endif
99bbc707 WB	398	#ifdef CONFIG_NET_FLOW_LIMIT
	399	{
	400	.procname = "flow_limit_cpu_bitmap",
	401	.mode = 0644,
	402	.proc_handler = flow_limit_cpu_sysctl
	403	},
	404	{
	405	.procname = "flow_limit_table_len",
	406	.data = &netdev_flow_limit_table_len,
	407	.maxlen = sizeof(int),
	408	.mode = 0644,
	409	.proc_handler = flow_limit_table_len_sysctl
	410	},
	411	#endif /* CONFIG_NET_FLOW_LIMIT */
e0d1095a	412	#ifdef CONFIG_NET_RX_BUSY_POLL
06021292	413	{
64b0dc51 ET	414	.procname = "busy_poll",
64b0dc51 ET	415	.data = &sysctl_net_busy_poll,
eb6db622	416	.maxlen = sizeof(unsigned int),
06021292	417	.mode = 0644,
95f25521 AD	418	.proc_handler = proc_dointvec_minmax,
95f25521 AD	419	.extra1 = &zero,
06021292	420	},
2d48d67f	421	{
64b0dc51 ET	422	.procname = "busy_read",
64b0dc51 ET	423	.data = &sysctl_net_busy_read,
2d48d67f ET	424	.maxlen = sizeof(unsigned int),
2d48d67f ET	425	.mode = 0644,
95f25521 AD	426	.proc_handler = proc_dointvec_minmax,
95f25521 AD	427	.extra1 = &zero,
2d48d67f	428	},
6da7c8fc	429	#endif
	430	#ifdef CONFIG_NET_SCHED
	431	{
	432	.procname = "default_qdisc",
	433	.mode = 0644,
	434	.maxlen = IFNAMSIZ,
	435	.proc_handler = set_default_qdisc
	436	},
06021292	437	#endif
1da177e4	438	#endif /* CONFIG_NET */
51b0bded	439	{
51b0bded SH	440	.procname = "netdev_budget",
	441	.data = &netdev_budget,
	442	.maxlen = sizeof(int),
	443	.mode = 0644,
6d9f239a	444	.proc_handler = proc_dointvec
51b0bded	445	},
a2a316fd	446	{
a2a316fd SH	447	.procname = "warnings",
	448	.data = &net_msg_warn,
	449	.maxlen = sizeof(int),
	450	.mode = 0644,
6d9f239a	451	.proc_handler = proc_dointvec
a2a316fd	452	},
5f74f82e HWR	453	{
	454	.procname = "max_skb_frags",
	455	.data = &sysctl_max_skb_frags,
	456	.maxlen = sizeof(int),
	457	.mode = 0644,
	458	.proc_handler = proc_dointvec_minmax,
	459	.extra1 = &one,
	460	.extra2 = &max_skb_frags,
	461	},
7acf8a1e MW	462	{
	463	.procname = "netdev_budget_usecs",
	464	.data = &netdev_budget_usecs,
	465	.maxlen = sizeof(unsigned int),
	466	.mode = 0644,
	467	.proc_handler = proc_dointvec_minmax,
	468	.extra1 = &zero,
	469	},
f8572d8f	470	{ }
1da177e4	471	};
33eb9cfc	472
d5a4502e PE	473	static struct ctl_table netns_core_table[] = {
d5a4502e PE	474	{
d5a4502e PE	475	.procname = "somaxconn",
	476	.data = &init_net.core.sysctl_somaxconn,
	477	.maxlen = sizeof(int),
	478	.mode = 0644,
5f671d6b	479	.extra1 = &zero,
5f671d6b	480	.proc_handler = proc_dointvec_minmax
d5a4502e	481	},
f8572d8f	482	{ }
d5a4502e PE	483	};
d5a4502e PE	484
024626e3	485	static __net_init int sysctl_core_net_init(struct net *net)
33eb9cfc	486	{
d5a4502e	487	struct ctl_table *tbl;
024626e3	488
d5a4502e	489	tbl = netns_core_table;
09ad9bc7	490	if (!net_eq(net, &init_net)) {
d5a4502e	491	tbl = kmemdup(tbl, sizeof(netns_core_table), GFP_KERNEL);
024626e3 PE	492	if (tbl == NULL)
	493	goto err_dup;
	494
d5a4502e	495	tbl[0].data = &net->core.sysctl_somaxconn;
464dc801 EB	496
	497	/* Don't export any sysctls to unprivileged users */
	498	if (net->user_ns != &init_user_ns) {
	499	tbl[0].procname = NULL;
	500	}
024626e3 PE	501	}
024626e3 PE	502
ec8f23ce	503	net->core.sysctl_hdr = register_net_sysctl(net, "net/core", tbl);
8efa6e93	504	if (net->core.sysctl_hdr == NULL)
024626e3	505	goto err_reg;
33eb9cfc	506
024626e3 PE	507	return 0;
	508
	509	err_reg:
d5a4502e	510	if (tbl != netns_core_table)
024626e3 PE	511	kfree(tbl);
	512	err_dup:
	513	return -ENOMEM;
	514	}
	515
	516	static __net_exit void sysctl_core_net_exit(struct net *net)
	517	{
	518	struct ctl_table *tbl;
	519
8efa6e93 PE	520	tbl = net->core.sysctl_hdr->ctl_table_arg;
8efa6e93 PE	521	unregister_net_sysctl_table(net->core.sysctl_hdr);
d5a4502e	522	BUG_ON(tbl == netns_core_table);
024626e3 PE	523	kfree(tbl);
	524	}
	525
	526	static __net_initdata struct pernet_operations sysctl_core_ops = {
	527	.init = sysctl_core_net_init,
	528	.exit = sysctl_core_net_exit,
	529	};
	530
	531	static __init int sysctl_core_init(void)
	532	{
43444757	533	register_net_sysctl(&init_net, "net/core", net_core_table);
024626e3	534	return register_pernet_subsys(&sysctl_core_ops);
33eb9cfc PE	535	}
33eb9cfc PE	536
b27aeadb	537	fs_initcall(sysctl_core_init);