tun.c 38.5 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
/*
 *  TUN - Universal TUN/TAP device driver.
 *  Copyright (C) 1999-2002 Maxim Krasnyansky <maxk@qualcomm.com>
 *
 *  This program is free software; you can redistribute it and/or modify
 *  it under the terms of the GNU General Public License as published by
 *  the Free Software Foundation; either version 2 of the License, or
 *  (at your option) any later version.
 *
 *  This program is distributed in the hope that it will be useful,
 *  but WITHOUT ANY WARRANTY; without even the implied warranty of
 *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
 *  GNU General Public License for more details.
 *
 *  $Id: tun.c,v 1.15 2002/03/01 02:44:24 maxk Exp $
 */

/*
 *  Changes:
 *
21 22 23
 *  Mike Kershaw <dragorn@kismetwireless.net> 2005/08/14
 *    Add TUNSETLINK ioctl to set the link encapsulation
 *
L
Linus Torvalds 已提交
24
 *  Mark Smith <markzzzsmith@yahoo.com.au>
J
Joe Perches 已提交
25
 *    Use eth_random_addr() for tap MAC address.
L
Linus Torvalds 已提交
26 27 28 29 30 31 32 33 34 35 36
 *
 *  Harald Roelle <harald.roelle@ifi.lmu.de>  2004/04/20
 *    Fixes in packet dropping, queue length setting and queue wakeup.
 *    Increased default tx queue length.
 *    Added ethtool API.
 *    Minor cleanups
 *
 *  Daniel Podlejski <underley@underley.eu.org>
 *    Modifications for 2.3.99-pre5 kernel.
 */

37 38
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt

L
Linus Torvalds 已提交
39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
#define DRV_NAME	"tun"
#define DRV_VERSION	"1.6"
#define DRV_DESCRIPTION	"Universal TUN/TAP device driver"
#define DRV_COPYRIGHT	"(C) 1999-2004 Max Krasnyansky <maxk@qualcomm.com>"

#include <linux/module.h>
#include <linux/errno.h>
#include <linux/kernel.h>
#include <linux/major.h>
#include <linux/slab.h>
#include <linux/poll.h>
#include <linux/fcntl.h>
#include <linux/init.h>
#include <linux/skbuff.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/miscdevice.h>
#include <linux/ethtool.h>
#include <linux/rtnetlink.h>
58
#include <linux/compat.h>
L
Linus Torvalds 已提交
59 60 61 62 63
#include <linux/if.h>
#include <linux/if_arp.h>
#include <linux/if_ether.h>
#include <linux/if_tun.h>
#include <linux/crc32.h>
64
#include <linux/nsproxy.h>
65
#include <linux/virtio_net.h>
M
Michael S. Tsirkin 已提交
66
#include <linux/rcupdate.h>
67
#include <net/net_namespace.h>
68
#include <net/netns/generic.h>
69
#include <net/rtnetlink.h>
70
#include <net/sock.h>
L
Linus Torvalds 已提交
71 72 73

#include <asm/uaccess.h>

74 75 76
/* Uncomment to enable debugging */
/* #define TUN_DEBUG 1 */

L
Linus Torvalds 已提交
77 78
#ifdef TUN_DEBUG
static int debug;
79

80 81 82 83 84 85 86 87 88 89
#define tun_debug(level, tun, fmt, args...)			\
do {								\
	if (tun->debug)						\
		netdev_printk(level, tun->dev, fmt, ##args);	\
} while (0)
#define DBG1(level, fmt, args...)				\
do {								\
	if (debug == 2)						\
		printk(level fmt, ##args);			\
} while (0)
90
#else
91 92 93 94 95 96 97 98 99 100
#define tun_debug(level, tun, fmt, args...)			\
do {								\
	if (0)							\
		netdev_printk(level, tun->dev, fmt, ##args);	\
} while (0)
#define DBG1(level, fmt, args...)				\
do {								\
	if (0)							\
		printk(level fmt, ##args);			\
} while (0)
101 102
#endif

103 104 105 106 107 108 109
#define FLT_EXACT_COUNT 8
struct tap_filter {
	unsigned int    count;    /* Number of addrs. Zero means disabled */
	u32             mask[2];  /* Mask of the hashed addrs */
	unsigned char	addr[FLT_EXACT_COUNT][ETH_ALEN];
};

E
Eric W. Biederman 已提交
110
struct tun_file {
111
	atomic_t count;
E
Eric W. Biederman 已提交
112
	struct tun_struct *tun;
E
Eric W. Biederman 已提交
113
	struct net *net;
E
Eric W. Biederman 已提交
114 115
};

116 117
struct tun_sock;

118
struct tun_struct {
E
Eric W. Biederman 已提交
119
	struct tun_file		*tfile;
120
	unsigned int 		flags;
121 122 123 124
	uid_t			owner;
	gid_t			group;

	struct net_device	*dev;
125
	netdev_features_t	set_features;
126 127
#define TUN_USER_FEATURES (NETIF_F_HW_CSUM|NETIF_F_TSO_ECN|NETIF_F_TSO| \
			  NETIF_F_TSO6|NETIF_F_UFO)
128
	struct fasync_struct	*fasync;
129

130
	struct tap_filter       txflt;
131
	struct socket		socket;
132
	struct socket_wq	wq;
133 134 135

	int			vnet_hdr_sz;

136 137
#ifdef TUN_DEBUG
	int debug;
L
Linus Torvalds 已提交
138
#endif
139
};
L
Linus Torvalds 已提交
140

141 142 143 144 145 146 147 148 149 150
struct tun_sock {
	struct sock		sk;
	struct tun_struct	*tun;
};

static inline struct tun_sock *tun_sk(struct sock *sk)
{
	return container_of(sk, struct tun_sock, sk);
}

151 152
static int tun_attach(struct tun_struct *tun, struct file *file)
{
E
Eric W. Biederman 已提交
153
	struct tun_file *tfile = file->private_data;
154
	int err;
155 156 157

	ASSERT_RTNL();

158 159 160 161 162 163 164 165 166 167 168
	netif_tx_lock_bh(tun->dev);

	err = -EINVAL;
	if (tfile->tun)
		goto out;

	err = -EBUSY;
	if (tun->tfile)
		goto out;

	err = 0;
E
Eric W. Biederman 已提交
169 170
	tfile->tun = tun;
	tun->tfile = tfile;
171
	tun->socket.file = file;
172
	netif_carrier_on(tun->dev);
173
	dev_hold(tun->dev);
174
	sock_hold(tun->socket.sk);
175
	atomic_inc(&tfile->count);
176

177 178 179
out:
	netif_tx_unlock_bh(tun->dev);
	return err;
180 181
}

E
Eric W. Biederman 已提交
182 183 184
static void __tun_detach(struct tun_struct *tun)
{
	/* Detach from net device */
185
	netif_tx_lock_bh(tun->dev);
186
	netif_carrier_off(tun->dev);
E
Eric W. Biederman 已提交
187
	tun->tfile = NULL;
188
	tun->socket.file = NULL;
189
	netif_tx_unlock_bh(tun->dev);
E
Eric W. Biederman 已提交
190 191

	/* Drop read queue */
192
	skb_queue_purge(&tun->socket.sk->sk_receive_queue);
193 194 195 196 197 198 199 200 201 202

	/* Drop the extra count on the net device */
	dev_put(tun->dev);
}

static void tun_detach(struct tun_struct *tun)
{
	rtnl_lock();
	__tun_detach(tun);
	rtnl_unlock();
E
Eric W. Biederman 已提交
203 204 205 206
}

static struct tun_struct *__tun_get(struct tun_file *tfile)
{
207 208 209 210 211 212
	struct tun_struct *tun = NULL;

	if (atomic_inc_not_zero(&tfile->count))
		tun = tfile->tun;

	return tun;
E
Eric W. Biederman 已提交
213 214 215 216 217 218 219 220 221
}

static struct tun_struct *tun_get(struct file *file)
{
	return __tun_get(file->private_data);
}

static void tun_put(struct tun_struct *tun)
{
222 223 224 225
	struct tun_file *tfile = tun->tfile;

	if (atomic_dec_and_test(&tfile->count))
		tun_detach(tfile->tun);
E
Eric W. Biederman 已提交
226 227
}

228
/* TAP filtering */
229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277
static void addr_hash_set(u32 *mask, const u8 *addr)
{
	int n = ether_crc(ETH_ALEN, addr) >> 26;
	mask[n >> 5] |= (1 << (n & 31));
}

static unsigned int addr_hash_test(const u32 *mask, const u8 *addr)
{
	int n = ether_crc(ETH_ALEN, addr) >> 26;
	return mask[n >> 5] & (1 << (n & 31));
}

static int update_filter(struct tap_filter *filter, void __user *arg)
{
	struct { u8 u[ETH_ALEN]; } *addr;
	struct tun_filter uf;
	int err, alen, n, nexact;

	if (copy_from_user(&uf, arg, sizeof(uf)))
		return -EFAULT;

	if (!uf.count) {
		/* Disabled */
		filter->count = 0;
		return 0;
	}

	alen = ETH_ALEN * uf.count;
	addr = kmalloc(alen, GFP_KERNEL);
	if (!addr)
		return -ENOMEM;

	if (copy_from_user(addr, arg + sizeof(uf), alen)) {
		err = -EFAULT;
		goto done;
	}

	/* The filter is updated without holding any locks. Which is
	 * perfectly safe. We disable it first and in the worst
	 * case we'll accept a few undesired packets. */
	filter->count = 0;
	wmb();

	/* Use first set of addresses as an exact filter */
	for (n = 0; n < uf.count && n < FLT_EXACT_COUNT; n++)
		memcpy(filter->addr[n], addr[n].u, ETH_ALEN);

	nexact = n;

278 279
	/* Remaining multicast addresses are hashed,
	 * unicast will leave the filter disabled. */
280
	memset(filter->mask, 0, sizeof(filter->mask));
281 282 283 284 285
	for (; n < uf.count; n++) {
		if (!is_multicast_ether_addr(addr[n].u)) {
			err = 0; /* no filter */
			goto done;
		}
286
		addr_hash_set(filter->mask, addr[n].u);
287
	}
288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315

	/* For ALLMULTI just set the mask to all ones.
	 * This overrides the mask populated above. */
	if ((uf.flags & TUN_FLT_ALLMULTI))
		memset(filter->mask, ~0, sizeof(filter->mask));

	/* Now enable the filter */
	wmb();
	filter->count = nexact;

	/* Return the number of exact filters */
	err = nexact;

done:
	kfree(addr);
	return err;
}

/* Returns: 0 - drop, !=0 - accept */
static int run_filter(struct tap_filter *filter, const struct sk_buff *skb)
{
	/* Cannot use eth_hdr(skb) here because skb_mac_hdr() is incorrect
	 * at this point. */
	struct ethhdr *eh = (struct ethhdr *) skb->data;
	int i;

	/* Exact match */
	for (i = 0; i < filter->count; i++)
316
		if (ether_addr_equal(eh->h_dest, filter->addr[i]))
317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337
			return 1;

	/* Inexact match (multicast only) */
	if (is_multicast_ether_addr(eh->h_dest))
		return addr_hash_test(filter->mask, eh->h_dest);

	return 0;
}

/*
 * Checks whether the packet is accepted or not.
 * Returns: 0 - drop, !=0 - accept
 */
static int check_filter(struct tap_filter *filter, const struct sk_buff *skb)
{
	if (!filter->count)
		return 1;

	return run_filter(filter, skb);
}

L
Linus Torvalds 已提交
338 339
/* Network device part of the driver */

340
static const struct ethtool_ops tun_ethtool_ops;
L
Linus Torvalds 已提交
341

342 343 344 345 346 347 348 349 350
/* Net device detach from fd. */
static void tun_net_uninit(struct net_device *dev)
{
	struct tun_struct *tun = netdev_priv(dev);
	struct tun_file *tfile = tun->tfile;

	/* Inform the methods they need to stop using the dev.
	 */
	if (tfile) {
351
		wake_up_all(&tun->wq.wait);
352 353 354 355 356
		if (atomic_dec_and_test(&tfile->count))
			__tun_detach(tun);
	}
}

357 358 359 360
static void tun_free_netdev(struct net_device *dev)
{
	struct tun_struct *tun = netdev_priv(dev);

361 362
	BUG_ON(!test_bit(SOCK_EXTERNALLY_ALLOCATED, &tun->socket.flags));

363
	sk_release_kernel(tun->socket.sk);
364 365
}

L
Linus Torvalds 已提交
366 367 368 369 370 371 372 373 374 375 376 377 378 379 380
/* Net device open. */
static int tun_net_open(struct net_device *dev)
{
	netif_start_queue(dev);
	return 0;
}

/* Net device close. */
static int tun_net_close(struct net_device *dev)
{
	netif_stop_queue(dev);
	return 0;
}

/* Net device start xmit */
381
static netdev_tx_t tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
L
Linus Torvalds 已提交
382 383 384
{
	struct tun_struct *tun = netdev_priv(dev);

385
	tun_debug(KERN_INFO, tun, "tun_net_xmit %d\n", skb->len);
L
Linus Torvalds 已提交
386 387

	/* Drop packet if interface is not attached */
E
Eric W. Biederman 已提交
388
	if (!tun->tfile)
L
Linus Torvalds 已提交
389 390
		goto drop;

391 392 393 394 395 396
	/* Drop if the filter does not like it.
	 * This is a noop if the filter is disabled.
	 * Filter can be enabled only for the TAP devices. */
	if (!check_filter(&tun->txflt, skb))
		goto drop;

M
Michael S. Tsirkin 已提交
397 398 399 400
	if (tun->socket.sk->sk_filter &&
	    sk_filter(tun->socket.sk, skb))
		goto drop;

401
	if (skb_queue_len(&tun->socket.sk->sk_receive_queue) >= dev->tx_queue_len) {
L
Linus Torvalds 已提交
402 403 404 405 406 407 408
		if (!(tun->flags & TUN_ONE_QUEUE)) {
			/* Normal queueing mode. */
			/* Packet scheduler handles dropping of further packets. */
			netif_stop_queue(dev);

			/* We won't see all dropped packets individually, so overrun
			 * error is more appropriate. */
409
			dev->stats.tx_fifo_errors++;
L
Linus Torvalds 已提交
410 411 412 413 414 415 416
		} else {
			/* Single queue mode.
			 * Driver handles dropping of all packets itself. */
			goto drop;
		}
	}

M
Michael S. Tsirkin 已提交
417 418 419 420
	/* Orphan the skb - required as we might hang on to it
	 * for indefinite time. */
	skb_orphan(skb);

421
	/* Enqueue packet */
422
	skb_queue_tail(&tun->socket.sk->sk_receive_queue, skb);
L
Linus Torvalds 已提交
423 424 425 426

	/* Notify and wake up reader process */
	if (tun->flags & TUN_FASYNC)
		kill_fasync(&tun->fasync, SIGIO, POLL_IN);
427
	wake_up_interruptible_poll(&tun->wq.wait, POLLIN |
428
				   POLLRDNORM | POLLRDBAND);
429
	return NETDEV_TX_OK;
L
Linus Torvalds 已提交
430 431

drop:
432
	dev->stats.tx_dropped++;
L
Linus Torvalds 已提交
433
	kfree_skb(skb);
434
	return NETDEV_TX_OK;
L
Linus Torvalds 已提交
435 436
}

437
static void tun_net_mclist(struct net_device *dev)
L
Linus Torvalds 已提交
438
{
439 440 441 442 443
	/*
	 * This callback is supposed to deal with mc filter in
	 * _rx_ path and has nothing to do with the _tx_ path.
	 * In rx path we always accept everything userspace gives us.
	 */
L
Linus Torvalds 已提交
444 445
}

E
Ed Swierk 已提交
446 447 448 449 450 451 452 453 454 455 456 457
#define MIN_MTU 68
#define MAX_MTU 65535

static int
tun_net_change_mtu(struct net_device *dev, int new_mtu)
{
	if (new_mtu < MIN_MTU || new_mtu + dev->hard_header_len > MAX_MTU)
		return -EINVAL;
	dev->mtu = new_mtu;
	return 0;
}

458 459
static netdev_features_t tun_net_fix_features(struct net_device *dev,
	netdev_features_t features)
460 461 462 463 464
{
	struct tun_struct *tun = netdev_priv(dev);

	return (features & tun->set_features) | (features & ~TUN_USER_FEATURES);
}
465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481
#ifdef CONFIG_NET_POLL_CONTROLLER
static void tun_poll_controller(struct net_device *dev)
{
	/*
	 * Tun only receives frames when:
	 * 1) the char device endpoint gets data from user space
	 * 2) the tun socket gets a sendmsg call from user space
	 * Since both of those are syncronous operations, we are guaranteed
	 * never to have pending data when we poll for it
	 * so theres nothing to do here but return.
	 * We need this though so netpoll recognizes us as an interface that
	 * supports polling, which enables bridge devices in virt setups to
	 * still use netconsole
	 */
	return;
}
#endif
S
Stephen Hemminger 已提交
482
static const struct net_device_ops tun_netdev_ops = {
483
	.ndo_uninit		= tun_net_uninit,
S
Stephen Hemminger 已提交
484 485
	.ndo_open		= tun_net_open,
	.ndo_stop		= tun_net_close,
486
	.ndo_start_xmit		= tun_net_xmit,
S
Stephen Hemminger 已提交
487
	.ndo_change_mtu		= tun_net_change_mtu,
488
	.ndo_fix_features	= tun_net_fix_features,
489 490 491
#ifdef CONFIG_NET_POLL_CONTROLLER
	.ndo_poll_controller	= tun_poll_controller,
#endif
S
Stephen Hemminger 已提交
492 493 494
};

static const struct net_device_ops tap_netdev_ops = {
495
	.ndo_uninit		= tun_net_uninit,
S
Stephen Hemminger 已提交
496 497
	.ndo_open		= tun_net_open,
	.ndo_stop		= tun_net_close,
498
	.ndo_start_xmit		= tun_net_xmit,
S
Stephen Hemminger 已提交
499
	.ndo_change_mtu		= tun_net_change_mtu,
500
	.ndo_fix_features	= tun_net_fix_features,
501
	.ndo_set_rx_mode	= tun_net_mclist,
S
Stephen Hemminger 已提交
502 503
	.ndo_set_mac_address	= eth_mac_addr,
	.ndo_validate_addr	= eth_validate_addr,
504 505 506
#ifdef CONFIG_NET_POLL_CONTROLLER
	.ndo_poll_controller	= tun_poll_controller,
#endif
S
Stephen Hemminger 已提交
507 508
};

L
Linus Torvalds 已提交
509 510 511 512
/* Initialize net device. */
static void tun_net_init(struct net_device *dev)
{
	struct tun_struct *tun = netdev_priv(dev);
513

L
Linus Torvalds 已提交
514 515
	switch (tun->flags & TUN_TYPE_MASK) {
	case TUN_TUN_DEV:
S
Stephen Hemminger 已提交
516 517
		dev->netdev_ops = &tun_netdev_ops;

L
Linus Torvalds 已提交
518 519 520 521 522 523
		/* Point-to-Point TUN Device */
		dev->hard_header_len = 0;
		dev->addr_len = 0;
		dev->mtu = 1500;

		/* Zero header length */
524
		dev->type = ARPHRD_NONE;
L
Linus Torvalds 已提交
525 526 527 528 529
		dev->flags = IFF_POINTOPOINT | IFF_NOARP | IFF_MULTICAST;
		dev->tx_queue_len = TUN_READQ_SIZE;  /* We prefer our own queue length */
		break;

	case TUN_TAP_DEV:
K
Kusanagi Kouichi 已提交
530
		dev->netdev_ops = &tap_netdev_ops;
L
Linus Torvalds 已提交
531 532
		/* Ethernet TAP Device */
		ether_setup(dev);
533
		dev->priv_flags &= ~IFF_TX_SKB_SHARING;
534

535
		eth_hw_addr_random(dev);
536

L
Linus Torvalds 已提交
537 538 539 540 541 542 543 544 545
		dev->tx_queue_len = TUN_READQ_SIZE;  /* We prefer our own queue length */
		break;
	}
}

/* Character device part */

/* Poll */
static unsigned int tun_chr_poll(struct file *file, poll_table * wait)
546
{
547 548
	struct tun_file *tfile = file->private_data;
	struct tun_struct *tun = __tun_get(tfile);
549
	struct sock *sk;
550
	unsigned int mask = 0;
L
Linus Torvalds 已提交
551 552

	if (!tun)
553
		return POLLERR;
L
Linus Torvalds 已提交
554

555
	sk = tun->socket.sk;
556

557
	tun_debug(KERN_INFO, tun, "tun_chr_poll\n");
L
Linus Torvalds 已提交
558

559
	poll_wait(file, &tun->wq.wait, wait);
560

561
	if (!skb_queue_empty(&sk->sk_receive_queue))
L
Linus Torvalds 已提交
562 563
		mask |= POLLIN | POLLRDNORM;

564 565 566 567 568
	if (sock_writeable(sk) ||
	    (!test_and_set_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags) &&
	     sock_writeable(sk)))
		mask |= POLLOUT | POLLWRNORM;

569 570 571
	if (tun->dev->reg_state != NETREG_REGISTERED)
		mask = POLLERR;

E
Eric W. Biederman 已提交
572
	tun_put(tun);
L
Linus Torvalds 已提交
573 574 575
	return mask;
}

576 577
/* prepad is the amount to reserve at front.  len is length after that.
 * linear is a hint as to how much to copy (usually headers). */
578 579 580
static struct sk_buff *tun_alloc_skb(struct tun_struct *tun,
				     size_t prepad, size_t len,
				     size_t linear, int noblock)
581
{
582
	struct sock *sk = tun->socket.sk;
583
	struct sk_buff *skb;
584
	int err;
585

586 587
	sock_update_classid(sk);

588
	/* Under a page?  Don't bother with paged skb. */
H
Herbert Xu 已提交
589
	if (prepad + len < PAGE_SIZE || !linear)
590
		linear = len;
591

592 593
	skb = sock_alloc_send_pskb(sk, prepad + linear, len - linear, noblock,
				   &err);
594
	if (!skb)
595
		return ERR_PTR(err);
596 597 598

	skb_reserve(skb, prepad);
	skb_put(skb, linear);
599 600
	skb->data_len = len - linear;
	skb->len += len - linear;
601 602 603 604

	return skb;
}

L
Linus Torvalds 已提交
605
/* Get packet from user space buffer */
606 607 608
static ssize_t tun_get_user(struct tun_struct *tun,
			    const struct iovec *iv, size_t count,
			    int noblock)
L
Linus Torvalds 已提交
609
{
610
	struct tun_pi pi = { 0, cpu_to_be16(ETH_P_IP) };
L
Linus Torvalds 已提交
611
	struct sk_buff *skb;
612
	size_t len = count, align = NET_SKB_PAD;
613
	struct virtio_net_hdr gso = { 0 };
614
	int offset = 0;
L
Linus Torvalds 已提交
615 616 617 618 619

	if (!(tun->flags & TUN_NO_PI)) {
		if ((len -= sizeof(pi)) > count)
			return -EINVAL;

620
		if (memcpy_fromiovecend((void *)&pi, iv, 0, sizeof(pi)))
L
Linus Torvalds 已提交
621
			return -EFAULT;
622
		offset += sizeof(pi);
L
Linus Torvalds 已提交
623 624
	}

625
	if (tun->flags & TUN_VNET_HDR) {
626
		if ((len -= tun->vnet_hdr_sz) > count)
627 628
			return -EINVAL;

629
		if (memcpy_fromiovecend((void *)&gso, iv, offset, sizeof(gso)))
630 631
			return -EFAULT;

632 633 634 635
		if ((gso.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) &&
		    gso.csum_start + gso.csum_offset + 2 > gso.hdr_len)
			gso.hdr_len = gso.csum_start + gso.csum_offset + 2;

636 637
		if (gso.hdr_len > len)
			return -EINVAL;
638
		offset += tun->vnet_hdr_sz;
639 640
	}

641
	if ((tun->flags & TUN_TYPE_MASK) == TUN_TAP_DEV) {
642
		align += NET_IP_ALIGN;
H
Herbert Xu 已提交
643 644
		if (unlikely(len < ETH_HLEN ||
			     (gso.hdr_len && gso.hdr_len < ETH_HLEN)))
645 646
			return -EINVAL;
	}
647

648 649 650 651 652
	skb = tun_alloc_skb(tun, align, len, gso.hdr_len, noblock);
	if (IS_ERR(skb)) {
		if (PTR_ERR(skb) != -EAGAIN)
			tun->dev->stats.rx_dropped++;
		return PTR_ERR(skb);
L
Linus Torvalds 已提交
653 654
	}

655
	if (skb_copy_datagram_from_iovec(skb, 0, iv, offset, len)) {
656
		tun->dev->stats.rx_dropped++;
D
Dave Jones 已提交
657
		kfree_skb(skb);
L
Linus Torvalds 已提交
658
		return -EFAULT;
D
Dave Jones 已提交
659
	}
L
Linus Torvalds 已提交
660

661 662 663 664 665 666 667
	if (gso.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) {
		if (!skb_partial_csum_set(skb, gso.csum_start,
					  gso.csum_offset)) {
			tun->dev->stats.rx_frame_errors++;
			kfree_skb(skb);
			return -EINVAL;
		}
668
	}
669

L
Linus Torvalds 已提交
670 671
	switch (tun->flags & TUN_TYPE_MASK) {
	case TUN_TUN_DEV:
672 673 674 675 676 677 678 679 680 681 682 683 684 685 686
		if (tun->flags & TUN_NO_PI) {
			switch (skb->data[0] & 0xf0) {
			case 0x40:
				pi.proto = htons(ETH_P_IP);
				break;
			case 0x60:
				pi.proto = htons(ETH_P_IPV6);
				break;
			default:
				tun->dev->stats.rx_dropped++;
				kfree_skb(skb);
				return -EINVAL;
			}
		}

687
		skb_reset_mac_header(skb);
L
Linus Torvalds 已提交
688
		skb->protocol = pi.proto;
689
		skb->dev = tun->dev;
L
Linus Torvalds 已提交
690 691 692 693
		break;
	case TUN_TAP_DEV:
		skb->protocol = eth_type_trans(skb, tun->dev);
		break;
694
	}
L
Linus Torvalds 已提交
695

696 697 698 699 700 701 702 703 704
	if (gso.gso_type != VIRTIO_NET_HDR_GSO_NONE) {
		pr_debug("GSO!\n");
		switch (gso.gso_type & ~VIRTIO_NET_HDR_GSO_ECN) {
		case VIRTIO_NET_HDR_GSO_TCPV4:
			skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4;
			break;
		case VIRTIO_NET_HDR_GSO_TCPV6:
			skb_shinfo(skb)->gso_type = SKB_GSO_TCPV6;
			break;
705 706 707
		case VIRTIO_NET_HDR_GSO_UDP:
			skb_shinfo(skb)->gso_type = SKB_GSO_UDP;
			break;
708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727
		default:
			tun->dev->stats.rx_frame_errors++;
			kfree_skb(skb);
			return -EINVAL;
		}

		if (gso.gso_type & VIRTIO_NET_HDR_GSO_ECN)
			skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_ECN;

		skb_shinfo(skb)->gso_size = gso.gso_size;
		if (skb_shinfo(skb)->gso_size == 0) {
			tun->dev->stats.rx_frame_errors++;
			kfree_skb(skb);
			return -EINVAL;
		}

		/* Header must be checked, and gso_segs computed. */
		skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY;
		skb_shinfo(skb)->gso_segs = 0;
	}
728

L
Linus Torvalds 已提交
729
	netif_rx_ni(skb);
730

731 732
	tun->dev->stats.rx_packets++;
	tun->dev->stats.rx_bytes += len;
L
Linus Torvalds 已提交
733 734

	return count;
735
}
L
Linus Torvalds 已提交
736

737 738
static ssize_t tun_chr_aio_write(struct kiocb *iocb, const struct iovec *iv,
			      unsigned long count, loff_t pos)
L
Linus Torvalds 已提交
739
{
740
	struct file *file = iocb->ki_filp;
H
Herbert Xu 已提交
741
	struct tun_struct *tun = tun_get(file);
E
Eric W. Biederman 已提交
742
	ssize_t result;
L
Linus Torvalds 已提交
743 744 745 746

	if (!tun)
		return -EBADFD;

747
	tun_debug(KERN_INFO, tun, "tun_chr_write %ld\n", count);
L
Linus Torvalds 已提交
748

749
	result = tun_get_user(tun, iv, iov_length(iv, count),
750
			      file->f_flags & O_NONBLOCK);
E
Eric W. Biederman 已提交
751 752 753

	tun_put(tun);
	return result;
L
Linus Torvalds 已提交
754 755 756
}

/* Put packet to the user space buffer */
757 758 759
static ssize_t tun_put_user(struct tun_struct *tun,
			    struct sk_buff *skb,
			    const struct iovec *iv, int len)
L
Linus Torvalds 已提交
760 761 762 763 764 765 766 767 768 769 770 771
{
	struct tun_pi pi = { 0, skb->protocol };
	ssize_t total = 0;

	if (!(tun->flags & TUN_NO_PI)) {
		if ((len -= sizeof(pi)) < 0)
			return -EINVAL;

		if (len < skb->len) {
			/* Packet will be striped */
			pi.flags |= TUN_PKT_STRIP;
		}
772

773
		if (memcpy_toiovecend(iv, (void *) &pi, 0, sizeof(pi)))
L
Linus Torvalds 已提交
774 775
			return -EFAULT;
		total += sizeof(pi);
776
	}
L
Linus Torvalds 已提交
777

778 779
	if (tun->flags & TUN_VNET_HDR) {
		struct virtio_net_hdr gso = { 0 }; /* no info leak */
780
		if ((len -= tun->vnet_hdr_sz) < 0)
781 782 783 784 785 786 787 788 789 790 791 792
			return -EINVAL;

		if (skb_is_gso(skb)) {
			struct skb_shared_info *sinfo = skb_shinfo(skb);

			/* This is a hint as to how much should be linear. */
			gso.hdr_len = skb_headlen(skb);
			gso.gso_size = sinfo->gso_size;
			if (sinfo->gso_type & SKB_GSO_TCPV4)
				gso.gso_type = VIRTIO_NET_HDR_GSO_TCPV4;
			else if (sinfo->gso_type & SKB_GSO_TCPV6)
				gso.gso_type = VIRTIO_NET_HDR_GSO_TCPV6;
793 794
			else if (sinfo->gso_type & SKB_GSO_UDP)
				gso.gso_type = VIRTIO_NET_HDR_GSO_UDP;
795
			else {
796
				pr_err("unexpected GSO type: "
797 798 799 800 801 802 803 804 805 806
				       "0x%x, gso_size %d, hdr_len %d\n",
				       sinfo->gso_type, gso.gso_size,
				       gso.hdr_len);
				print_hex_dump(KERN_ERR, "tun: ",
					       DUMP_PREFIX_NONE,
					       16, 1, skb->head,
					       min((int)gso.hdr_len, 64), true);
				WARN_ON_ONCE(1);
				return -EINVAL;
			}
807 808 809 810 811 812 813
			if (sinfo->gso_type & SKB_GSO_TCP_ECN)
				gso.gso_type |= VIRTIO_NET_HDR_GSO_ECN;
		} else
			gso.gso_type = VIRTIO_NET_HDR_GSO_NONE;

		if (skb->ip_summed == CHECKSUM_PARTIAL) {
			gso.flags = VIRTIO_NET_HDR_F_NEEDS_CSUM;
814
			gso.csum_start = skb_checksum_start_offset(skb);
815
			gso.csum_offset = skb->csum_offset;
816 817
		} else if (skb->ip_summed == CHECKSUM_UNNECESSARY) {
			gso.flags = VIRTIO_NET_HDR_F_DATA_VALID;
818 819
		} /* else everything is zero */

820 821
		if (unlikely(memcpy_toiovecend(iv, (void *)&gso, total,
					       sizeof(gso))))
822
			return -EFAULT;
823
		total += tun->vnet_hdr_sz;
824 825
	}

L
Linus Torvalds 已提交
826 827
	len = min_t(int, skb->len, len);

828
	skb_copy_datagram_const_iovec(skb, 0, iv, total, len);
829
	total += skb->len;
L
Linus Torvalds 已提交
830

831 832
	tun->dev->stats.tx_packets++;
	tun->dev->stats.tx_bytes += len;
L
Linus Torvalds 已提交
833 834 835 836

	return total;
}

837 838 839
static ssize_t tun_do_read(struct tun_struct *tun,
			   struct kiocb *iocb, const struct iovec *iv,
			   ssize_t len, int noblock)
L
Linus Torvalds 已提交
840 841 842
{
	DECLARE_WAITQUEUE(wait, current);
	struct sk_buff *skb;
843
	ssize_t ret = 0;
L
Linus Torvalds 已提交
844

845
	tun_debug(KERN_INFO, tun, "tun_chr_read\n");
L
Linus Torvalds 已提交
846

847 848
	if (unlikely(!noblock))
		add_wait_queue(&tun->wq.wait, &wait);
L
Linus Torvalds 已提交
849 850 851 852
	while (len) {
		current->state = TASK_INTERRUPTIBLE;

		/* Read frames from the queue */
853
		if (!(skb=skb_dequeue(&tun->socket.sk->sk_receive_queue))) {
854
			if (noblock) {
L
Linus Torvalds 已提交
855 856 857 858 859 860 861
				ret = -EAGAIN;
				break;
			}
			if (signal_pending(current)) {
				ret = -ERESTARTSYS;
				break;
			}
862 863 864 865
			if (tun->dev->reg_state != NETREG_REGISTERED) {
				ret = -EIO;
				break;
			}
L
Linus Torvalds 已提交
866 867 868 869 870 871 872

			/* Nothing to read, let's sleep */
			schedule();
			continue;
		}
		netif_wake_queue(tun->dev);

873
		ret = tun_put_user(tun, skb, iv, len);
874 875
		kfree_skb(skb);
		break;
L
Linus Torvalds 已提交
876 877 878
	}

	current->state = TASK_RUNNING;
879 880
	if (unlikely(!noblock))
		remove_wait_queue(&tun->wq.wait, &wait);
L
Linus Torvalds 已提交
881

882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902
	return ret;
}

static ssize_t tun_chr_aio_read(struct kiocb *iocb, const struct iovec *iv,
			    unsigned long count, loff_t pos)
{
	struct file *file = iocb->ki_filp;
	struct tun_file *tfile = file->private_data;
	struct tun_struct *tun = __tun_get(tfile);
	ssize_t len, ret;

	if (!tun)
		return -EBADFD;
	len = iov_length(iv, count);
	if (len < 0) {
		ret = -EINVAL;
		goto out;
	}

	ret = tun_do_read(tun, iocb, iv, len, file->f_flags & O_NONBLOCK);
	ret = min_t(ssize_t, ret, len);
E
Eric W. Biederman 已提交
903 904
out:
	tun_put(tun);
L
Linus Torvalds 已提交
905 906 907 908 909 910 911 912
	return ret;
}

static void tun_setup(struct net_device *dev)
{
	struct tun_struct *tun = netdev_priv(dev);

	tun->owner = -1;
913
	tun->group = -1;
L
Linus Torvalds 已提交
914 915

	dev->ethtool_ops = &tun_ethtool_ops;
916
	dev->destructor = tun_free_netdev;
L
Linus Torvalds 已提交
917 918
}

919 920 921 922 923 924 925 926 927 928 929 930 931 932 933
/* Trivial set of netlink ops to allow deleting tun or tap
 * device with netlink.
 */
static int tun_validate(struct nlattr *tb[], struct nlattr *data[])
{
	return -EINVAL;
}

static struct rtnl_link_ops tun_link_ops __read_mostly = {
	.kind		= DRV_NAME,
	.priv_size	= sizeof(struct tun_struct),
	.setup		= tun_setup,
	.validate	= tun_validate,
};

934 935 936
static void tun_sock_write_space(struct sock *sk)
{
	struct tun_struct *tun;
937
	wait_queue_head_t *wqueue;
938 939 940 941 942 943 944

	if (!sock_writeable(sk))
		return;

	if (!test_and_clear_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags))
		return;

945 946 947
	wqueue = sk_sleep(sk);
	if (wqueue && waitqueue_active(wqueue))
		wake_up_interruptible_sync_poll(wqueue, POLLOUT |
948
						POLLWRNORM | POLLWRBAND);
H
Herbert Xu 已提交
949

950
	tun = tun_sk(sk)->tun;
951 952 953 954 955
	kill_fasync(&tun->fasync, SIGIO, POLL_OUT);
}

static void tun_sock_destruct(struct sock *sk)
{
956
	free_netdev(tun_sk(sk)->tun->dev);
957 958
}

959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983
static int tun_sendmsg(struct kiocb *iocb, struct socket *sock,
		       struct msghdr *m, size_t total_len)
{
	struct tun_struct *tun = container_of(sock, struct tun_struct, socket);
	return tun_get_user(tun, m->msg_iov, total_len,
			    m->msg_flags & MSG_DONTWAIT);
}

static int tun_recvmsg(struct kiocb *iocb, struct socket *sock,
		       struct msghdr *m, size_t total_len,
		       int flags)
{
	struct tun_struct *tun = container_of(sock, struct tun_struct, socket);
	int ret;
	if (flags & ~(MSG_DONTWAIT|MSG_TRUNC))
		return -EINVAL;
	ret = tun_do_read(tun, iocb, m->msg_iov, total_len,
			  flags & MSG_DONTWAIT);
	if (ret > total_len) {
		m->msg_flags |= MSG_TRUNC;
		ret = flags & MSG_TRUNC ? ret : total_len;
	}
	return ret;
}

984 985 986 987 988 989 990
static int tun_release(struct socket *sock)
{
	if (sock->sk)
		sock_put(sock->sk);
	return 0;
}

991 992 993 994
/* Ops structure to mimic raw sockets with tun */
static const struct proto_ops tun_socket_ops = {
	.sendmsg = tun_sendmsg,
	.recvmsg = tun_recvmsg,
995
	.release = tun_release,
996 997
};

998 999 1000 1001 1002
static struct proto tun_proto = {
	.name		= "tun",
	.owner		= THIS_MODULE,
	.obj_size	= sizeof(struct tun_sock),
};
1003

1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049
static int tun_flags(struct tun_struct *tun)
{
	int flags = 0;

	if (tun->flags & TUN_TUN_DEV)
		flags |= IFF_TUN;
	else
		flags |= IFF_TAP;

	if (tun->flags & TUN_NO_PI)
		flags |= IFF_NO_PI;

	if (tun->flags & TUN_ONE_QUEUE)
		flags |= IFF_ONE_QUEUE;

	if (tun->flags & TUN_VNET_HDR)
		flags |= IFF_VNET_HDR;

	return flags;
}

static ssize_t tun_show_flags(struct device *dev, struct device_attribute *attr,
			      char *buf)
{
	struct tun_struct *tun = netdev_priv(to_net_dev(dev));
	return sprintf(buf, "0x%x\n", tun_flags(tun));
}

static ssize_t tun_show_owner(struct device *dev, struct device_attribute *attr,
			      char *buf)
{
	struct tun_struct *tun = netdev_priv(to_net_dev(dev));
	return sprintf(buf, "%d\n", tun->owner);
}

static ssize_t tun_show_group(struct device *dev, struct device_attribute *attr,
			      char *buf)
{
	struct tun_struct *tun = netdev_priv(to_net_dev(dev));
	return sprintf(buf, "%d\n", tun->group);
}

static DEVICE_ATTR(tun_flags, 0444, tun_show_flags, NULL);
static DEVICE_ATTR(owner, 0444, tun_show_owner, NULL);
static DEVICE_ATTR(group, 0444, tun_show_group, NULL);

1050
static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
L
Linus Torvalds 已提交
1051
{
1052
	struct sock *sk;
L
Linus Torvalds 已提交
1053 1054 1055 1056
	struct tun_struct *tun;
	struct net_device *dev;
	int err;

1057 1058
	dev = __dev_get_by_name(net, ifr->ifr_name);
	if (dev) {
P
Paul Moore 已提交
1059 1060
		const struct cred *cred = current_cred();

1061 1062
		if (ifr->ifr_flags & IFF_TUN_EXCL)
			return -EBUSY;
1063 1064 1065 1066 1067 1068 1069
		if ((ifr->ifr_flags & IFF_TUN) && dev->netdev_ops == &tun_netdev_ops)
			tun = netdev_priv(dev);
		else if ((ifr->ifr_flags & IFF_TAP) && dev->netdev_ops == &tap_netdev_ops)
			tun = netdev_priv(dev);
		else
			return -EINVAL;

P
Paul Moore 已提交
1070 1071 1072 1073
		if (((tun->owner != -1 && cred->euid != tun->owner) ||
		     (tun->group != -1 && !in_egroup_p(tun->group))) &&
		    !capable(CAP_NET_ADMIN))
			return -EPERM;
1074
		err = security_tun_dev_attach(tun->socket.sk);
P
Paul Moore 已提交
1075 1076 1077
		if (err < 0)
			return err;

1078 1079 1080
		err = tun_attach(tun, file);
		if (err < 0)
			return err;
1081
	}
L
Linus Torvalds 已提交
1082 1083 1084 1085
	else {
		char *name;
		unsigned long flags = 0;

1086 1087
		if (!capable(CAP_NET_ADMIN))
			return -EPERM;
P
Paul Moore 已提交
1088 1089 1090
		err = security_tun_dev_create();
		if (err < 0)
			return err;
1091

L
Linus Torvalds 已提交
1092 1093 1094 1095 1096 1097 1098 1099 1100
		/* Set dev type */
		if (ifr->ifr_flags & IFF_TUN) {
			/* TUN device */
			flags |= TUN_TUN_DEV;
			name = "tun%d";
		} else if (ifr->ifr_flags & IFF_TAP) {
			/* TAP device */
			flags |= TUN_TAP_DEV;
			name = "tap%d";
1101
		} else
1102
			return -EINVAL;
1103

L
Linus Torvalds 已提交
1104 1105 1106 1107 1108 1109 1110 1111
		if (*ifr->ifr_name)
			name = ifr->ifr_name;

		dev = alloc_netdev(sizeof(struct tun_struct), name,
				   tun_setup);
		if (!dev)
			return -ENOMEM;

1112
		dev_net_set(dev, net);
1113
		dev->rtnl_link_ops = &tun_link_ops;
S
Stephen Hemminger 已提交
1114

L
Linus Torvalds 已提交
1115 1116 1117
		tun = netdev_priv(dev);
		tun->dev = dev;
		tun->flags = flags;
1118
		tun->txflt.count = 0;
1119
		tun->vnet_hdr_sz = sizeof(struct virtio_net_hdr);
1120
		set_bit(SOCK_EXTERNALLY_ALLOCATED, &tun->socket.flags);
L
Linus Torvalds 已提交
1121

1122
		err = -ENOMEM;
1123
		sk = sk_alloc(&init_net, AF_UNSPEC, GFP_KERNEL, &tun_proto);
1124 1125 1126
		if (!sk)
			goto err_free_dev;

1127
		sk_change_net(sk, net);
1128 1129
		tun->socket.wq = &tun->wq;
		init_waitqueue_head(&tun->wq.wait);
1130
		tun->socket.ops = &tun_socket_ops;
1131 1132 1133 1134
		sock_init_data(&tun->socket, sk);
		sk->sk_write_space = tun_sock_write_space;
		sk->sk_sndbuf = INT_MAX;

1135
		tun_sk(sk)->tun = tun;
1136

P
Paul Moore 已提交
1137 1138
		security_tun_dev_post_create(sk);

L
Linus Torvalds 已提交
1139 1140
		tun_net_init(dev);

1141 1142 1143 1144
		dev->hw_features = NETIF_F_SG | NETIF_F_FRAGLIST |
			TUN_USER_FEATURES;
		dev->features = dev->hw_features;

L
Linus Torvalds 已提交
1145 1146
		err = register_netdevice(tun->dev);
		if (err < 0)
1147 1148
			goto err_free_sk;

1149 1150 1151
		if (device_create_file(&tun->dev->dev, &dev_attr_tun_flags) ||
		    device_create_file(&tun->dev->dev, &dev_attr_owner) ||
		    device_create_file(&tun->dev->dev, &dev_attr_group))
1152
			pr_err("Failed to create tun sysfs files\n");
1153

1154
		sk->sk_destruct = tun_sock_destruct;
1155 1156 1157

		err = tun_attach(tun, file);
		if (err < 0)
1158
			goto failed;
L
Linus Torvalds 已提交
1159 1160
	}

1161
	tun_debug(KERN_INFO, tun, "tun_set_iff\n");
L
Linus Torvalds 已提交
1162 1163 1164

	if (ifr->ifr_flags & IFF_NO_PI)
		tun->flags |= TUN_NO_PI;
1165 1166
	else
		tun->flags &= ~TUN_NO_PI;
L
Linus Torvalds 已提交
1167 1168 1169

	if (ifr->ifr_flags & IFF_ONE_QUEUE)
		tun->flags |= TUN_ONE_QUEUE;
1170 1171
	else
		tun->flags &= ~TUN_ONE_QUEUE;
L
Linus Torvalds 已提交
1172

1173 1174 1175 1176 1177
	if (ifr->ifr_flags & IFF_VNET_HDR)
		tun->flags |= TUN_VNET_HDR;
	else
		tun->flags &= ~TUN_VNET_HDR;

1178 1179 1180 1181 1182 1183
	/* Make sure persistent devices do not get stuck in
	 * xoff state.
	 */
	if (netif_running(tun->dev))
		netif_wake_queue(tun->dev);

L
Linus Torvalds 已提交
1184 1185 1186
	strcpy(ifr->ifr_name, tun->dev->name);
	return 0;

1187
 err_free_sk:
1188
	tun_free_netdev(dev);
L
Linus Torvalds 已提交
1189 1190 1191 1192 1193 1194
 err_free_dev:
	free_netdev(dev);
 failed:
	return err;
}

1195 1196
static int tun_get_iff(struct net *net, struct tun_struct *tun,
		       struct ifreq *ifr)
1197
{
1198
	tun_debug(KERN_INFO, tun, "tun_get_iff\n");
1199 1200 1201

	strcpy(ifr->ifr_name, tun->dev->name);

1202
	ifr->ifr_flags = tun_flags(tun);
1203 1204 1205 1206

	return 0;
}

1207 1208
/* This is like a cut-down ethtool ops, except done via tun fd so no
 * privs required. */
1209
static int set_offload(struct tun_struct *tun, unsigned long arg)
1210
{
1211
	netdev_features_t features = 0;
1212 1213

	if (arg & TUN_F_CSUM) {
1214
		features |= NETIF_F_HW_CSUM;
1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227
		arg &= ~TUN_F_CSUM;

		if (arg & (TUN_F_TSO4|TUN_F_TSO6)) {
			if (arg & TUN_F_TSO_ECN) {
				features |= NETIF_F_TSO_ECN;
				arg &= ~TUN_F_TSO_ECN;
			}
			if (arg & TUN_F_TSO4)
				features |= NETIF_F_TSO;
			if (arg & TUN_F_TSO6)
				features |= NETIF_F_TSO6;
			arg &= ~(TUN_F_TSO4|TUN_F_TSO6);
		}
1228 1229 1230 1231 1232

		if (arg & TUN_F_UFO) {
			features |= NETIF_F_UFO;
			arg &= ~TUN_F_UFO;
		}
1233 1234 1235 1236 1237 1238 1239
	}

	/* This gives the user a way to test for new features in future by
	 * trying to set them. */
	if (arg)
		return -EINVAL;

1240 1241
	tun->set_features = features;
	netdev_update_features(tun->dev);
1242 1243 1244 1245

	return 0;
}

1246 1247
static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
			    unsigned long arg, int ifreq_len)
L
Linus Torvalds 已提交
1248
{
E
Eric W. Biederman 已提交
1249
	struct tun_file *tfile = file->private_data;
E
Eric W. Biederman 已提交
1250
	struct tun_struct *tun;
L
Linus Torvalds 已提交
1251
	void __user* argp = (void __user*)arg;
M
Michael S. Tsirkin 已提交
1252
	struct sock_fprog fprog;
L
Linus Torvalds 已提交
1253
	struct ifreq ifr;
1254
	int sndbuf;
1255
	int vnet_hdr_sz;
1256
	int ret;
L
Linus Torvalds 已提交
1257 1258

	if (cmd == TUNSETIFF || _IOC_TYPE(cmd) == 0x89)
1259
		if (copy_from_user(&ifr, argp, ifreq_len))
L
Linus Torvalds 已提交
1260 1261
			return -EFAULT;

E
Eric W. Biederman 已提交
1262 1263 1264 1265 1266 1267 1268 1269 1270
	if (cmd == TUNGETFEATURES) {
		/* Currently this just means: "what IFF flags are valid?".
		 * This is needed because we never checked for invalid flags on
		 * TUNSETIFF. */
		return put_user(IFF_TUN | IFF_TAP | IFF_NO_PI | IFF_ONE_QUEUE |
				IFF_VNET_HDR,
				(unsigned int __user*)argp);
	}

1271 1272
	rtnl_lock();

E
Eric W. Biederman 已提交
1273
	tun = __tun_get(tfile);
L
Linus Torvalds 已提交
1274 1275 1276
	if (cmd == TUNSETIFF && !tun) {
		ifr.ifr_name[IFNAMSIZ-1] = '\0';

1277
		ret = tun_set_iff(tfile->net, file, &ifr);
L
Linus Torvalds 已提交
1278

1279 1280
		if (ret)
			goto unlock;
L
Linus Torvalds 已提交
1281

1282
		if (copy_to_user(argp, &ifr, ifreq_len))
1283 1284
			ret = -EFAULT;
		goto unlock;
L
Linus Torvalds 已提交
1285 1286
	}

1287
	ret = -EBADFD;
L
Linus Torvalds 已提交
1288
	if (!tun)
1289
		goto unlock;
L
Linus Torvalds 已提交
1290

1291
	tun_debug(KERN_INFO, tun, "tun_chr_ioctl cmd %d\n", cmd);
L
Linus Torvalds 已提交
1292

E
Eric W. Biederman 已提交
1293
	ret = 0;
L
Linus Torvalds 已提交
1294
	switch (cmd) {
1295
	case TUNGETIFF:
1296
		ret = tun_get_iff(current->nsproxy->net_ns, tun, &ifr);
1297
		if (ret)
E
Eric W. Biederman 已提交
1298
			break;
1299

1300
		if (copy_to_user(argp, &ifr, ifreq_len))
E
Eric W. Biederman 已提交
1301
			ret = -EFAULT;
1302 1303
		break;

L
Linus Torvalds 已提交
1304 1305 1306
	case TUNSETNOCSUM:
		/* Disable/Enable checksum */

1307 1308
		/* [unimplemented] */
		tun_debug(KERN_INFO, tun, "ignored: set checksum %s\n",
1309
			  arg ? "disabled" : "enabled");
L
Linus Torvalds 已提交
1310 1311 1312 1313 1314 1315 1316 1317 1318
		break;

	case TUNSETPERSIST:
		/* Disable/Enable persist mode */
		if (arg)
			tun->flags |= TUN_PERSIST;
		else
			tun->flags &= ~TUN_PERSIST;

1319 1320
		tun_debug(KERN_INFO, tun, "persist %s\n",
			  arg ? "enabled" : "disabled");
L
Linus Torvalds 已提交
1321 1322 1323 1324 1325 1326
		break;

	case TUNSETOWNER:
		/* Set owner of the device */
		tun->owner = (uid_t) arg;

1327
		tun_debug(KERN_INFO, tun, "owner set to %d\n", tun->owner);
L
Linus Torvalds 已提交
1328 1329
		break;

1330 1331 1332 1333
	case TUNSETGROUP:
		/* Set group of the device */
		tun->group= (gid_t) arg;

1334
		tun_debug(KERN_INFO, tun, "group set to %d\n", tun->group);
1335 1336
		break;

1337 1338 1339
	case TUNSETLINK:
		/* Only allow setting the type when the interface is down */
		if (tun->dev->flags & IFF_UP) {
1340 1341
			tun_debug(KERN_INFO, tun,
				  "Linktype set failed because interface is up\n");
1342
			ret = -EBUSY;
1343 1344
		} else {
			tun->dev->type = (int) arg;
1345 1346
			tun_debug(KERN_INFO, tun, "linktype set to %d\n",
				  tun->dev->type);
1347
			ret = 0;
1348
		}
E
Eric W. Biederman 已提交
1349
		break;
1350

L
Linus Torvalds 已提交
1351 1352 1353 1354 1355
#ifdef TUN_DEBUG
	case TUNSETDEBUG:
		tun->debug = arg;
		break;
#endif
1356
	case TUNSETOFFLOAD:
1357
		ret = set_offload(tun, arg);
E
Eric W. Biederman 已提交
1358
		break;
1359

1360 1361
	case TUNSETTXFILTER:
		/* Can be set only for TAPs */
E
Eric W. Biederman 已提交
1362
		ret = -EINVAL;
1363
		if ((tun->flags & TUN_TYPE_MASK) != TUN_TAP_DEV)
E
Eric W. Biederman 已提交
1364
			break;
H
Harvey Harrison 已提交
1365
		ret = update_filter(&tun->txflt, (void __user *)arg);
E
Eric W. Biederman 已提交
1366
		break;
L
Linus Torvalds 已提交
1367 1368

	case SIOCGIFHWADDR:
1369
		/* Get hw address */
1370 1371
		memcpy(ifr.ifr_hwaddr.sa_data, tun->dev->dev_addr, ETH_ALEN);
		ifr.ifr_hwaddr.sa_family = tun->dev->type;
1372
		if (copy_to_user(argp, &ifr, ifreq_len))
E
Eric W. Biederman 已提交
1373 1374
			ret = -EFAULT;
		break;
L
Linus Torvalds 已提交
1375 1376

	case SIOCSIFHWADDR:
1377
		/* Set hw address */
1378 1379
		tun_debug(KERN_DEBUG, tun, "set hw address: %pM\n",
			  ifr.ifr_hwaddr.sa_data);
1380 1381

		ret = dev_set_mac_address(tun->dev, &ifr.ifr_hwaddr);
E
Eric W. Biederman 已提交
1382
		break;
1383 1384

	case TUNGETSNDBUF:
1385
		sndbuf = tun->socket.sk->sk_sndbuf;
1386 1387 1388 1389 1390 1391 1392 1393 1394 1395
		if (copy_to_user(argp, &sndbuf, sizeof(sndbuf)))
			ret = -EFAULT;
		break;

	case TUNSETSNDBUF:
		if (copy_from_user(&sndbuf, argp, sizeof(sndbuf))) {
			ret = -EFAULT;
			break;
		}

1396
		tun->socket.sk->sk_sndbuf = sndbuf;
1397 1398
		break;

1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417
	case TUNGETVNETHDRSZ:
		vnet_hdr_sz = tun->vnet_hdr_sz;
		if (copy_to_user(argp, &vnet_hdr_sz, sizeof(vnet_hdr_sz)))
			ret = -EFAULT;
		break;

	case TUNSETVNETHDRSZ:
		if (copy_from_user(&vnet_hdr_sz, argp, sizeof(vnet_hdr_sz))) {
			ret = -EFAULT;
			break;
		}
		if (vnet_hdr_sz < (int)sizeof(struct virtio_net_hdr)) {
			ret = -EINVAL;
			break;
		}

		tun->vnet_hdr_sz = vnet_hdr_sz;
		break;

M
Michael S. Tsirkin 已提交
1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437
	case TUNATTACHFILTER:
		/* Can be set only for TAPs */
		ret = -EINVAL;
		if ((tun->flags & TUN_TYPE_MASK) != TUN_TAP_DEV)
			break;
		ret = -EFAULT;
		if (copy_from_user(&fprog, argp, sizeof(fprog)))
			break;

		ret = sk_attach_filter(&fprog, tun->socket.sk);
		break;

	case TUNDETACHFILTER:
		/* Can be set only for TAPs */
		ret = -EINVAL;
		if ((tun->flags & TUN_TYPE_MASK) != TUN_TAP_DEV)
			break;
		ret = sk_detach_filter(tun->socket.sk);
		break;

L
Linus Torvalds 已提交
1438
	default:
E
Eric W. Biederman 已提交
1439 1440
		ret = -EINVAL;
		break;
1441
	}
L
Linus Torvalds 已提交
1442

1443 1444 1445 1446
unlock:
	rtnl_unlock();
	if (tun)
		tun_put(tun);
E
Eric W. Biederman 已提交
1447
	return ret;
L
Linus Torvalds 已提交
1448 1449
}

1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484
static long tun_chr_ioctl(struct file *file,
			  unsigned int cmd, unsigned long arg)
{
	return __tun_chr_ioctl(file, cmd, arg, sizeof (struct ifreq));
}

#ifdef CONFIG_COMPAT
static long tun_chr_compat_ioctl(struct file *file,
			 unsigned int cmd, unsigned long arg)
{
	switch (cmd) {
	case TUNSETIFF:
	case TUNGETIFF:
	case TUNSETTXFILTER:
	case TUNGETSNDBUF:
	case TUNSETSNDBUF:
	case SIOCGIFHWADDR:
	case SIOCSIFHWADDR:
		arg = (unsigned long)compat_ptr(arg);
		break;
	default:
		arg = (compat_ulong_t)arg;
		break;
	}

	/*
	 * compat_ifreq is shorter than ifreq, so we must not access beyond
	 * the end of that structure. All fields that are used in this
	 * driver are compatible though, we don't need to convert the
	 * contents.
	 */
	return __tun_chr_ioctl(file, cmd, arg, sizeof(struct compat_ifreq));
}
#endif /* CONFIG_COMPAT */

L
Linus Torvalds 已提交
1485 1486
static int tun_chr_fasync(int fd, struct file *file, int on)
{
E
Eric W. Biederman 已提交
1487
	struct tun_struct *tun = tun_get(file);
L
Linus Torvalds 已提交
1488 1489 1490 1491 1492
	int ret;

	if (!tun)
		return -EBADFD;

1493
	tun_debug(KERN_INFO, tun, "tun_chr_fasync %d\n", on);
L
Linus Torvalds 已提交
1494 1495

	if ((ret = fasync_helper(fd, file, on, &tun->fasync)) < 0)
J
Jonathan Corbet 已提交
1496
		goto out;
1497

L
Linus Torvalds 已提交
1498
	if (on) {
1499
		ret = __f_setown(file, task_pid(current), PIDTYPE_PID, 0);
L
Linus Torvalds 已提交
1500
		if (ret)
J
Jonathan Corbet 已提交
1501
			goto out;
L
Linus Torvalds 已提交
1502
		tun->flags |= TUN_FASYNC;
1503
	} else
L
Linus Torvalds 已提交
1504
		tun->flags &= ~TUN_FASYNC;
J
Jonathan Corbet 已提交
1505 1506
	ret = 0;
out:
E
Eric W. Biederman 已提交
1507
	tun_put(tun);
J
Jonathan Corbet 已提交
1508
	return ret;
L
Linus Torvalds 已提交
1509 1510 1511 1512
}

static int tun_chr_open(struct inode *inode, struct file * file)
{
E
Eric W. Biederman 已提交
1513
	struct tun_file *tfile;
T
Thomas Gleixner 已提交
1514

1515
	DBG1(KERN_INFO, "tunX: tun_chr_open\n");
E
Eric W. Biederman 已提交
1516 1517 1518 1519

	tfile = kmalloc(sizeof(*tfile), GFP_KERNEL);
	if (!tfile)
		return -ENOMEM;
1520
	atomic_set(&tfile->count, 0);
E
Eric W. Biederman 已提交
1521
	tfile->tun = NULL;
E
Eric W. Biederman 已提交
1522
	tfile->net = get_net(current->nsproxy->net_ns);
E
Eric W. Biederman 已提交
1523
	file->private_data = tfile;
L
Linus Torvalds 已提交
1524 1525 1526 1527 1528
	return 0;
}

static int tun_chr_close(struct inode *inode, struct file *file)
{
E
Eric W. Biederman 已提交
1529
	struct tun_file *tfile = file->private_data;
E
Eric W. Biederman 已提交
1530
	struct tun_struct *tun;
L
Linus Torvalds 已提交
1531

E
Eric W. Biederman 已提交
1532
	tun = __tun_get(tfile);
E
Eric W. Biederman 已提交
1533
	if (tun) {
H
Herbert Xu 已提交
1534 1535
		struct net_device *dev = tun->dev;

1536
		tun_debug(KERN_INFO, tun, "tun_chr_close\n");
L
Linus Torvalds 已提交
1537

E
Eric W. Biederman 已提交
1538
		__tun_detach(tun);
L
Linus Torvalds 已提交
1539

D
Daniel Mack 已提交
1540
		/* If desirable, unregister the netdevice. */
H
Herbert Xu 已提交
1541 1542 1543 1544 1545 1546
		if (!(tun->flags & TUN_PERSIST)) {
			rtnl_lock();
			if (dev->reg_state == NETREG_REGISTERED)
				unregister_netdevice(dev);
			rtnl_unlock();
		}
E
Eric W. Biederman 已提交
1547
	}
L
Linus Torvalds 已提交
1548

1549 1550
	tun = tfile->tun;
	if (tun)
1551
		sock_put(tun->socket.sk);
1552

E
Eric W. Biederman 已提交
1553
	put_net(tfile->net);
E
Eric W. Biederman 已提交
1554
	kfree(tfile);
L
Linus Torvalds 已提交
1555 1556 1557 1558

	return 0;
}

1559
static const struct file_operations tun_fops = {
1560
	.owner	= THIS_MODULE,
L
Linus Torvalds 已提交
1561
	.llseek = no_llseek,
1562 1563 1564 1565
	.read  = do_sync_read,
	.aio_read  = tun_chr_aio_read,
	.write = do_sync_write,
	.aio_write = tun_chr_aio_write,
L
Linus Torvalds 已提交
1566
	.poll	= tun_chr_poll,
1567 1568 1569 1570
	.unlocked_ioctl	= tun_chr_ioctl,
#ifdef CONFIG_COMPAT
	.compat_ioctl = tun_chr_compat_ioctl,
#endif
L
Linus Torvalds 已提交
1571 1572
	.open	= tun_chr_open,
	.release = tun_chr_close,
1573
	.fasync = tun_chr_fasync
L
Linus Torvalds 已提交
1574 1575 1576 1577 1578
};

static struct miscdevice tun_miscdev = {
	.minor = TUN_MINOR,
	.name = "tun",
1579
	.nodename = "net/tun",
L
Linus Torvalds 已提交
1580 1581 1582 1583 1584 1585 1586 1587 1588
	.fops = &tun_fops,
};

/* ethtool interface */

static int tun_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
	cmd->supported		= 0;
	cmd->advertising	= 0;
1589
	ethtool_cmd_speed_set(cmd, SPEED_10);
L
Linus Torvalds 已提交
1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603
	cmd->duplex		= DUPLEX_FULL;
	cmd->port		= PORT_TP;
	cmd->phy_address	= 0;
	cmd->transceiver	= XCVR_INTERNAL;
	cmd->autoneg		= AUTONEG_DISABLE;
	cmd->maxtxpkt		= 0;
	cmd->maxrxpkt		= 0;
	return 0;
}

static void tun_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
{
	struct tun_struct *tun = netdev_priv(dev);

1604 1605
	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
L
Linus Torvalds 已提交
1606 1607 1608

	switch (tun->flags & TUN_TYPE_MASK) {
	case TUN_TUN_DEV:
1609
		strlcpy(info->bus_info, "tun", sizeof(info->bus_info));
L
Linus Torvalds 已提交
1610 1611
		break;
	case TUN_TAP_DEV:
1612
		strlcpy(info->bus_info, "tap", sizeof(info->bus_info));
L
Linus Torvalds 已提交
1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634
		break;
	}
}

static u32 tun_get_msglevel(struct net_device *dev)
{
#ifdef TUN_DEBUG
	struct tun_struct *tun = netdev_priv(dev);
	return tun->debug;
#else
	return -EOPNOTSUPP;
#endif
}

static void tun_set_msglevel(struct net_device *dev, u32 value)
{
#ifdef TUN_DEBUG
	struct tun_struct *tun = netdev_priv(dev);
	tun->debug = value;
#endif
}

1635
static const struct ethtool_ops tun_ethtool_ops = {
L
Linus Torvalds 已提交
1636 1637 1638 1639
	.get_settings	= tun_get_settings,
	.get_drvinfo	= tun_get_drvinfo,
	.get_msglevel	= tun_get_msglevel,
	.set_msglevel	= tun_set_msglevel,
1640
	.get_link	= ethtool_op_get_link,
L
Linus Torvalds 已提交
1641 1642
};

1643

L
Linus Torvalds 已提交
1644 1645 1646 1647
static int __init tun_init(void)
{
	int ret = 0;

1648 1649
	pr_info("%s, %s\n", DRV_DESCRIPTION, DRV_VERSION);
	pr_info("%s\n", DRV_COPYRIGHT);
L
Linus Torvalds 已提交
1650

1651
	ret = rtnl_link_register(&tun_link_ops);
1652
	if (ret) {
1653
		pr_err("Can't register link_ops\n");
1654
		goto err_linkops;
1655 1656
	}

L
Linus Torvalds 已提交
1657
	ret = misc_register(&tun_miscdev);
1658
	if (ret) {
1659
		pr_err("Can't register misc device %d\n", TUN_MINOR);
1660 1661
		goto err_misc;
	}
1662
	return  0;
1663
err_misc:
1664 1665
	rtnl_link_unregister(&tun_link_ops);
err_linkops:
L
Linus Torvalds 已提交
1666 1667 1668 1669 1670
	return ret;
}

static void tun_cleanup(void)
{
1671
	misc_deregister(&tun_miscdev);
1672
	rtnl_link_unregister(&tun_link_ops);
L
Linus Torvalds 已提交
1673 1674
}

1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691
/* Get an underlying socket object from tun file.  Returns error unless file is
 * attached to a device.  The returned object works like a packet socket, it
 * can be used for sock_sendmsg/sock_recvmsg.  The caller is responsible for
 * holding a reference to the file for as long as the socket is in use. */
struct socket *tun_get_socket(struct file *file)
{
	struct tun_struct *tun;
	if (file->f_op != &tun_fops)
		return ERR_PTR(-EINVAL);
	tun = tun_get(file);
	if (!tun)
		return ERR_PTR(-EBADFD);
	tun_put(tun);
	return &tun->socket;
}
EXPORT_SYMBOL_GPL(tun_get_socket);

L
Linus Torvalds 已提交
1692 1693 1694 1695 1696 1697
module_init(tun_init);
module_exit(tun_cleanup);
MODULE_DESCRIPTION(DRV_DESCRIPTION);
MODULE_AUTHOR(DRV_COPYRIGHT);
MODULE_LICENSE("GPL");
MODULE_ALIAS_MISCDEV(TUN_MINOR);
1698
MODULE_ALIAS("devname:net/tun");