sysctl_net_ipv4.c 34.8 KB
Newer Older
1
// SPDX-License-Identifier: GPL-2.0
L
Linus Torvalds 已提交
2 3 4 5 6 7 8 9 10 11
/*
 * sysctl_net_ipv4.c: sysctl interface to net IPV4 subsystem.
 *
 * Begun April 1, 1996, Mike Shaver.
 * Added /proc/sys/net/ipv4 directory entry (empty =) ). [MS]
 */

#include <linux/mm.h>
#include <linux/module.h>
#include <linux/sysctl.h>
12
#include <linux/igmp.h>
13
#include <linux/inetdevice.h>
14
#include <linux/seqlock.h>
15
#include <linux/init.h>
16
#include <linux/slab.h>
17
#include <linux/nsproxy.h>
G
Glauber Costa 已提交
18
#include <linux/swap.h>
L
Linus Torvalds 已提交
19
#include <net/snmp.h>
20
#include <net/icmp.h>
L
Linus Torvalds 已提交
21 22 23
#include <net/ip.h>
#include <net/route.h>
#include <net/tcp.h>
H
Hideo Aoki 已提交
24
#include <net/udp.h>
P
Paul Moore 已提交
25
#include <net/cipso_ipv4.h>
26
#include <net/inet_frag.h>
27
#include <net/ping.h>
28
#include <net/protocol.h>
29
#include <net/netevent.h>
L
Linus Torvalds 已提交
30

31
static int two = 2;
N
Nandita Dukkipati 已提交
32
static int four = 4;
33
static int thousand = 1000;
E
Eric Dumazet 已提交
34
static int gso_max_segs = GSO_MAX_SEGS;
35
static int tcp_retr1_max = 255;
L
Linus Torvalds 已提交
36 37
static int ip_local_port_range_min[] = { 1, 1 };
static int ip_local_port_range_max[] = { 65535, 65535 };
38 39
static int tcp_adv_win_scale_min = -31;
static int tcp_adv_win_scale_max = 31;
E
Eric Dumazet 已提交
40 41
static int tcp_min_snd_mss_min = TCP_MIN_SND_MSS;
static int tcp_min_snd_mss_max = 65535;
42 43
static int ip_privileged_port_min;
static int ip_privileged_port_max = 65535;
E
Eric Dumazet 已提交
44 45
static int ip_ttl_min = 1;
static int ip_ttl_max = 255;
46 47
static int tcp_syn_retries_min = 1;
static int tcp_syn_retries_max = MAX_TCP_SYNCNT;
48 49
static int ip_ping_group_range_min[] = { 0, 0 };
static int ip_ping_group_range_max[] = { GID_T_MAX, GID_T_MAX };
E
Eric Dumazet 已提交
50
static int comp_sack_nr_max = 255;
51
static u32 u32_max_div_HZ = UINT_MAX / HZ;
52
static int one_day_secs = 24 * 3600;
L
Linus Torvalds 已提交
53

54 55 56
/* obsolete */
static int sysctl_tcp_low_latency __read_mostly;

57
/* Update system visible IP port range */
58
static void set_local_port_range(struct net *net, int range[2])
59
{
60 61
	bool same_parity = !((range[0] ^ range[1]) & 1);

62
	write_seqlock_bh(&net->ipv4.ip_local_ports.lock);
63 64 65 66
	if (same_parity && !net->ipv4.ip_local_ports.warned) {
		net->ipv4.ip_local_ports.warned = true;
		pr_err_ratelimited("ip_local_port_range: prefer different parity for start/end values.\n");
	}
67 68
	net->ipv4.ip_local_ports.range[0] = range[0];
	net->ipv4.ip_local_ports.range[1] = range[1];
69
	write_sequnlock_bh(&net->ipv4.ip_local_ports.lock);
70 71 72
}

/* Validate changes from /proc interface. */
73
static int ipv4_local_port_range(struct ctl_table *table, int write,
74
				 void *buffer, size_t *lenp, loff_t *ppos)
75
{
76
	struct net *net =
77
		container_of(table->data, struct net, ipv4.ip_local_ports.range);
78
	int ret;
E
Eric Dumazet 已提交
79
	int range[2];
80
	struct ctl_table tmp = {
81 82 83 84 85 86 87
		.data = &range,
		.maxlen = sizeof(range),
		.mode = table->mode,
		.extra1 = &ip_local_port_range_min,
		.extra2 = &ip_local_port_range_max,
	};

88 89
	inet_get_local_port_range(net, &range[0], &range[1]);

90
	ret = proc_dointvec_minmax(&tmp, write, buffer, lenp, ppos);
91 92

	if (write && ret == 0) {
93 94 95 96 97 98
		/* Ensure that the upper limit is not smaller than the lower,
		 * and that the lower does not encroach upon the privileged
		 * port limit.
		 */
		if ((range[1] < range[0]) ||
		    (range[0] < net->ipv4.sysctl_ip_prot_sock))
99 100
			ret = -EINVAL;
		else
101
			set_local_port_range(net, range);
102 103 104 105 106
	}

	return ret;
}

107 108
/* Validate changes from /proc interface. */
static int ipv4_privileged_ports(struct ctl_table *table, int write,
109
				void *buffer, size_t *lenp, loff_t *ppos)
110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140
{
	struct net *net = container_of(table->data, struct net,
	    ipv4.sysctl_ip_prot_sock);
	int ret;
	int pports;
	int range[2];
	struct ctl_table tmp = {
		.data = &pports,
		.maxlen = sizeof(pports),
		.mode = table->mode,
		.extra1 = &ip_privileged_port_min,
		.extra2 = &ip_privileged_port_max,
	};

	pports = net->ipv4.sysctl_ip_prot_sock;

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

	if (write && ret == 0) {
		inet_get_local_port_range(net, &range[0], &range[1]);
		/* Ensure that the local port range doesn't overlap with the
		 * privileged port range.
		 */
		if (range[0] < pports)
			ret = -EINVAL;
		else
			net->ipv4.sysctl_ip_prot_sock = pports;
	}

	return ret;
}
141

142
static void inet_get_ping_group_range_table(struct ctl_table *table, kgid_t *low, kgid_t *high)
143
{
144
	kgid_t *data = table->data;
145
	struct net *net =
146
		container_of(table->data, struct net, ipv4.ping_group_range.range);
147
	unsigned int seq;
148
	do {
149
		seq = read_seqbegin(&net->ipv4.ping_group_range.lock);
150 151 152

		*low = data[0];
		*high = data[1];
153
	} while (read_seqretry(&net->ipv4.ping_group_range.lock, seq));
154 155 156
}

/* Update system visible IP port range */
157
static void set_ping_group_range(struct ctl_table *table, kgid_t low, kgid_t high)
158
{
159
	kgid_t *data = table->data;
160
	struct net *net =
161
		container_of(table->data, struct net, ipv4.ping_group_range.range);
162
	write_seqlock(&net->ipv4.ping_group_range.lock);
163 164
	data[0] = low;
	data[1] = high;
165
	write_sequnlock(&net->ipv4.ping_group_range.lock);
166 167 168
}

/* Validate changes from /proc interface. */
169
static int ipv4_ping_group_range(struct ctl_table *table, int write,
170
				 void *buffer, size_t *lenp, loff_t *ppos)
171
{
172
	struct user_namespace *user_ns = current_user_ns();
173
	int ret;
174 175
	gid_t urange[2];
	kgid_t low, high;
176
	struct ctl_table tmp = {
177 178
		.data = &urange,
		.maxlen = sizeof(urange),
179 180 181 182 183
		.mode = table->mode,
		.extra1 = &ip_ping_group_range_min,
		.extra2 = &ip_ping_group_range_max,
	};

184 185 186
	inet_get_ping_group_range_table(table, &low, &high);
	urange[0] = from_kgid_munged(user_ns, low);
	urange[1] = from_kgid_munged(user_ns, high);
187 188
	ret = proc_dointvec_minmax(&tmp, write, buffer, lenp, ppos);

189 190 191
	if (write && ret == 0) {
		low = make_kgid(user_ns, urange[0]);
		high = make_kgid(user_ns, urange[1]);
192 193 194
		if (!gid_valid(low) || !gid_valid(high))
			return -EINVAL;
		if (urange[1] < urange[0] || gid_lt(high, low)) {
195 196 197 198 199
			low = make_kgid(&init_user_ns, 1);
			high = make_kgid(&init_user_ns, 0);
		}
		set_ping_group_range(table, low, high);
	}
200 201 202 203

	return ret;
}

204
static int ipv4_fwd_update_priority(struct ctl_table *table, int write,
205
				    void *buffer, size_t *lenp, loff_t *ppos)
206 207 208 209 210 211 212 213 214 215 216 217 218 219
{
	struct net *net;
	int ret;

	net = container_of(table->data, struct net,
			   ipv4.sysctl_ip_fwd_update_priority);
	ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
	if (write && ret == 0)
		call_netevent_notifiers(NETEVENT_IPV4_FWD_UPDATE_PRIORITY_UPDATE,
					net);

	return ret;
}

220
static int proc_tcp_congestion_control(struct ctl_table *ctl, int write,
221
				       void *buffer, size_t *lenp, loff_t *ppos)
222
{
223 224
	struct net *net = container_of(ctl->data, struct net,
				       ipv4.tcp_congestion_control);
225
	char val[TCP_CA_NAME_MAX];
226
	struct ctl_table tbl = {
227 228 229 230 231
		.data = val,
		.maxlen = TCP_CA_NAME_MAX,
	};
	int ret;

232
	tcp_get_default_congestion_control(net, val);
233

234
	ret = proc_dostring(&tbl, write, buffer, lenp, ppos);
235
	if (write && ret == 0)
236
		ret = tcp_set_default_congestion_control(net, val);
237 238 239
	return ret;
}

240
static int proc_tcp_available_congestion_control(struct ctl_table *ctl,
241 242
						 int write, void *buffer,
						 size_t *lenp, loff_t *ppos)
243
{
244
	struct ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX, };
245 246 247 248 249 250
	int ret;

	tbl.data = kmalloc(tbl.maxlen, GFP_USER);
	if (!tbl.data)
		return -ENOMEM;
	tcp_get_available_congestion_control(tbl.data, TCP_CA_BUF_MAX);
251
	ret = proc_dostring(&tbl, write, buffer, lenp, ppos);
252 253 254 255
	kfree(tbl.data);
	return ret;
}

256
static int proc_allowed_congestion_control(struct ctl_table *ctl,
257 258
					   int write, void *buffer,
					   size_t *lenp, loff_t *ppos)
259
{
260
	struct ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX };
261 262 263 264 265 266 267
	int ret;

	tbl.data = kmalloc(tbl.maxlen, GFP_USER);
	if (!tbl.data)
		return -ENOMEM;

	tcp_get_allowed_congestion_control(tbl.data, tbl.maxlen);
268
	ret = proc_dostring(&tbl, write, buffer, lenp, ppos);
269 270 271 272 273 274
	if (write && ret == 0)
		ret = tcp_set_allowed_congestion_control(tbl.data);
	kfree(tbl.data);
	return ret;
}

275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292
static int sscanf_key(char *buf, __le32 *key)
{
	u32 user_key[4];
	int i, ret = 0;

	if (sscanf(buf, "%x-%x-%x-%x", user_key, user_key + 1,
		   user_key + 2, user_key + 3) != 4) {
		ret = -EINVAL;
	} else {
		for (i = 0; i < ARRAY_SIZE(user_key); i++)
			key[i] = cpu_to_le32(user_key[i]);
	}
	pr_debug("proc TFO key set 0x%x-%x-%x-%x <- 0x%s: %u\n",
		 user_key[0], user_key[1], user_key[2], user_key[3], buf, ret);

	return ret;
}

293
static int proc_tcp_fastopen_key(struct ctl_table *table, int write,
294
				 void *buffer, size_t *lenp, loff_t *ppos)
295
{
296 297
	struct net *net = container_of(table->data, struct net,
	    ipv4.sysctl_tcp_fastopen);
298 299 300 301 302 303 304 305 306 307 308
	/* maxlen to print the list of keys in hex (*2), with dashes
	 * separating doublewords and a comma in between keys.
	 */
	struct ctl_table tbl = { .maxlen = ((TCP_FASTOPEN_KEY_LENGTH *
					    2 * TCP_FASTOPEN_KEY_MAX) +
					    (TCP_FASTOPEN_KEY_MAX * 5)) };
	struct tcp_fastopen_context *ctx;
	u32 user_key[TCP_FASTOPEN_KEY_MAX * 4];
	__le32 key[TCP_FASTOPEN_KEY_MAX * 4];
	char *backup_data;
	int ret, i = 0, off = 0, n_keys = 0;
309 310 311 312 313 314

	tbl.data = kmalloc(tbl.maxlen, GFP_KERNEL);
	if (!tbl.data)
		return -ENOMEM;

	rcu_read_lock();
315 316 317 318 319
	ctx = rcu_dereference(net->ipv4.tcp_fastopen_ctx);
	if (ctx) {
		n_keys = tcp_fastopen_context_len(ctx);
		memcpy(&key[0], &ctx->key[0], TCP_FASTOPEN_KEY_LENGTH * n_keys);
	}
320 321
	rcu_read_unlock();

322 323 324 325 326 327
	if (!n_keys) {
		memset(&key[0], 0, TCP_FASTOPEN_KEY_LENGTH);
		n_keys = 1;
	}

	for (i = 0; i < n_keys * 4; i++)
328 329
		user_key[i] = le32_to_cpu(key[i]);

330 331 332 333 334 335 336
	for (i = 0; i < n_keys; i++) {
		off += snprintf(tbl.data + off, tbl.maxlen - off,
				"%08x-%08x-%08x-%08x",
				user_key[i * 4],
				user_key[i * 4 + 1],
				user_key[i * 4 + 2],
				user_key[i * 4 + 3]);
337 338 339 340

		if (WARN_ON_ONCE(off >= tbl.maxlen - 1))
			break;

341 342 343 344
		if (i + 1 < n_keys)
			off += snprintf(tbl.data + off, tbl.maxlen - off, ",");
	}

345 346 347
	ret = proc_dostring(&tbl, write, buffer, lenp, ppos);

	if (write && ret == 0) {
348 349 350 351 352 353
		backup_data = strchr(tbl.data, ',');
		if (backup_data) {
			*backup_data = '\0';
			backup_data++;
		}
		if (sscanf_key(tbl.data, key)) {
354 355 356
			ret = -EINVAL;
			goto bad_key;
		}
357 358 359 360 361 362 363
		if (backup_data) {
			if (sscanf_key(backup_data, key + 4)) {
				ret = -EINVAL;
				goto bad_key;
			}
		}
		tcp_fastopen_reset_cipher(net, NULL, key,
364
					  backup_data ? key + 4 : NULL);
365 366 367 368 369 370 371
	}

bad_key:
	kfree(tbl.data);
	return ret;
}

372 373 374 375 376 377 378
static void proc_configure_early_demux(int enabled, int protocol)
{
	struct net_protocol *ipprot;
#if IS_ENABLED(CONFIG_IPV6)
	struct inet6_protocol *ip6prot;
#endif

379 380
	rcu_read_lock();

381 382 383 384 385 386 387 388 389 390 391
	ipprot = rcu_dereference(inet_protos[protocol]);
	if (ipprot)
		ipprot->early_demux = enabled ? ipprot->early_demux_handler :
						NULL;

#if IS_ENABLED(CONFIG_IPV6)
	ip6prot = rcu_dereference(inet6_protos[protocol]);
	if (ip6prot)
		ip6prot->early_demux = enabled ? ip6prot->early_demux_handler :
						 NULL;
#endif
392
	rcu_read_unlock();
393 394 395
}

static int proc_tcp_early_demux(struct ctl_table *table, int write,
396
				void *buffer, size_t *lenp, loff_t *ppos)
397 398 399 400 401 402 403 404 405 406 407 408 409 410 411
{
	int ret = 0;

	ret = proc_dointvec(table, write, buffer, lenp, ppos);

	if (write && !ret) {
		int enabled = init_net.ipv4.sysctl_tcp_early_demux;

		proc_configure_early_demux(enabled, IPPROTO_TCP);
	}

	return ret;
}

static int proc_udp_early_demux(struct ctl_table *table, int write,
412
				void *buffer, size_t *lenp, loff_t *ppos)
413 414 415 416 417 418 419 420 421 422 423 424 425 426
{
	int ret = 0;

	ret = proc_dointvec(table, write, buffer, lenp, ppos);

	if (write && !ret) {
		int enabled = init_net.ipv4.sysctl_udp_early_demux;

		proc_configure_early_demux(enabled, IPPROTO_UDP);
	}

	return ret;
}

427
static int proc_tfo_blackhole_detect_timeout(struct ctl_table *table,
428
					     int write, void *buffer,
429 430
					     size_t *lenp, loff_t *ppos)
{
431 432
	struct net *net = container_of(table->data, struct net,
	    ipv4.sysctl_tcp_fastopen_blackhole_timeout);
433 434 435 436
	int ret;

	ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
	if (write && ret == 0)
437
		atomic_set(&net->ipv4.tfo_active_disable_times, 0);
D
Dave Watson 已提交
438 439 440 441 442

	return ret;
}

static int proc_tcp_available_ulp(struct ctl_table *ctl,
443
				  int write, void *buffer, size_t *lenp,
D
Dave Watson 已提交
444 445 446 447 448 449 450 451 452 453 454 455
				  loff_t *ppos)
{
	struct ctl_table tbl = { .maxlen = TCP_ULP_BUF_MAX, };
	int ret;

	tbl.data = kmalloc(tbl.maxlen, GFP_USER);
	if (!tbl.data)
		return -ENOMEM;
	tcp_get_available_ulp(tbl.data, TCP_ULP_BUF_MAX);
	ret = proc_dostring(&tbl, write, buffer, lenp, ppos);
	kfree(tbl.data);

456 457 458
	return ret;
}

459 460
#ifdef CONFIG_IP_ROUTE_MULTIPATH
static int proc_fib_multipath_hash_policy(struct ctl_table *table, int write,
461
					  void *buffer, size_t *lenp,
462 463 464 465 466 467 468 469
					  loff_t *ppos)
{
	struct net *net = container_of(table->data, struct net,
	    ipv4.sysctl_fib_multipath_hash_policy);
	int ret;

	ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
	if (write && ret == 0)
470
		call_netevent_notifiers(NETEVENT_IPV4_MPATH_HASH_UPDATE, net);
471 472 473 474 475

	return ret;
}
#endif

476
static struct ctl_table ipv4_table[] = {
L
Linus Torvalds 已提交
477 478 479 480 481
	{
		.procname	= "tcp_max_orphans",
		.data		= &sysctl_tcp_max_orphans,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
482
		.proc_handler	= proc_dointvec
L
Linus Torvalds 已提交
483 484 485 486 487 488
	},
	{
		.procname	= "inet_peer_threshold",
		.data		= &inet_peer_threshold,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
489
		.proc_handler	= proc_dointvec
L
Linus Torvalds 已提交
490 491 492 493 494 495
	},
	{
		.procname	= "inet_peer_minttl",
		.data		= &inet_peer_minttl,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
496
		.proc_handler	= proc_dointvec_jiffies,
L
Linus Torvalds 已提交
497 498 499 500 501 502
	},
	{
		.procname	= "inet_peer_maxttl",
		.data		= &inet_peer_maxttl,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
503
		.proc_handler	= proc_dointvec_jiffies,
L
Linus Torvalds 已提交
504
	},
505 506 507 508 509 510 511
	{
		.procname	= "tcp_mem",
		.maxlen		= sizeof(sysctl_tcp_mem),
		.data		= &sysctl_tcp_mem,
		.mode		= 0644,
		.proc_handler	= proc_doulongvec_minmax,
	},
L
Linus Torvalds 已提交
512 513 514 515 516
	{
		.procname	= "tcp_low_latency",
		.data		= &sysctl_tcp_low_latency,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
517
		.proc_handler	= proc_dointvec
L
Linus Torvalds 已提交
518
	},
P
Paul Moore 已提交
519 520 521 522 523 524
#ifdef CONFIG_NETLABEL
	{
		.procname	= "cipso_cache_enable",
		.data		= &cipso_v4_cache_enabled,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
525
		.proc_handler	= proc_dointvec,
P
Paul Moore 已提交
526 527 528 529 530 531
	},
	{
		.procname	= "cipso_cache_bucket_size",
		.data		= &cipso_v4_cache_bucketsize,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
532
		.proc_handler	= proc_dointvec,
P
Paul Moore 已提交
533 534 535 536 537 538
	},
	{
		.procname	= "cipso_rbm_optfmt",
		.data		= &cipso_v4_rbm_optfmt,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
539
		.proc_handler	= proc_dointvec,
P
Paul Moore 已提交
540 541 542 543 544 545
	},
	{
		.procname	= "cipso_rbm_strictvalid",
		.data		= &cipso_v4_rbm_strictvalid,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
546
		.proc_handler	= proc_dointvec,
P
Paul Moore 已提交
547 548
	},
#endif /* CONFIG_NETLABEL */
D
Dave Watson 已提交
549 550 551 552 553 554
	{
		.procname	= "tcp_available_ulp",
		.maxlen		= TCP_ULP_BUF_MAX,
		.mode		= 0444,
		.proc_handler   = proc_tcp_available_ulp,
	},
555 556 557 558 559 560
	{
		.procname	= "icmp_msgs_per_sec",
		.data		= &sysctl_icmp_msgs_per_sec,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
561
		.extra1		= SYSCTL_ZERO,
562 563 564 565 566 567 568
	},
	{
		.procname	= "icmp_msgs_burst",
		.data		= &sysctl_icmp_msgs_burst,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
569
		.extra1		= SYSCTL_ZERO,
570
	},
H
Hideo Aoki 已提交
571 572 573 574 575
	{
		.procname	= "udp_mem",
		.data		= &sysctl_udp_mem,
		.maxlen		= sizeof(sysctl_udp_mem),
		.mode		= 0644,
E
Eric Dumazet 已提交
576
		.proc_handler	= proc_doulongvec_minmax,
H
Hideo Aoki 已提交
577
	},
578 579 580 581 582 583 584 585 586
	{
		.procname	= "fib_sync_mem",
		.data		= &sysctl_fib_sync_mem,
		.maxlen		= sizeof(sysctl_fib_sync_mem),
		.mode		= 0644,
		.proc_handler	= proc_douintvec_minmax,
		.extra1		= &sysctl_fib_sync_mem_min,
		.extra2		= &sysctl_fib_sync_mem_max,
	},
E
Eric Dumazet 已提交
587 588 589 590 591 592
	{
		.procname	= "tcp_rx_skb_cache",
		.data		= &tcp_rx_skb_cache_key.key,
		.mode		= 0644,
		.proc_handler	= proc_do_static_key,
	},
E
Eric Dumazet 已提交
593 594 595 596 597 598
	{
		.procname	= "tcp_tx_skb_cache",
		.data		= &tcp_tx_skb_cache_key.key,
		.mode		= 0644,
		.proc_handler	= proc_do_static_key,
	},
599
	{ }
L
Linus Torvalds 已提交
600
};
601

602 603 604 605 606 607
static struct ctl_table ipv4_net_table[] = {
	{
		.procname	= "icmp_echo_ignore_all",
		.data		= &init_net.ipv4.sysctl_icmp_echo_ignore_all,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
608
		.proc_handler	= proc_dointvec
609 610 611 612 613 614
	},
	{
		.procname	= "icmp_echo_ignore_broadcasts",
		.data		= &init_net.ipv4.sysctl_icmp_echo_ignore_broadcasts,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
615
		.proc_handler	= proc_dointvec
616 617 618 619 620 621
	},
	{
		.procname	= "icmp_ignore_bogus_error_responses",
		.data		= &init_net.ipv4.sysctl_icmp_ignore_bogus_error_responses,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
622
		.proc_handler	= proc_dointvec
623 624 625 626 627 628
	},
	{
		.procname	= "icmp_errors_use_inbound_ifaddr",
		.data		= &init_net.ipv4.sysctl_icmp_errors_use_inbound_ifaddr,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
629
		.proc_handler	= proc_dointvec
630 631 632 633 634 635
	},
	{
		.procname	= "icmp_ratelimit",
		.data		= &init_net.ipv4.sysctl_icmp_ratelimit,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
636
		.proc_handler	= proc_dointvec_ms_jiffies,
637 638 639 640 641 642
	},
	{
		.procname	= "icmp_ratemask",
		.data		= &init_net.ipv4.sysctl_icmp_ratemask,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
643
		.proc_handler	= proc_dointvec
644
	},
645 646
	{
		.procname	= "ping_group_range",
647
		.data		= &init_net.ipv4.ping_group_range.range,
648
		.maxlen		= sizeof(gid_t)*2,
649 650 651
		.mode		= 0644,
		.proc_handler	= ipv4_ping_group_range,
	},
652 653 654 655 656 657 658
#ifdef CONFIG_NET_L3_MASTER_DEV
	{
		.procname	= "raw_l3mdev_accept",
		.data		= &init_net.ipv4.sysctl_raw_l3mdev_accept,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
659 660
		.extra1		= SYSCTL_ZERO,
		.extra2		= SYSCTL_ONE,
661 662
	},
#endif
663 664 665 666 667 668 669
	{
		.procname	= "tcp_ecn",
		.data		= &init_net.ipv4.sysctl_tcp_ecn,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
670 671 672 673 674 675 676
	{
		.procname	= "tcp_ecn_fallback",
		.data		= &init_net.ipv4.sysctl_tcp_ecn_fallback,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
677 678 679 680 681 682 683
	{
		.procname	= "ip_dynaddr",
		.data		= &init_net.ipv4.sysctl_ip_dynaddr,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
684 685 686 687 688 689 690
	{
		.procname	= "ip_early_demux",
		.data		= &init_net.ipv4.sysctl_ip_early_demux,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
691 692 693 694 695 696 697 698 699 700 701 702 703 704
	{
		.procname       = "udp_early_demux",
		.data           = &init_net.ipv4.sysctl_udp_early_demux,
		.maxlen         = sizeof(int),
		.mode           = 0644,
		.proc_handler   = proc_udp_early_demux
	},
	{
		.procname       = "tcp_early_demux",
		.data           = &init_net.ipv4.sysctl_tcp_early_demux,
		.maxlen         = sizeof(int),
		.mode           = 0644,
		.proc_handler   = proc_tcp_early_demux
	},
705 706 707 708 709 710 711 712 713
	{
		.procname	= "ip_default_ttl",
		.data		= &init_net.ipv4.sysctl_ip_default_ttl,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &ip_ttl_min,
		.extra2		= &ip_ttl_max,
	},
714 715
	{
		.procname	= "ip_local_port_range",
716 717
		.maxlen		= sizeof(init_net.ipv4.ip_local_ports.range),
		.data		= &init_net.ipv4.ip_local_ports.range,
718 719 720
		.mode		= 0644,
		.proc_handler	= ipv4_local_port_range,
	},
721 722 723 724 725 726 727
	{
		.procname	= "ip_local_reserved_ports",
		.data		= &init_net.ipv4.sysctl_local_reserved_ports,
		.maxlen		= 65536,
		.mode		= 0644,
		.proc_handler	= proc_do_large_bitmap,
	},
728 729 730 731 732 733 734
	{
		.procname	= "ip_no_pmtu_disc",
		.data		= &init_net.ipv4.sysctl_ip_no_pmtu_disc,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
735 736 737 738 739 740 741
	{
		.procname	= "ip_forward_use_pmtu",
		.data		= &init_net.ipv4.sysctl_ip_fwd_use_pmtu,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
742 743 744 745 746
	{
		.procname	= "ip_forward_update_priority",
		.data		= &init_net.ipv4.sysctl_ip_fwd_update_priority,
		.maxlen		= sizeof(int),
		.mode		= 0644,
747
		.proc_handler   = ipv4_fwd_update_priority,
748 749
		.extra1		= SYSCTL_ZERO,
		.extra2		= SYSCTL_ONE,
750
	},
751 752 753 754 755 756 757
	{
		.procname	= "ip_nonlocal_bind",
		.data		= &init_net.ipv4.sysctl_ip_nonlocal_bind,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
758 759 760 761 762 763 764 765 766
	{
		.procname	= "ip_autobind_reuse",
		.data		= &init_net.ipv4.sysctl_ip_autobind_reuse,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1         = SYSCTL_ZERO,
		.extra2         = SYSCTL_ONE,
	},
767 768 769 770 771 772 773
	{
		.procname	= "fwmark_reflect",
		.data		= &init_net.ipv4.sysctl_fwmark_reflect,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
774 775 776 777 778 779 780
	{
		.procname	= "tcp_fwmark_accept",
		.data		= &init_net.ipv4.sysctl_tcp_fwmark_accept,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
781 782 783 784 785 786 787
#ifdef CONFIG_NET_L3_MASTER_DEV
	{
		.procname	= "tcp_l3mdev_accept",
		.data		= &init_net.ipv4.sysctl_tcp_l3mdev_accept,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
788 789
		.extra1		= SYSCTL_ZERO,
		.extra2		= SYSCTL_ONE,
790 791
	},
#endif
F
Fan Du 已提交
792 793 794 795 796 797 798 799 800 801 802 803 804 805
	{
		.procname	= "tcp_mtu_probing",
		.data		= &init_net.ipv4.sysctl_tcp_mtu_probing,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
	{
		.procname	= "tcp_base_mss",
		.data		= &init_net.ipv4.sysctl_tcp_base_mss,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
E
Eric Dumazet 已提交
806 807 808 809 810 811 812 813 814
	{
		.procname	= "tcp_min_snd_mss",
		.data		= &init_net.ipv4.sysctl_tcp_min_snd_mss,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &tcp_min_snd_mss_min,
		.extra2		= &tcp_min_snd_mss_max,
	},
815 816 817 818 819 820 821 822 823
	{
		.procname	= "tcp_mtu_probe_floor",
		.data		= &init_net.ipv4.sysctl_tcp_mtu_probe_floor,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &tcp_min_snd_mss_min,
		.extra2		= &tcp_min_snd_mss_max,
	},
824 825 826 827 828 829 830
	{
		.procname	= "tcp_probe_threshold",
		.data		= &init_net.ipv4.sysctl_tcp_probe_threshold,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
831 832 833
	{
		.procname	= "tcp_probe_interval",
		.data		= &init_net.ipv4.sysctl_tcp_probe_interval,
834
		.maxlen		= sizeof(u32),
835
		.mode		= 0644,
836 837
		.proc_handler	= proc_douintvec_minmax,
		.extra2		= &u32_max_div_HZ,
838
	},
839 840
	{
		.procname	= "igmp_link_local_mcast_reports",
841
		.data		= &init_net.ipv4.sysctl_igmp_llm_reports,
842 843 844 845
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
846 847 848 849 850 851 852
	{
		.procname	= "igmp_max_memberships",
		.data		= &init_net.ipv4.sysctl_igmp_max_memberships,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
853 854 855 856 857 858 859
	{
		.procname	= "igmp_max_msf",
		.data		= &init_net.ipv4.sysctl_igmp_max_msf,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
860 861 862 863 864 865 866
#ifdef CONFIG_IP_MULTICAST
	{
		.procname	= "igmp_qrv",
		.data		= &init_net.ipv4.sysctl_igmp_qrv,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
867
		.extra1		= SYSCTL_ONE
868 869
	},
#endif
870 871 872 873 874 875 876
	{
		.procname	= "tcp_congestion_control",
		.data		= &init_net.ipv4.tcp_congestion_control,
		.mode		= 0644,
		.maxlen		= TCP_CA_NAME_MAX,
		.proc_handler	= proc_tcp_congestion_control,
	},
877 878 879 880 881 882 883 884 885 886 887 888
	{
		.procname	= "tcp_available_congestion_control",
		.maxlen		= TCP_CA_BUF_MAX,
		.mode		= 0444,
		.proc_handler   = proc_tcp_available_congestion_control,
	},
	{
		.procname	= "tcp_allowed_congestion_control",
		.maxlen		= TCP_CA_BUF_MAX,
		.mode		= 0644,
		.proc_handler   = proc_allowed_congestion_control,
	},
889 890 891 892 893 894 895
	{
		.procname	= "tcp_keepalive_time",
		.data		= &init_net.ipv4.sysctl_tcp_keepalive_time,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_jiffies,
	},
896 897 898 899 900 901 902
	{
		.procname	= "tcp_keepalive_probes",
		.data		= &init_net.ipv4.sysctl_tcp_keepalive_probes,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
903 904 905 906 907 908 909
	{
		.procname	= "tcp_keepalive_intvl",
		.data		= &init_net.ipv4.sysctl_tcp_keepalive_intvl,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_jiffies,
	},
910 911 912 913 914 915 916 917 918
	{
		.procname	= "tcp_syn_retries",
		.data		= &init_net.ipv4.sysctl_tcp_syn_retries,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &tcp_syn_retries_min,
		.extra2		= &tcp_syn_retries_max
	},
919 920 921 922 923 924 925
	{
		.procname	= "tcp_synack_retries",
		.data		= &init_net.ipv4.sysctl_tcp_synack_retries,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
926 927 928 929 930 931 932 933 934
#ifdef CONFIG_SYN_COOKIES
	{
		.procname	= "tcp_syncookies",
		.data		= &init_net.ipv4.sysctl_tcp_syncookies,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
#endif
935 936 937 938 939 940 941
	{
		.procname	= "tcp_reordering",
		.data		= &init_net.ipv4.sysctl_tcp_reordering,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
942 943 944 945 946 947 948 949
	{
		.procname	= "tcp_retries1",
		.data		= &init_net.ipv4.sysctl_tcp_retries1,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra2		= &tcp_retr1_max
	},
950 951 952 953 954 955 956
	{
		.procname	= "tcp_retries2",
		.data		= &init_net.ipv4.sysctl_tcp_retries2,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
957 958 959 960 961 962 963
	{
		.procname	= "tcp_orphan_retries",
		.data		= &init_net.ipv4.sysctl_tcp_orphan_retries,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
964 965 966 967 968 969 970
	{
		.procname	= "tcp_fin_timeout",
		.data		= &init_net.ipv4.sysctl_tcp_fin_timeout,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_jiffies,
	},
971 972 973 974 975
	{
		.procname	= "tcp_notsent_lowat",
		.data		= &init_net.ipv4.sysctl_tcp_notsent_lowat,
		.maxlen		= sizeof(unsigned int),
		.mode		= 0644,
976
		.proc_handler	= proc_douintvec,
977
	},
978 979 980 981 982
	{
		.procname	= "tcp_tw_reuse",
		.data		= &init_net.ipv4.sysctl_tcp_tw_reuse,
		.maxlen		= sizeof(int),
		.mode		= 0644,
983
		.proc_handler	= proc_dointvec_minmax,
984
		.extra1		= SYSCTL_ZERO,
985
		.extra2		= &two,
986
	},
987 988 989 990 991 992 993
	{
		.procname	= "tcp_max_tw_buckets",
		.data		= &init_net.ipv4.tcp_death_row.sysctl_max_tw_buckets,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
994 995 996 997 998 999 1000
	{
		.procname	= "tcp_max_syn_backlog",
		.data		= &init_net.ipv4.sysctl_max_syn_backlog,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1001 1002 1003 1004 1005 1006 1007
	{
		.procname	= "tcp_fastopen",
		.data		= &init_net.ipv4.sysctl_tcp_fastopen,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
1008 1009 1010 1011
	{
		.procname	= "tcp_fastopen_key",
		.mode		= 0600,
		.data		= &init_net.ipv4.sysctl_tcp_fastopen,
1012 1013 1014 1015 1016 1017
		/* maxlen to print the list of keys in hex (*2), with dashes
		 * separating doublewords and a comma in between keys.
		 */
		.maxlen		= ((TCP_FASTOPEN_KEY_LENGTH *
				   2 * TCP_FASTOPEN_KEY_MAX) +
				   (TCP_FASTOPEN_KEY_MAX * 5)),
1018 1019
		.proc_handler	= proc_tcp_fastopen_key,
	},
1020 1021 1022 1023 1024 1025
	{
		.procname	= "tcp_fastopen_blackhole_timeout_sec",
		.data		= &init_net.ipv4.sysctl_tcp_fastopen_blackhole_timeout,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_tfo_blackhole_detect_timeout,
1026
		.extra1		= SYSCTL_ZERO,
1027
	},
1028 1029 1030 1031 1032 1033 1034
#ifdef CONFIG_IP_ROUTE_MULTIPATH
	{
		.procname	= "fib_multipath_use_neigh",
		.data		= &init_net.ipv4.sysctl_fib_multipath_use_neigh,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
1035 1036
		.extra1		= SYSCTL_ZERO,
		.extra2		= SYSCTL_ONE,
1037 1038 1039 1040 1041 1042
	},
	{
		.procname	= "fib_multipath_hash_policy",
		.data		= &init_net.ipv4.sysctl_fib_multipath_hash_policy,
		.maxlen		= sizeof(int),
		.mode		= 0644,
1043
		.proc_handler	= proc_fib_multipath_hash_policy,
1044
		.extra1		= SYSCTL_ZERO,
1045
		.extra2		= &two,
1046 1047
	},
#endif
1048 1049 1050 1051 1052 1053 1054
	{
		.procname	= "ip_unprivileged_port_start",
		.maxlen		= sizeof(int),
		.data		= &init_net.ipv4.sysctl_ip_prot_sock,
		.mode		= 0644,
		.proc_handler	= ipv4_privileged_ports,
	},
1055 1056 1057 1058 1059 1060 1061
#ifdef CONFIG_NET_L3_MASTER_DEV
	{
		.procname	= "udp_l3mdev_accept",
		.data		= &init_net.ipv4.sysctl_udp_l3mdev_accept,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
1062 1063
		.extra1		= SYSCTL_ZERO,
		.extra2		= SYSCTL_ONE,
1064 1065
	},
#endif
E
Eric Dumazet 已提交
1066 1067 1068 1069 1070 1071 1072
	{
		.procname	= "tcp_sack",
		.data		= &init_net.ipv4.sysctl_tcp_sack,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1073 1074 1075 1076 1077 1078 1079
	{
		.procname	= "tcp_window_scaling",
		.data		= &init_net.ipv4.sysctl_tcp_window_scaling,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1080 1081 1082 1083 1084 1085 1086
	{
		.procname	= "tcp_timestamps",
		.data		= &init_net.ipv4.sysctl_tcp_timestamps,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1087 1088 1089 1090 1091 1092
	{
		.procname	= "tcp_early_retrans",
		.data		= &init_net.ipv4.sysctl_tcp_early_retrans,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
1093
		.extra1		= SYSCTL_ZERO,
1094 1095
		.extra2		= &four,
	},
1096 1097 1098 1099 1100 1101 1102
	{
		.procname	= "tcp_recovery",
		.data		= &init_net.ipv4.sysctl_tcp_recovery,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
1103 1104 1105 1106 1107 1108 1109
	{
		.procname       = "tcp_thin_linear_timeouts",
		.data           = &init_net.ipv4.sysctl_tcp_thin_linear_timeouts,
		.maxlen         = sizeof(int),
		.mode           = 0644,
		.proc_handler   = proc_dointvec
	},
1110 1111 1112 1113 1114 1115 1116
	{
		.procname	= "tcp_slow_start_after_idle",
		.data		= &init_net.ipv4.sysctl_tcp_slow_start_after_idle,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1117 1118 1119 1120 1121 1122 1123
	{
		.procname	= "tcp_retrans_collapse",
		.data		= &init_net.ipv4.sysctl_tcp_retrans_collapse,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1124 1125 1126 1127 1128 1129 1130
	{
		.procname	= "tcp_stdurg",
		.data		= &init_net.ipv4.sysctl_tcp_stdurg,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1131 1132 1133 1134 1135 1136 1137
	{
		.procname	= "tcp_rfc1337",
		.data		= &init_net.ipv4.sysctl_tcp_rfc1337,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1138 1139 1140 1141 1142 1143 1144
	{
		.procname	= "tcp_abort_on_overflow",
		.data		= &init_net.ipv4.sysctl_tcp_abort_on_overflow,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
E
Eric Dumazet 已提交
1145 1146 1147 1148 1149 1150 1151
	{
		.procname	= "tcp_fack",
		.data		= &init_net.ipv4.sysctl_tcp_fack,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1152 1153 1154 1155 1156 1157 1158
	{
		.procname	= "tcp_max_reordering",
		.data		= &init_net.ipv4.sysctl_tcp_max_reordering,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
E
Eric Dumazet 已提交
1159 1160 1161 1162 1163 1164 1165
	{
		.procname	= "tcp_dsack",
		.data		= &init_net.ipv4.sysctl_tcp_dsack,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1166 1167 1168 1169 1170 1171 1172
	{
		.procname	= "tcp_app_win",
		.data		= &init_net.ipv4.sysctl_tcp_app_win,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1173 1174 1175 1176 1177 1178 1179 1180 1181
	{
		.procname	= "tcp_adv_win_scale",
		.data		= &init_net.ipv4.sysctl_tcp_adv_win_scale,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &tcp_adv_win_scale_min,
		.extra2		= &tcp_adv_win_scale_max,
	},
E
Eric Dumazet 已提交
1182 1183 1184 1185 1186 1187 1188
	{
		.procname	= "tcp_frto",
		.data		= &init_net.ipv4.sysctl_tcp_frto,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1189 1190 1191 1192 1193 1194 1195
	{
		.procname	= "tcp_no_metrics_save",
		.data		= &init_net.ipv4.sysctl_tcp_nometrics_save,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
1196 1197 1198 1199 1200 1201 1202 1203 1204
	{
		.procname	= "tcp_no_ssthresh_metrics_save",
		.data		= &init_net.ipv4.sysctl_tcp_no_ssthresh_metrics_save,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= SYSCTL_ZERO,
		.extra2		= SYSCTL_ONE,
	},
1205 1206 1207 1208 1209 1210 1211
	{
		.procname	= "tcp_moderate_rcvbuf",
		.data		= &init_net.ipv4.sysctl_tcp_moderate_rcvbuf,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
1212 1213 1214 1215 1216 1217 1218
	{
		.procname	= "tcp_tso_win_divisor",
		.data		= &init_net.ipv4.sysctl_tcp_tso_win_divisor,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
1219 1220 1221 1222 1223 1224 1225
	{
		.procname	= "tcp_workaround_signed_windows",
		.data		= &init_net.ipv4.sysctl_tcp_workaround_signed_windows,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1226 1227 1228 1229 1230 1231 1232
	{
		.procname	= "tcp_limit_output_bytes",
		.data		= &init_net.ipv4.sysctl_tcp_limit_output_bytes,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1233 1234 1235 1236 1237 1238 1239
	{
		.procname	= "tcp_challenge_ack_limit",
		.data		= &init_net.ipv4.sysctl_tcp_challenge_ack_limit,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1240 1241 1242 1243 1244 1245
	{
		.procname	= "tcp_min_tso_segs",
		.data		= &init_net.ipv4.sysctl_tcp_min_tso_segs,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
1246
		.extra1		= SYSCTL_ONE,
1247 1248
		.extra2		= &gso_max_segs,
	},
1249 1250 1251 1252 1253
	{
		.procname	= "tcp_min_rtt_wlen",
		.data		= &init_net.ipv4.sysctl_tcp_min_rtt_wlen,
		.maxlen		= sizeof(int),
		.mode		= 0644,
1254
		.proc_handler	= proc_dointvec_minmax,
1255
		.extra1		= SYSCTL_ZERO,
1256
		.extra2		= &one_day_secs
1257
	},
1258 1259 1260 1261 1262 1263
	{
		.procname	= "tcp_autocorking",
		.data		= &init_net.ipv4.sysctl_tcp_autocorking,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
1264 1265
		.extra1		= SYSCTL_ZERO,
		.extra2		= SYSCTL_ONE,
1266
	},
1267 1268 1269 1270 1271 1272 1273
	{
		.procname	= "tcp_invalid_ratelimit",
		.data		= &init_net.ipv4.sysctl_tcp_invalid_ratelimit,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_ms_jiffies,
	},
1274 1275 1276 1277 1278 1279
	{
		.procname	= "tcp_pacing_ss_ratio",
		.data		= &init_net.ipv4.sysctl_tcp_pacing_ss_ratio,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
1280
		.extra1		= SYSCTL_ZERO,
1281 1282
		.extra2		= &thousand,
	},
1283 1284 1285 1286 1287 1288
	{
		.procname	= "tcp_pacing_ca_ratio",
		.data		= &init_net.ipv4.sysctl_tcp_pacing_ca_ratio,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
1289
		.extra1		= SYSCTL_ZERO,
1290 1291
		.extra2		= &thousand,
	},
1292 1293 1294 1295 1296 1297
	{
		.procname	= "tcp_wmem",
		.data		= &init_net.ipv4.sysctl_tcp_wmem,
		.maxlen		= sizeof(init_net.ipv4.sysctl_tcp_wmem),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
1298
		.extra1		= SYSCTL_ONE,
1299 1300 1301 1302 1303 1304 1305
	},
	{
		.procname	= "tcp_rmem",
		.data		= &init_net.ipv4.sysctl_tcp_rmem,
		.maxlen		= sizeof(init_net.ipv4.sysctl_tcp_rmem),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
1306
		.extra1		= SYSCTL_ONE,
1307
	},
1308 1309 1310 1311 1312 1313 1314
	{
		.procname	= "tcp_comp_sack_delay_ns",
		.data		= &init_net.ipv4.sysctl_tcp_comp_sack_delay_ns,
		.maxlen		= sizeof(unsigned long),
		.mode		= 0644,
		.proc_handler	= proc_doulongvec_minmax,
	},
E
Eric Dumazet 已提交
1315 1316 1317 1318 1319 1320
	{
		.procname	= "tcp_comp_sack_nr",
		.data		= &init_net.ipv4.sysctl_tcp_comp_sack_nr,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
1321
		.extra1		= SYSCTL_ZERO,
E
Eric Dumazet 已提交
1322 1323
		.extra2		= &comp_sack_nr_max,
	},
1324 1325 1326 1327 1328 1329
	{
		.procname	= "udp_rmem_min",
		.data		= &init_net.ipv4.sysctl_udp_rmem_min,
		.maxlen		= sizeof(init_net.ipv4.sysctl_udp_rmem_min),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
1330
		.extra1		= SYSCTL_ONE
1331 1332 1333 1334 1335 1336 1337
	},
	{
		.procname	= "udp_wmem_min",
		.data		= &init_net.ipv4.sysctl_udp_wmem_min,
		.maxlen		= sizeof(init_net.ipv4.sysctl_udp_wmem_min),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
1338
		.extra1		= SYSCTL_ONE
1339
	},
1340 1341 1342
	{ }
};

1343 1344
static __net_init int ipv4_sysctl_init_net(struct net *net)
{
1345 1346 1347
	struct ctl_table *table;

	table = ipv4_net_table;
O
Octavian Purdila 已提交
1348
	if (!net_eq(net, &init_net)) {
1349 1350
		int i;

1351
		table = kmemdup(table, sizeof(ipv4_net_table), GFP_KERNEL);
1352
		if (!table)
1353 1354
			goto err_alloc;

1355 1356 1357
		/* Update the variables to point into the current struct net */
		for (i = 0; i < ARRAY_SIZE(ipv4_net_table) - 1; i++)
			table[i].data += (void *)net - (void *)&init_net;
1358 1359
	}

1360
	net->ipv4.ipv4_hdr = register_net_sysctl(net, "net/ipv4", table);
1361
	if (!net->ipv4.ipv4_hdr)
1362 1363
		goto err_reg;

1364 1365 1366 1367
	net->ipv4.sysctl_local_reserved_ports = kzalloc(65536 / 8, GFP_KERNEL);
	if (!net->ipv4.sysctl_local_reserved_ports)
		goto err_ports;

1368
	return 0;
1369

1370 1371
err_ports:
	unregister_net_sysctl_table(net->ipv4.ipv4_hdr);
1372
err_reg:
O
Octavian Purdila 已提交
1373
	if (!net_eq(net, &init_net))
1374 1375 1376
		kfree(table);
err_alloc:
	return -ENOMEM;
1377 1378 1379 1380
}

static __net_exit void ipv4_sysctl_exit_net(struct net *net)
{
1381 1382
	struct ctl_table *table;

1383
	kfree(net->ipv4.sysctl_local_reserved_ports);
1384 1385 1386
	table = net->ipv4.ipv4_hdr->ctl_table_arg;
	unregister_net_sysctl_table(net->ipv4.ipv4_hdr);
	kfree(table);
1387 1388 1389 1390 1391 1392 1393
}

static __net_initdata struct pernet_operations ipv4_sysctl_ops = {
	.init = ipv4_sysctl_init_net,
	.exit = ipv4_sysctl_exit_net,
};

1394 1395 1396 1397
static __init int sysctl_ipv4_init(void)
{
	struct ctl_table_header *hdr;

1398
	hdr = register_net_sysctl(&init_net, "net/ipv4", ipv4_table);
1399
	if (!hdr)
1400 1401 1402
		return -ENOMEM;

	if (register_pernet_subsys(&ipv4_sysctl_ops)) {
1403
		unregister_net_sysctl_table(hdr);
1404 1405 1406 1407
		return -ENOMEM;
	}

	return 0;
1408 1409 1410
}

__initcall(sysctl_ipv4_init);