sysctl_net_ipv4.c 29.2 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10
/*
 * 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>
11
#include <linux/igmp.h>
12
#include <linux/inetdevice.h>
13
#include <linux/seqlock.h>
14
#include <linux/init.h>
15
#include <linux/slab.h>
16
#include <linux/nsproxy.h>
G
Glauber Costa 已提交
17
#include <linux/swap.h>
L
Linus Torvalds 已提交
18
#include <net/snmp.h>
19
#include <net/icmp.h>
L
Linus Torvalds 已提交
20 21 22
#include <net/ip.h>
#include <net/route.h>
#include <net/tcp.h>
H
Hideo Aoki 已提交
23
#include <net/udp.h>
P
Paul Moore 已提交
24
#include <net/cipso_ipv4.h>
25
#include <net/inet_frag.h>
26
#include <net/ping.h>
27
#include <net/protocol.h>
L
Linus Torvalds 已提交
28

H
Herbert Xu 已提交
29
static int zero;
30
static int one = 1;
N
Nandita Dukkipati 已提交
31
static int four = 4;
32
static int thousand = 1000;
E
Eric Dumazet 已提交
33
static int gso_max_segs = GSO_MAX_SEGS;
34
static int tcp_retr1_max = 255;
L
Linus Torvalds 已提交
35 36
static int ip_local_port_range_min[] = { 1, 1 };
static int ip_local_port_range_max[] = { 65535, 65535 };
37 38
static int tcp_adv_win_scale_min = -31;
static int tcp_adv_win_scale_max = 31;
39 40
static int ip_privileged_port_min;
static int ip_privileged_port_max = 65535;
E
Eric Dumazet 已提交
41 42
static int ip_ttl_min = 1;
static int ip_ttl_max = 255;
43 44
static int tcp_syn_retries_min = 1;
static int tcp_syn_retries_max = MAX_TCP_SYNCNT;
45 46
static int ip_ping_group_range_min[] = { 0, 0 };
static int ip_ping_group_range_max[] = { GID_T_MAX, GID_T_MAX };
L
Linus Torvalds 已提交
47

48 49 50
/* obsolete */
static int sysctl_tcp_low_latency __read_mostly;

51
/* Update system visible IP port range */
52
static void set_local_port_range(struct net *net, int range[2])
53
{
54 55
	bool same_parity = !((range[0] ^ range[1]) & 1);

56
	write_seqlock_bh(&net->ipv4.ip_local_ports.lock);
57 58 59 60
	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");
	}
61 62
	net->ipv4.ip_local_ports.range[0] = range[0];
	net->ipv4.ip_local_ports.range[1] = range[1];
63
	write_sequnlock_bh(&net->ipv4.ip_local_ports.lock);
64 65 66
}

/* Validate changes from /proc interface. */
67
static int ipv4_local_port_range(struct ctl_table *table, int write,
68 69 70
				 void __user *buffer,
				 size_t *lenp, loff_t *ppos)
{
71
	struct net *net =
72
		container_of(table->data, struct net, ipv4.ip_local_ports.range);
73
	int ret;
E
Eric Dumazet 已提交
74
	int range[2];
75
	struct ctl_table tmp = {
76 77 78 79 80 81 82
		.data = &range,
		.maxlen = sizeof(range),
		.mode = table->mode,
		.extra1 = &ip_local_port_range_min,
		.extra2 = &ip_local_port_range_max,
	};

83 84
	inet_get_local_port_range(net, &range[0], &range[1]);

85
	ret = proc_dointvec_minmax(&tmp, write, buffer, lenp, ppos);
86 87

	if (write && ret == 0) {
88 89 90 91 92 93
		/* 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))
94 95
			ret = -EINVAL;
		else
96
			set_local_port_range(net, range);
97 98 99 100 101
	}

	return ret;
}

102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135
/* Validate changes from /proc interface. */
static int ipv4_privileged_ports(struct ctl_table *table, int write,
				void __user *buffer, size_t *lenp, loff_t *ppos)
{
	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;
}
136

137
static void inet_get_ping_group_range_table(struct ctl_table *table, kgid_t *low, kgid_t *high)
138
{
139
	kgid_t *data = table->data;
140
	struct net *net =
141
		container_of(table->data, struct net, ipv4.ping_group_range.range);
142
	unsigned int seq;
143
	do {
144
		seq = read_seqbegin(&net->ipv4.ping_group_range.lock);
145 146 147

		*low = data[0];
		*high = data[1];
148
	} while (read_seqretry(&net->ipv4.ping_group_range.lock, seq));
149 150 151
}

/* Update system visible IP port range */
152
static void set_ping_group_range(struct ctl_table *table, kgid_t low, kgid_t high)
153
{
154
	kgid_t *data = table->data;
155
	struct net *net =
156
		container_of(table->data, struct net, ipv4.ping_group_range.range);
157
	write_seqlock(&net->ipv4.ping_group_range.lock);
158 159
	data[0] = low;
	data[1] = high;
160
	write_sequnlock(&net->ipv4.ping_group_range.lock);
161 162 163
}

/* Validate changes from /proc interface. */
164
static int ipv4_ping_group_range(struct ctl_table *table, int write,
165 166 167
				 void __user *buffer,
				 size_t *lenp, loff_t *ppos)
{
168
	struct user_namespace *user_ns = current_user_ns();
169
	int ret;
170 171
	gid_t urange[2];
	kgid_t low, high;
172
	struct ctl_table tmp = {
173 174
		.data = &urange,
		.maxlen = sizeof(urange),
175 176 177 178 179
		.mode = table->mode,
		.extra1 = &ip_ping_group_range_min,
		.extra2 = &ip_ping_group_range_max,
	};

180 181 182
	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);
183 184
	ret = proc_dointvec_minmax(&tmp, write, buffer, lenp, ppos);

185 186 187 188 189 190 191 192 193 194
	if (write && ret == 0) {
		low = make_kgid(user_ns, urange[0]);
		high = make_kgid(user_ns, urange[1]);
		if (!gid_valid(low) || !gid_valid(high) ||
		    (urange[1] < urange[0]) || gid_lt(high, low)) {
			low = make_kgid(&init_user_ns, 1);
			high = make_kgid(&init_user_ns, 0);
		}
		set_ping_group_range(table, low, high);
	}
195 196 197 198

	return ret;
}

199
static int proc_tcp_congestion_control(struct ctl_table *ctl, int write,
200 201 202
				       void __user *buffer, size_t *lenp, loff_t *ppos)
{
	char val[TCP_CA_NAME_MAX];
203
	struct ctl_table tbl = {
204 205 206 207 208 209 210
		.data = val,
		.maxlen = TCP_CA_NAME_MAX,
	};
	int ret;

	tcp_get_default_congestion_control(val);

211
	ret = proc_dostring(&tbl, write, buffer, lenp, ppos);
212 213 214 215 216
	if (write && ret == 0)
		ret = tcp_set_default_congestion_control(val);
	return ret;
}

217
static int proc_tcp_available_congestion_control(struct ctl_table *ctl,
218
						 int write,
219 220 221
						 void __user *buffer, size_t *lenp,
						 loff_t *ppos)
{
222
	struct ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX, };
223 224 225 226 227 228
	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);
229
	ret = proc_dostring(&tbl, write, buffer, lenp, ppos);
230 231 232 233
	kfree(tbl.data);
	return ret;
}

234
static int proc_allowed_congestion_control(struct ctl_table *ctl,
235
					   int write,
236 237 238
					   void __user *buffer, size_t *lenp,
					   loff_t *ppos)
{
239
	struct ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX };
240 241 242 243 244 245 246
	int ret;

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

	tcp_get_allowed_congestion_control(tbl.data, tbl.maxlen);
247
	ret = proc_dostring(&tbl, write, buffer, lenp, ppos);
248 249 250 251 252 253
	if (write && ret == 0)
		ret = tcp_set_allowed_congestion_control(tbl.data);
	kfree(tbl.data);
	return ret;
}

254 255 256
static int proc_tcp_fastopen_key(struct ctl_table *ctl, int write,
				 void __user *buffer, size_t *lenp,
				 loff_t *ppos)
257
{
258
	struct ctl_table tbl = { .maxlen = (TCP_FASTOPEN_KEY_LENGTH * 2 + 10) };
259 260 261 262 263 264 265 266 267 268 269 270
	struct tcp_fastopen_context *ctxt;
	int ret;
	u32  user_key[4]; /* 16 bytes, matching TCP_FASTOPEN_KEY_LENGTH */

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

	rcu_read_lock();
	ctxt = rcu_dereference(tcp_fastopen_ctx);
	if (ctxt)
		memcpy(user_key, ctxt->key, TCP_FASTOPEN_KEY_LENGTH);
271 272
	else
		memset(user_key, 0, sizeof(user_key));
273 274 275 276 277 278 279 280 281 282 283 284
	rcu_read_unlock();

	snprintf(tbl.data, tbl.maxlen, "%08x-%08x-%08x-%08x",
		user_key[0], user_key[1], user_key[2], user_key[3]);
	ret = proc_dostring(&tbl, write, buffer, lenp, ppos);

	if (write && ret == 0) {
		if (sscanf(tbl.data, "%x-%x-%x-%x", user_key, user_key + 1,
			   user_key + 2, user_key + 3) != 4) {
			ret = -EINVAL;
			goto bad_key;
		}
285 286 287 288 289
		/* Generate a dummy secret but don't publish it. This
		 * is needed so we don't regenerate a new key on the
		 * first invocation of tcp_fastopen_cookie_gen
		 */
		tcp_fastopen_init_key_once(false);
290 291 292 293 294 295 296 297 298 299 300
		tcp_fastopen_reset_cipher(user_key, TCP_FASTOPEN_KEY_LENGTH);
	}

bad_key:
	pr_debug("proc FO key set 0x%x-%x-%x-%x <- 0x%s: %u\n",
	       user_key[0], user_key[1], user_key[2], user_key[3],
	       (char *)tbl.data, ret);
	kfree(tbl.data);
	return ret;
}

301 302 303 304 305 306 307
static void proc_configure_early_demux(int enabled, int protocol)
{
	struct net_protocol *ipprot;
#if IS_ENABLED(CONFIG_IPV6)
	struct inet6_protocol *ip6prot;
#endif

308 309
	rcu_read_lock();

310 311 312 313 314 315 316 317 318 319 320
	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
321
	rcu_read_unlock();
322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355
}

static int proc_tcp_early_demux(struct ctl_table *table, int write,
				void __user *buffer, size_t *lenp, loff_t *ppos)
{
	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,
				void __user *buffer, size_t *lenp, loff_t *ppos)
{
	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;
}

356 357 358 359 360 361 362 363 364 365
static int proc_tfo_blackhole_detect_timeout(struct ctl_table *table,
					     int write,
					     void __user *buffer,
					     size_t *lenp, loff_t *ppos)
{
	int ret;

	ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
	if (write && ret == 0)
		tcp_fastopen_active_timeout_reset();
D
Dave Watson 已提交
366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384

	return ret;
}

static int proc_tcp_available_ulp(struct ctl_table *ctl,
				  int write,
				  void __user *buffer, size_t *lenp,
				  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);

385 386 387
	return ret;
}

388
static struct ctl_table ipv4_table[] = {
389
	{
L
Linus Torvalds 已提交
390 391 392 393
		.procname	= "tcp_retrans_collapse",
		.data		= &sysctl_tcp_retrans_collapse,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
394
		.proc_handler	= proc_dointvec
L
Linus Torvalds 已提交
395 396 397 398 399 400
	},
	{
		.procname	= "tcp_max_orphans",
		.data		= &sysctl_tcp_max_orphans,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
401
		.proc_handler	= proc_dointvec
L
Linus Torvalds 已提交
402
	},
Y
Yuchung Cheng 已提交
403 404 405 406 407 408 409
	{
		.procname	= "tcp_fastopen",
		.data		= &sysctl_tcp_fastopen,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
410 411 412 413 414 415
	{
		.procname	= "tcp_fastopen_key",
		.mode		= 0600,
		.maxlen		= ((TCP_FASTOPEN_KEY_LENGTH * 2) + 10),
		.proc_handler	= proc_tcp_fastopen_key,
	},
416 417 418 419 420 421 422 423
	{
		.procname	= "tcp_fastopen_blackhole_timeout_sec",
		.data		= &sysctl_tcp_fastopen_blackhole_timeout,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_tfo_blackhole_detect_timeout,
		.extra1		= &zero,
	},
L
Linus Torvalds 已提交
424 425 426 427 428
	{
		.procname	= "tcp_abort_on_overflow",
		.data		= &sysctl_tcp_abort_on_overflow,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
429
		.proc_handler	= proc_dointvec
L
Linus Torvalds 已提交
430 431 432 433 434 435
	},
	{
		.procname	= "tcp_stdurg",
		.data		= &sysctl_tcp_stdurg,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
436
		.proc_handler	= proc_dointvec
L
Linus Torvalds 已提交
437 438 439 440 441 442
	},
	{
		.procname	= "tcp_rfc1337",
		.data		= &sysctl_tcp_rfc1337,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
443
		.proc_handler	= proc_dointvec
L
Linus Torvalds 已提交
444 445 446 447 448 449
	},
	{
		.procname	= "inet_peer_threshold",
		.data		= &inet_peer_threshold,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
450
		.proc_handler	= proc_dointvec
L
Linus Torvalds 已提交
451 452 453 454 455 456
	},
	{
		.procname	= "inet_peer_minttl",
		.data		= &inet_peer_minttl,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
457
		.proc_handler	= proc_dointvec_jiffies,
L
Linus Torvalds 已提交
458 459 460 461 462 463
	},
	{
		.procname	= "inet_peer_maxttl",
		.data		= &inet_peer_maxttl,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
464
		.proc_handler	= proc_dointvec_jiffies,
L
Linus Torvalds 已提交
465 466 467 468 469 470
	},
	{
		.procname	= "tcp_fack",
		.data		= &sysctl_tcp_fack,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
471
		.proc_handler	= proc_dointvec
L
Linus Torvalds 已提交
472
	},
Y
Yuchung Cheng 已提交
473 474 475 476 477 478 479
	{
		.procname	= "tcp_recovery",
		.data		= &sysctl_tcp_recovery,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
480 481 482 483 484 485 486
	{
		.procname	= "tcp_max_reordering",
		.data		= &sysctl_tcp_max_reordering,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
L
Linus Torvalds 已提交
487 488 489 490 491
	{
		.procname	= "tcp_dsack",
		.data		= &sysctl_tcp_dsack,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
492
		.proc_handler	= proc_dointvec
L
Linus Torvalds 已提交
493
	},
494 495 496 497 498 499 500
	{
		.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
501 502 503 504 505
	{
		.procname	= "tcp_wmem",
		.data		= &sysctl_tcp_wmem,
		.maxlen		= sizeof(sysctl_tcp_wmem),
		.mode		= 0644,
506
		.proc_handler	= proc_dointvec_minmax,
507
		.extra1		= &one,
L
Linus Torvalds 已提交
508 509 510 511 512 513
	},
	{
		.procname	= "tcp_rmem",
		.data		= &sysctl_tcp_rmem,
		.maxlen		= sizeof(sysctl_tcp_rmem),
		.mode		= 0644,
514
		.proc_handler	= proc_dointvec_minmax,
515
		.extra1		= &one,
L
Linus Torvalds 已提交
516 517 518 519 520 521
	},
	{
		.procname	= "tcp_app_win",
		.data		= &sysctl_tcp_app_win,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
522
		.proc_handler	= proc_dointvec
L
Linus Torvalds 已提交
523 524 525 526 527 528
	},
	{
		.procname	= "tcp_adv_win_scale",
		.data		= &sysctl_tcp_adv_win_scale,
		.maxlen		= sizeof(int),
		.mode		= 0644,
529 530 531
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &tcp_adv_win_scale_min,
		.extra2		= &tcp_adv_win_scale_max,
L
Linus Torvalds 已提交
532 533 534 535 536
	},
	{
		.procname	= "tcp_frto",
		.data		= &sysctl_tcp_frto,
		.maxlen		= sizeof(int),
537 538 539 540 541 542 543
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
	{
		.procname	= "tcp_min_rtt_wlen",
		.data		= &sysctl_tcp_min_rtt_wlen,
		.maxlen		= sizeof(int),
L
Linus Torvalds 已提交
544
		.mode		= 0644,
A
Alexey Dobriyan 已提交
545
		.proc_handler	= proc_dointvec
L
Linus Torvalds 已提交
546 547 548 549 550 551
	},
	{
		.procname	= "tcp_low_latency",
		.data		= &sysctl_tcp_low_latency,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
552
		.proc_handler	= proc_dointvec
L
Linus Torvalds 已提交
553 554 555 556 557 558
	},
	{
		.procname	= "tcp_no_metrics_save",
		.data		= &sysctl_tcp_nometrics_save,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
559
		.proc_handler	= proc_dointvec,
L
Linus Torvalds 已提交
560 561 562 563 564 565
	},
	{
		.procname	= "tcp_moderate_rcvbuf",
		.data		= &sysctl_tcp_moderate_rcvbuf,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
566
		.proc_handler	= proc_dointvec,
L
Linus Torvalds 已提交
567 568 569 570 571 572
	},
	{
		.procname	= "tcp_tso_win_divisor",
		.data		= &sysctl_tcp_tso_win_divisor,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
573
		.proc_handler	= proc_dointvec,
L
Linus Torvalds 已提交
574 575
	},
	{
576
		.procname	= "tcp_congestion_control",
L
Linus Torvalds 已提交
577
		.mode		= 0644,
578
		.maxlen		= TCP_CA_NAME_MAX,
A
Alexey Dobriyan 已提交
579
		.proc_handler	= proc_tcp_congestion_control,
L
Linus Torvalds 已提交
580
	},
581
	{
582 583 584 585
		.procname	= "tcp_workaround_signed_windows",
		.data		= &sysctl_tcp_workaround_signed_windows,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
586
		.proc_handler	= proc_dointvec
587
	},
E
Eric Dumazet 已提交
588 589 590 591 592 593 594
	{
		.procname	= "tcp_limit_output_bytes",
		.data		= &sysctl_tcp_limit_output_bytes,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
E
Eric Dumazet 已提交
595 596 597 598 599 600 601
	{
		.procname	= "tcp_challenge_ack_limit",
		.data		= &sysctl_tcp_challenge_ack_limit,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
602 603 604 605 606
	{
		.procname	= "tcp_slow_start_after_idle",
		.data		= &sysctl_tcp_slow_start_after_idle,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
607
		.proc_handler	= proc_dointvec
608
	},
P
Paul Moore 已提交
609 610 611 612 613 614
#ifdef CONFIG_NETLABEL
	{
		.procname	= "cipso_cache_enable",
		.data		= &cipso_v4_cache_enabled,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
615
		.proc_handler	= proc_dointvec,
P
Paul Moore 已提交
616 617 618 619 620 621
	},
	{
		.procname	= "cipso_cache_bucket_size",
		.data		= &cipso_v4_cache_bucketsize,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
622
		.proc_handler	= proc_dointvec,
P
Paul Moore 已提交
623 624 625 626 627 628
	},
	{
		.procname	= "cipso_rbm_optfmt",
		.data		= &cipso_v4_rbm_optfmt,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
629
		.proc_handler	= proc_dointvec,
P
Paul Moore 已提交
630 631 632 633 634 635
	},
	{
		.procname	= "cipso_rbm_strictvalid",
		.data		= &cipso_v4_rbm_strictvalid,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
636
		.proc_handler	= proc_dointvec,
P
Paul Moore 已提交
637 638
	},
#endif /* CONFIG_NETLABEL */
639 640 641 642
	{
		.procname	= "tcp_available_congestion_control",
		.maxlen		= TCP_CA_BUF_MAX,
		.mode		= 0444,
A
Alexey Dobriyan 已提交
643
		.proc_handler   = proc_tcp_available_congestion_control,
644
	},
645 646 647 648
	{
		.procname	= "tcp_allowed_congestion_control",
		.maxlen		= TCP_CA_BUF_MAX,
		.mode		= 0644,
A
Alexey Dobriyan 已提交
649
		.proc_handler   = proc_allowed_congestion_control,
650
	},
A
Andreas Petlund 已提交
651 652 653 654 655 656 657
	{
		.procname       = "tcp_thin_linear_timeouts",
		.data           = &sysctl_tcp_thin_linear_timeouts,
		.maxlen         = sizeof(int),
		.mode           = 0644,
		.proc_handler   = proc_dointvec
	},
Y
Yuchung Cheng 已提交
658 659 660 661 662 663 664
	{
		.procname	= "tcp_early_retrans",
		.data		= &sysctl_tcp_early_retrans,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &zero,
N
Nandita Dukkipati 已提交
665
		.extra2		= &four,
Y
Yuchung Cheng 已提交
666
	},
E
Eric Dumazet 已提交
667 668 669 670 671 672
	{
		.procname	= "tcp_min_tso_segs",
		.data		= &sysctl_tcp_min_tso_segs,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
673
		.extra1		= &one,
E
Eric Dumazet 已提交
674 675
		.extra2		= &gso_max_segs,
	},
676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693
	{
		.procname	= "tcp_pacing_ss_ratio",
		.data		= &sysctl_tcp_pacing_ss_ratio,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &zero,
		.extra2		= &thousand,
	},
	{
		.procname	= "tcp_pacing_ca_ratio",
		.data		= &sysctl_tcp_pacing_ca_ratio,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &zero,
		.extra2		= &thousand,
	},
E
Eric Dumazet 已提交
694 695 696 697 698 699 700 701 702
	{
		.procname	= "tcp_autocorking",
		.data		= &sysctl_tcp_autocorking,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &zero,
		.extra2		= &one,
	},
703 704 705 706 707 708 709
	{
		.procname	= "tcp_invalid_ratelimit",
		.data		= &sysctl_tcp_invalid_ratelimit,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_ms_jiffies,
	},
D
Dave Watson 已提交
710 711 712 713 714 715
	{
		.procname	= "tcp_available_ulp",
		.maxlen		= TCP_ULP_BUF_MAX,
		.mode		= 0444,
		.proc_handler   = proc_tcp_available_ulp,
	},
716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731
	{
		.procname	= "icmp_msgs_per_sec",
		.data		= &sysctl_icmp_msgs_per_sec,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &zero,
	},
	{
		.procname	= "icmp_msgs_burst",
		.data		= &sysctl_icmp_msgs_burst,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &zero,
	},
H
Hideo Aoki 已提交
732 733 734 735 736
	{
		.procname	= "udp_mem",
		.data		= &sysctl_udp_mem,
		.maxlen		= sizeof(sysctl_udp_mem),
		.mode		= 0644,
E
Eric Dumazet 已提交
737
		.proc_handler	= proc_doulongvec_minmax,
H
Hideo Aoki 已提交
738 739 740 741 742 743
	},
	{
		.procname	= "udp_rmem_min",
		.data		= &sysctl_udp_rmem_min,
		.maxlen		= sizeof(sysctl_udp_rmem_min),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
744
		.proc_handler	= proc_dointvec_minmax,
745
		.extra1		= &one
H
Hideo Aoki 已提交
746 747 748 749 750 751
	},
	{
		.procname	= "udp_wmem_min",
		.data		= &sysctl_udp_wmem_min,
		.maxlen		= sizeof(sysctl_udp_wmem_min),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
752
		.proc_handler	= proc_dointvec_minmax,
753
		.extra1		= &one
H
Hideo Aoki 已提交
754
	},
755
	{ }
L
Linus Torvalds 已提交
756
};
757

758 759 760 761 762 763
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
764
		.proc_handler	= proc_dointvec
765 766 767 768 769 770
	},
	{
		.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
771
		.proc_handler	= proc_dointvec
772 773 774 775 776 777
	},
	{
		.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
778
		.proc_handler	= proc_dointvec
779 780 781 782 783 784
	},
	{
		.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
785
		.proc_handler	= proc_dointvec
786 787 788 789 790 791
	},
	{
		.procname	= "icmp_ratelimit",
		.data		= &init_net.ipv4.sysctl_icmp_ratelimit,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
792
		.proc_handler	= proc_dointvec_ms_jiffies,
793 794 795 796 797 798
	},
	{
		.procname	= "icmp_ratemask",
		.data		= &init_net.ipv4.sysctl_icmp_ratemask,
		.maxlen		= sizeof(int),
		.mode		= 0644,
A
Alexey Dobriyan 已提交
799
		.proc_handler	= proc_dointvec
800
	},
801 802
	{
		.procname	= "ping_group_range",
803
		.data		= &init_net.ipv4.ping_group_range.range,
804
		.maxlen		= sizeof(gid_t)*2,
805 806 807
		.mode		= 0644,
		.proc_handler	= ipv4_ping_group_range,
	},
808 809 810 811 812 813 814
	{
		.procname	= "tcp_ecn",
		.data		= &init_net.ipv4.sysctl_tcp_ecn,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
815 816 817 818 819 820 821
	{
		.procname	= "tcp_ecn_fallback",
		.data		= &init_net.ipv4.sysctl_tcp_ecn_fallback,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
822 823 824 825 826 827 828
	{
		.procname	= "ip_dynaddr",
		.data		= &init_net.ipv4.sysctl_ip_dynaddr,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
829 830 831 832 833 834 835
	{
		.procname	= "ip_early_demux",
		.data		= &init_net.ipv4.sysctl_ip_early_demux,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
836 837 838 839 840 841 842 843 844 845 846 847 848 849
	{
		.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
	},
850 851 852 853 854 855 856 857 858
	{
		.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,
	},
859 860
	{
		.procname	= "ip_local_port_range",
861 862
		.maxlen		= sizeof(init_net.ipv4.ip_local_ports.range),
		.data		= &init_net.ipv4.ip_local_ports.range,
863 864 865
		.mode		= 0644,
		.proc_handler	= ipv4_local_port_range,
	},
866 867 868 869 870 871 872
	{
		.procname	= "ip_local_reserved_ports",
		.data		= &init_net.ipv4.sysctl_local_reserved_ports,
		.maxlen		= 65536,
		.mode		= 0644,
		.proc_handler	= proc_do_large_bitmap,
	},
873 874 875 876 877 878 879
	{
		.procname	= "ip_no_pmtu_disc",
		.data		= &init_net.ipv4.sysctl_ip_no_pmtu_disc,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
880 881 882 883 884 885 886
	{
		.procname	= "ip_forward_use_pmtu",
		.data		= &init_net.ipv4.sysctl_ip_fwd_use_pmtu,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
887 888 889 890 891 892 893
	{
		.procname	= "ip_nonlocal_bind",
		.data		= &init_net.ipv4.sysctl_ip_nonlocal_bind,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
894 895 896 897 898 899 900
	{
		.procname	= "fwmark_reflect",
		.data		= &init_net.ipv4.sysctl_fwmark_reflect,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
901 902 903 904 905 906 907
	{
		.procname	= "tcp_fwmark_accept",
		.data		= &init_net.ipv4.sysctl_tcp_fwmark_accept,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
908 909 910 911 912 913 914 915 916 917 918
#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,
		.extra1		= &zero,
		.extra2		= &one,
	},
#endif
F
Fan Du 已提交
919 920 921 922 923 924 925 926 927 928 929 930 931 932
	{
		.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,
	},
933 934 935 936 937 938 939
	{
		.procname	= "tcp_probe_threshold",
		.data		= &init_net.ipv4.sysctl_tcp_probe_threshold,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
940 941 942 943 944 945 946
	{
		.procname	= "tcp_probe_interval",
		.data		= &init_net.ipv4.sysctl_tcp_probe_interval,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
947 948
	{
		.procname	= "igmp_link_local_mcast_reports",
949
		.data		= &init_net.ipv4.sysctl_igmp_llm_reports,
950 951 952 953
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
954 955 956 957 958 959 960
	{
		.procname	= "igmp_max_memberships",
		.data		= &init_net.ipv4.sysctl_igmp_max_memberships,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
961 962 963 964 965 966 967
	{
		.procname	= "igmp_max_msf",
		.data		= &init_net.ipv4.sysctl_igmp_max_msf,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
968 969 970 971 972 973 974 975 976 977
#ifdef CONFIG_IP_MULTICAST
	{
		.procname	= "igmp_qrv",
		.data		= &init_net.ipv4.sysctl_igmp_qrv,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &one
	},
#endif
978 979 980 981 982 983 984
	{
		.procname	= "tcp_keepalive_time",
		.data		= &init_net.ipv4.sysctl_tcp_keepalive_time,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_jiffies,
	},
985 986 987 988 989 990 991
	{
		.procname	= "tcp_keepalive_probes",
		.data		= &init_net.ipv4.sysctl_tcp_keepalive_probes,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
992 993 994 995 996 997 998
	{
		.procname	= "tcp_keepalive_intvl",
		.data		= &init_net.ipv4.sysctl_tcp_keepalive_intvl,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_jiffies,
	},
999 1000 1001 1002 1003 1004 1005 1006 1007
	{
		.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
	},
1008 1009 1010 1011 1012 1013 1014
	{
		.procname	= "tcp_synack_retries",
		.data		= &init_net.ipv4.sysctl_tcp_synack_retries,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1015 1016 1017 1018 1019 1020 1021 1022 1023
#ifdef CONFIG_SYN_COOKIES
	{
		.procname	= "tcp_syncookies",
		.data		= &init_net.ipv4.sysctl_tcp_syncookies,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
#endif
1024 1025 1026 1027 1028 1029 1030
	{
		.procname	= "tcp_reordering",
		.data		= &init_net.ipv4.sysctl_tcp_reordering,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1031 1032 1033 1034 1035 1036 1037 1038
	{
		.procname	= "tcp_retries1",
		.data		= &init_net.ipv4.sysctl_tcp_retries1,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra2		= &tcp_retr1_max
	},
1039 1040 1041 1042 1043 1044 1045
	{
		.procname	= "tcp_retries2",
		.data		= &init_net.ipv4.sysctl_tcp_retries2,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1046 1047 1048 1049 1050 1051 1052
	{
		.procname	= "tcp_orphan_retries",
		.data		= &init_net.ipv4.sysctl_tcp_orphan_retries,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1053 1054 1055 1056 1057 1058 1059
	{
		.procname	= "tcp_fin_timeout",
		.data		= &init_net.ipv4.sysctl_tcp_fin_timeout,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_jiffies,
	},
1060 1061 1062 1063 1064
	{
		.procname	= "tcp_notsent_lowat",
		.data		= &init_net.ipv4.sysctl_tcp_notsent_lowat,
		.maxlen		= sizeof(unsigned int),
		.mode		= 0644,
1065
		.proc_handler	= proc_douintvec,
1066
	},
1067 1068 1069 1070 1071 1072 1073
	{
		.procname	= "tcp_tw_reuse",
		.data		= &init_net.ipv4.sysctl_tcp_tw_reuse,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1074 1075 1076 1077 1078 1079 1080
	{
		.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
	},
1081 1082 1083 1084 1085 1086 1087
	{
		.procname	= "tcp_max_syn_backlog",
		.data		= &init_net.ipv4.sysctl_max_syn_backlog,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1088 1089 1090 1091 1092 1093 1094
#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,
1095 1096 1097 1098 1099 1100 1101 1102 1103
		.extra1		= &zero,
		.extra2		= &one,
	},
	{
		.procname	= "fib_multipath_hash_policy",
		.data		= &init_net.ipv4.sysctl_fib_multipath_hash_policy,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
1104 1105 1106 1107
		.extra1		= &zero,
		.extra2		= &one,
	},
#endif
1108 1109 1110 1111 1112 1113 1114
	{
		.procname	= "ip_unprivileged_port_start",
		.maxlen		= sizeof(int),
		.data		= &init_net.ipv4.sysctl_ip_prot_sock,
		.mode		= 0644,
		.proc_handler	= ipv4_privileged_ports,
	},
1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125
#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,
		.extra1		= &zero,
		.extra2		= &one,
	},
#endif
E
Eric Dumazet 已提交
1126 1127 1128 1129 1130 1131 1132
	{
		.procname	= "tcp_sack",
		.data		= &init_net.ipv4.sysctl_tcp_sack,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1133 1134 1135 1136 1137 1138 1139
	{
		.procname	= "tcp_window_scaling",
		.data		= &init_net.ipv4.sysctl_tcp_window_scaling,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1140 1141 1142 1143 1144 1145 1146
	{
		.procname	= "tcp_timestamps",
		.data		= &init_net.ipv4.sysctl_tcp_timestamps,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
1147 1148 1149
	{ }
};

1150 1151
static __net_init int ipv4_sysctl_init_net(struct net *net)
{
1152 1153 1154
	struct ctl_table *table;

	table = ipv4_net_table;
O
Octavian Purdila 已提交
1155
	if (!net_eq(net, &init_net)) {
1156 1157
		int i;

1158
		table = kmemdup(table, sizeof(ipv4_net_table), GFP_KERNEL);
1159
		if (!table)
1160 1161
			goto err_alloc;

1162 1163 1164
		/* 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;
1165 1166
	}

1167
	net->ipv4.ipv4_hdr = register_net_sysctl(net, "net/ipv4", table);
1168
	if (!net->ipv4.ipv4_hdr)
1169 1170
		goto err_reg;

1171 1172 1173 1174
	net->ipv4.sysctl_local_reserved_ports = kzalloc(65536 / 8, GFP_KERNEL);
	if (!net->ipv4.sysctl_local_reserved_ports)
		goto err_ports;

1175
	return 0;
1176

1177 1178
err_ports:
	unregister_net_sysctl_table(net->ipv4.ipv4_hdr);
1179
err_reg:
O
Octavian Purdila 已提交
1180
	if (!net_eq(net, &init_net))
1181 1182 1183
		kfree(table);
err_alloc:
	return -ENOMEM;
1184 1185 1186 1187
}

static __net_exit void ipv4_sysctl_exit_net(struct net *net)
{
1188 1189
	struct ctl_table *table;

1190
	kfree(net->ipv4.sysctl_local_reserved_ports);
1191 1192 1193
	table = net->ipv4.ipv4_hdr->ctl_table_arg;
	unregister_net_sysctl_table(net->ipv4.ipv4_hdr);
	kfree(table);
1194 1195 1196 1197 1198 1199 1200
}

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

1201 1202 1203 1204
static __init int sysctl_ipv4_init(void)
{
	struct ctl_table_header *hdr;

1205
	hdr = register_net_sysctl(&init_net, "net/ipv4", ipv4_table);
1206
	if (!hdr)
1207 1208 1209
		return -ENOMEM;

	if (register_pernet_subsys(&ipv4_sysctl_ops)) {
1210
		unregister_net_sysctl_table(hdr);
1211 1212 1213 1214
		return -ENOMEM;
	}

	return 0;
1215 1216 1217
}

__initcall(sysctl_ipv4_init);