tun.c 38.4 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>
25
 *    Use random_ether_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
	}
288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337

	/* 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++)
		if (!compare_ether_addr(eh->h_dest, filter->addr[i]))
			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
	sk_release_kernel(tun->socket.sk);
362 363
}

L
Linus Torvalds 已提交
364 365 366 367 368 369 370 371 372 373 374 375 376 377 378
/* 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 */
379
static netdev_tx_t tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
L
Linus Torvalds 已提交
380 381 382
{
	struct tun_struct *tun = netdev_priv(dev);

383
	tun_debug(KERN_INFO, tun, "tun_net_xmit %d\n", skb->len);
L
Linus Torvalds 已提交
384 385

	/* Drop packet if interface is not attached */
E
Eric W. Biederman 已提交
386
	if (!tun->tfile)
L
Linus Torvalds 已提交
387 388
		goto drop;

389 390 391 392 393 394
	/* 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
395 396 397 398
	if (tun->socket.sk->sk_filter &&
	    sk_filter(tun->socket.sk, skb))
		goto drop;

399
	if (skb_queue_len(&tun->socket.sk->sk_receive_queue) >= dev->tx_queue_len) {
L
Linus Torvalds 已提交
400 401 402 403 404 405 406
		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. */
407
			dev->stats.tx_fifo_errors++;
L
Linus Torvalds 已提交
408 409 410 411 412 413 414
		} else {
			/* Single queue mode.
			 * Driver handles dropping of all packets itself. */
			goto drop;
		}
	}

M
Michael S. Tsirkin 已提交
415 416 417 418
	/* Orphan the skb - required as we might hang on to it
	 * for indefinite time. */
	skb_orphan(skb);

419
	/* Enqueue packet */
420
	skb_queue_tail(&tun->socket.sk->sk_receive_queue, skb);
L
Linus Torvalds 已提交
421 422 423 424

	/* Notify and wake up reader process */
	if (tun->flags & TUN_FASYNC)
		kill_fasync(&tun->fasync, SIGIO, POLL_IN);
425
	wake_up_interruptible_poll(&tun->wq.wait, POLLIN |
426
				   POLLRDNORM | POLLRDBAND);
427
	return NETDEV_TX_OK;
L
Linus Torvalds 已提交
428 429

drop:
430
	dev->stats.tx_dropped++;
L
Linus Torvalds 已提交
431
	kfree_skb(skb);
432
	return NETDEV_TX_OK;
L
Linus Torvalds 已提交
433 434
}

435
static void tun_net_mclist(struct net_device *dev)
L
Linus Torvalds 已提交
436
{
437 438 439 440 441
	/*
	 * 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
442 443
}

E
Ed Swierk 已提交
444 445 446 447 448 449 450 451 452 453 454 455
#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;
}

456 457
static netdev_features_t tun_net_fix_features(struct net_device *dev,
	netdev_features_t features)
458 459 460 461 462
{
	struct tun_struct *tun = netdev_priv(dev);

	return (features & tun->set_features) | (features & ~TUN_USER_FEATURES);
}
463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479
#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
480
static const struct net_device_ops tun_netdev_ops = {
481
	.ndo_uninit		= tun_net_uninit,
S
Stephen Hemminger 已提交
482 483
	.ndo_open		= tun_net_open,
	.ndo_stop		= tun_net_close,
484
	.ndo_start_xmit		= tun_net_xmit,
S
Stephen Hemminger 已提交
485
	.ndo_change_mtu		= tun_net_change_mtu,
486
	.ndo_fix_features	= tun_net_fix_features,
487 488 489
#ifdef CONFIG_NET_POLL_CONTROLLER
	.ndo_poll_controller	= tun_poll_controller,
#endif
S
Stephen Hemminger 已提交
490 491 492
};

static const struct net_device_ops tap_netdev_ops = {
493
	.ndo_uninit		= tun_net_uninit,
S
Stephen Hemminger 已提交
494 495
	.ndo_open		= tun_net_open,
	.ndo_stop		= tun_net_close,
496
	.ndo_start_xmit		= tun_net_xmit,
S
Stephen Hemminger 已提交
497
	.ndo_change_mtu		= tun_net_change_mtu,
498
	.ndo_fix_features	= tun_net_fix_features,
499
	.ndo_set_rx_mode	= tun_net_mclist,
S
Stephen Hemminger 已提交
500 501
	.ndo_set_mac_address	= eth_mac_addr,
	.ndo_validate_addr	= eth_validate_addr,
502 503 504
#ifdef CONFIG_NET_POLL_CONTROLLER
	.ndo_poll_controller	= tun_poll_controller,
#endif
S
Stephen Hemminger 已提交
505 506
};

L
Linus Torvalds 已提交
507 508 509 510
/* Initialize net device. */
static void tun_net_init(struct net_device *dev)
{
	struct tun_struct *tun = netdev_priv(dev);
511

L
Linus Torvalds 已提交
512 513
	switch (tun->flags & TUN_TYPE_MASK) {
	case TUN_TUN_DEV:
S
Stephen Hemminger 已提交
514 515
		dev->netdev_ops = &tun_netdev_ops;

L
Linus Torvalds 已提交
516 517 518 519 520 521
		/* Point-to-Point TUN Device */
		dev->hard_header_len = 0;
		dev->addr_len = 0;
		dev->mtu = 1500;

		/* Zero header length */
522
		dev->type = ARPHRD_NONE;
L
Linus Torvalds 已提交
523 524 525 526 527
		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
528
		dev->netdev_ops = &tap_netdev_ops;
L
Linus Torvalds 已提交
529 530
		/* Ethernet TAP Device */
		ether_setup(dev);
531
		dev->priv_flags &= ~IFF_TX_SKB_SHARING;
532

533
		eth_hw_addr_random(dev);
534

L
Linus Torvalds 已提交
535 536 537 538 539 540 541 542 543
		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)
544
{
545 546
	struct tun_file *tfile = file->private_data;
	struct tun_struct *tun = __tun_get(tfile);
547
	struct sock *sk;
548
	unsigned int mask = 0;
L
Linus Torvalds 已提交
549 550

	if (!tun)
551
		return POLLERR;
L
Linus Torvalds 已提交
552

553
	sk = tun->socket.sk;
554

555
	tun_debug(KERN_INFO, tun, "tun_chr_poll\n");
L
Linus Torvalds 已提交
556

557
	poll_wait(file, &tun->wq.wait, wait);
558

559
	if (!skb_queue_empty(&sk->sk_receive_queue))
L
Linus Torvalds 已提交
560 561
		mask |= POLLIN | POLLRDNORM;

562 563 564 565 566
	if (sock_writeable(sk) ||
	    (!test_and_set_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags) &&
	     sock_writeable(sk)))
		mask |= POLLOUT | POLLWRNORM;

567 568 569
	if (tun->dev->reg_state != NETREG_REGISTERED)
		mask = POLLERR;

E
Eric W. Biederman 已提交
570
	tun_put(tun);
L
Linus Torvalds 已提交
571 572 573
	return mask;
}

574 575
/* 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). */
576 577 578
static struct sk_buff *tun_alloc_skb(struct tun_struct *tun,
				     size_t prepad, size_t len,
				     size_t linear, int noblock)
579
{
580
	struct sock *sk = tun->socket.sk;
581
	struct sk_buff *skb;
582
	int err;
583

584 585
	sock_update_classid(sk);

586
	/* Under a page?  Don't bother with paged skb. */
H
Herbert Xu 已提交
587
	if (prepad + len < PAGE_SIZE || !linear)
588
		linear = len;
589

590 591
	skb = sock_alloc_send_pskb(sk, prepad + linear, len - linear, noblock,
				   &err);
592
	if (!skb)
593
		return ERR_PTR(err);
594 595 596

	skb_reserve(skb, prepad);
	skb_put(skb, linear);
597 598
	skb->data_len = len - linear;
	skb->len += len - linear;
599 600 601 602

	return skb;
}

L
Linus Torvalds 已提交
603
/* Get packet from user space buffer */
604 605 606
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
607
{
608
	struct tun_pi pi = { 0, cpu_to_be16(ETH_P_IP) };
L
Linus Torvalds 已提交
609
	struct sk_buff *skb;
610
	size_t len = count, align = NET_SKB_PAD;
611
	struct virtio_net_hdr gso = { 0 };
612
	int offset = 0;
L
Linus Torvalds 已提交
613 614 615 616 617

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

618
		if (memcpy_fromiovecend((void *)&pi, iv, 0, sizeof(pi)))
L
Linus Torvalds 已提交
619
			return -EFAULT;
620
		offset += sizeof(pi);
L
Linus Torvalds 已提交
621 622
	}

623
	if (tun->flags & TUN_VNET_HDR) {
624
		if ((len -= tun->vnet_hdr_sz) > count)
625 626
			return -EINVAL;

627
		if (memcpy_fromiovecend((void *)&gso, iv, offset, sizeof(gso)))
628 629
			return -EFAULT;

630 631 632 633
		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;

634 635
		if (gso.hdr_len > len)
			return -EINVAL;
636
		offset += tun->vnet_hdr_sz;
637 638
	}

639
	if ((tun->flags & TUN_TYPE_MASK) == TUN_TAP_DEV) {
640
		align += NET_IP_ALIGN;
H
Herbert Xu 已提交
641 642
		if (unlikely(len < ETH_HLEN ||
			     (gso.hdr_len && gso.hdr_len < ETH_HLEN)))
643 644
			return -EINVAL;
	}
645

646 647 648 649 650
	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
651 652
	}

653
	if (skb_copy_datagram_from_iovec(skb, 0, iv, offset, len)) {
654
		tun->dev->stats.rx_dropped++;
D
Dave Jones 已提交
655
		kfree_skb(skb);
L
Linus Torvalds 已提交
656
		return -EFAULT;
D
Dave Jones 已提交
657
	}
L
Linus Torvalds 已提交
658

659 660 661 662 663 664 665
	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;
		}
666
	}
667

L
Linus Torvalds 已提交
668 669
	switch (tun->flags & TUN_TYPE_MASK) {
	case TUN_TUN_DEV:
670 671 672 673 674 675 676 677 678 679 680 681 682 683 684
		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;
			}
		}

685
		skb_reset_mac_header(skb);
L
Linus Torvalds 已提交
686
		skb->protocol = pi.proto;
687
		skb->dev = tun->dev;
L
Linus Torvalds 已提交
688 689 690 691
		break;
	case TUN_TAP_DEV:
		skb->protocol = eth_type_trans(skb, tun->dev);
		break;
692
	}
L
Linus Torvalds 已提交
693

694 695 696 697 698 699 700 701 702
	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;
703 704 705
		case VIRTIO_NET_HDR_GSO_UDP:
			skb_shinfo(skb)->gso_type = SKB_GSO_UDP;
			break;
706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725
		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;
	}
726

L
Linus Torvalds 已提交
727
	netif_rx_ni(skb);
728

729 730
	tun->dev->stats.rx_packets++;
	tun->dev->stats.rx_bytes += len;
L
Linus Torvalds 已提交
731 732

	return count;
733
}
L
Linus Torvalds 已提交
734

735 736
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
737
{
738
	struct file *file = iocb->ki_filp;
H
Herbert Xu 已提交
739
	struct tun_struct *tun = tun_get(file);
E
Eric W. Biederman 已提交
740
	ssize_t result;
L
Linus Torvalds 已提交
741 742 743 744

	if (!tun)
		return -EBADFD;

745
	tun_debug(KERN_INFO, tun, "tun_chr_write %ld\n", count);
L
Linus Torvalds 已提交
746

747
	result = tun_get_user(tun, iv, iov_length(iv, count),
748
			      file->f_flags & O_NONBLOCK);
E
Eric W. Biederman 已提交
749 750 751

	tun_put(tun);
	return result;
L
Linus Torvalds 已提交
752 753 754
}

/* Put packet to the user space buffer */
755 756 757
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
758 759 760 761 762 763 764 765 766 767 768 769
{
	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;
		}
770

771
		if (memcpy_toiovecend(iv, (void *) &pi, 0, sizeof(pi)))
L
Linus Torvalds 已提交
772 773
			return -EFAULT;
		total += sizeof(pi);
774
	}
L
Linus Torvalds 已提交
775

776 777
	if (tun->flags & TUN_VNET_HDR) {
		struct virtio_net_hdr gso = { 0 }; /* no info leak */
778
		if ((len -= tun->vnet_hdr_sz) < 0)
779 780 781 782 783 784 785 786 787 788 789 790
			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;
791 792
			else if (sinfo->gso_type & SKB_GSO_UDP)
				gso.gso_type = VIRTIO_NET_HDR_GSO_UDP;
793
			else {
794
				pr_err("unexpected GSO type: "
795 796 797 798 799 800 801 802 803 804
				       "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;
			}
805 806 807 808 809 810 811
			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;
812
			gso.csum_start = skb_checksum_start_offset(skb);
813
			gso.csum_offset = skb->csum_offset;
814 815
		} else if (skb->ip_summed == CHECKSUM_UNNECESSARY) {
			gso.flags = VIRTIO_NET_HDR_F_DATA_VALID;
816 817
		} /* else everything is zero */

818 819
		if (unlikely(memcpy_toiovecend(iv, (void *)&gso, total,
					       sizeof(gso))))
820
			return -EFAULT;
821
		total += tun->vnet_hdr_sz;
822 823
	}

L
Linus Torvalds 已提交
824 825
	len = min_t(int, skb->len, len);

826
	skb_copy_datagram_const_iovec(skb, 0, iv, total, len);
827
	total += skb->len;
L
Linus Torvalds 已提交
828

829 830
	tun->dev->stats.tx_packets++;
	tun->dev->stats.tx_bytes += len;
L
Linus Torvalds 已提交
831 832 833 834

	return total;
}

835 836 837
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
838 839 840
{
	DECLARE_WAITQUEUE(wait, current);
	struct sk_buff *skb;
841
	ssize_t ret = 0;
L
Linus Torvalds 已提交
842

843
	tun_debug(KERN_INFO, tun, "tun_chr_read\n");
L
Linus Torvalds 已提交
844

845 846
	if (unlikely(!noblock))
		add_wait_queue(&tun->wq.wait, &wait);
L
Linus Torvalds 已提交
847 848 849 850
	while (len) {
		current->state = TASK_INTERRUPTIBLE;

		/* Read frames from the queue */
851
		if (!(skb=skb_dequeue(&tun->socket.sk->sk_receive_queue))) {
852
			if (noblock) {
L
Linus Torvalds 已提交
853 854 855 856 857 858 859
				ret = -EAGAIN;
				break;
			}
			if (signal_pending(current)) {
				ret = -ERESTARTSYS;
				break;
			}
860 861 862 863
			if (tun->dev->reg_state != NETREG_REGISTERED) {
				ret = -EIO;
				break;
			}
L
Linus Torvalds 已提交
864 865 866 867 868 869 870

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

871
		ret = tun_put_user(tun, skb, iv, len);
872 873
		kfree_skb(skb);
		break;
L
Linus Torvalds 已提交
874 875 876
	}

	current->state = TASK_RUNNING;
877 878
	if (unlikely(!noblock))
		remove_wait_queue(&tun->wq.wait, &wait);
L
Linus Torvalds 已提交
879

880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900
	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
901 902
out:
	tun_put(tun);
L
Linus Torvalds 已提交
903 904 905 906 907 908 909 910
	return ret;
}

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

	tun->owner = -1;
911
	tun->group = -1;
L
Linus Torvalds 已提交
912 913

	dev->ethtool_ops = &tun_ethtool_ops;
914
	dev->destructor = tun_free_netdev;
L
Linus Torvalds 已提交
915 916
}

917 918 919 920 921 922 923 924 925 926 927 928 929 930 931
/* 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,
};

932 933 934
static void tun_sock_write_space(struct sock *sk)
{
	struct tun_struct *tun;
935
	wait_queue_head_t *wqueue;
936 937 938 939 940 941 942

	if (!sock_writeable(sk))
		return;

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

943 944 945
	wqueue = sk_sleep(sk);
	if (wqueue && waitqueue_active(wqueue))
		wake_up_interruptible_sync_poll(wqueue, POLLOUT |
946
						POLLWRNORM | POLLWRBAND);
H
Herbert Xu 已提交
947

948
	tun = tun_sk(sk)->tun;
949 950 951 952 953
	kill_fasync(&tun->fasync, SIGIO, POLL_OUT);
}

static void tun_sock_destruct(struct sock *sk)
{
954
	free_netdev(tun_sk(sk)->tun->dev);
955 956
}

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
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;
}

982 983 984 985 986 987 988
static int tun_release(struct socket *sock)
{
	if (sock->sk)
		sock_put(sock->sk);
	return 0;
}

989 990 991 992
/* Ops structure to mimic raw sockets with tun */
static const struct proto_ops tun_socket_ops = {
	.sendmsg = tun_sendmsg,
	.recvmsg = tun_recvmsg,
993
	.release = tun_release,
994 995
};

996 997 998 999 1000
static struct proto tun_proto = {
	.name		= "tun",
	.owner		= THIS_MODULE,
	.obj_size	= sizeof(struct tun_sock),
};
1001

1002 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
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);

1048
static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
L
Linus Torvalds 已提交
1049
{
1050
	struct sock *sk;
L
Linus Torvalds 已提交
1051 1052 1053 1054
	struct tun_struct *tun;
	struct net_device *dev;
	int err;

1055 1056
	dev = __dev_get_by_name(net, ifr->ifr_name);
	if (dev) {
P
Paul Moore 已提交
1057 1058
		const struct cred *cred = current_cred();

1059 1060
		if (ifr->ifr_flags & IFF_TUN_EXCL)
			return -EBUSY;
1061 1062 1063 1064 1065 1066 1067
		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
1068 1069 1070 1071
		if (((tun->owner != -1 && cred->euid != tun->owner) ||
		     (tun->group != -1 && !in_egroup_p(tun->group))) &&
		    !capable(CAP_NET_ADMIN))
			return -EPERM;
1072
		err = security_tun_dev_attach(tun->socket.sk);
P
Paul Moore 已提交
1073 1074 1075
		if (err < 0)
			return err;

1076 1077 1078
		err = tun_attach(tun, file);
		if (err < 0)
			return err;
1079
	}
L
Linus Torvalds 已提交
1080 1081 1082 1083
	else {
		char *name;
		unsigned long flags = 0;

1084 1085
		if (!capable(CAP_NET_ADMIN))
			return -EPERM;
P
Paul Moore 已提交
1086 1087 1088
		err = security_tun_dev_create();
		if (err < 0)
			return err;
1089

L
Linus Torvalds 已提交
1090 1091 1092 1093 1094 1095 1096 1097 1098
		/* 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";
1099
		} else
1100
			return -EINVAL;
1101

L
Linus Torvalds 已提交
1102 1103 1104 1105 1106 1107 1108 1109
		if (*ifr->ifr_name)
			name = ifr->ifr_name;

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

1110
		dev_net_set(dev, net);
1111
		dev->rtnl_link_ops = &tun_link_ops;
S
Stephen Hemminger 已提交
1112

L
Linus Torvalds 已提交
1113 1114 1115
		tun = netdev_priv(dev);
		tun->dev = dev;
		tun->flags = flags;
1116
		tun->txflt.count = 0;
1117
		tun->vnet_hdr_sz = sizeof(struct virtio_net_hdr);
L
Linus Torvalds 已提交
1118

1119
		err = -ENOMEM;
1120
		sk = sk_alloc(&init_net, AF_UNSPEC, GFP_KERNEL, &tun_proto);
1121 1122 1123
		if (!sk)
			goto err_free_dev;

1124
		sk_change_net(sk, net);
1125 1126
		tun->socket.wq = &tun->wq;
		init_waitqueue_head(&tun->wq.wait);
1127
		tun->socket.ops = &tun_socket_ops;
1128 1129 1130 1131
		sock_init_data(&tun->socket, sk);
		sk->sk_write_space = tun_sock_write_space;
		sk->sk_sndbuf = INT_MAX;

1132
		tun_sk(sk)->tun = tun;
1133

P
Paul Moore 已提交
1134 1135
		security_tun_dev_post_create(sk);

L
Linus Torvalds 已提交
1136 1137
		tun_net_init(dev);

1138 1139 1140 1141
		dev->hw_features = NETIF_F_SG | NETIF_F_FRAGLIST |
			TUN_USER_FEATURES;
		dev->features = dev->hw_features;

L
Linus Torvalds 已提交
1142 1143
		err = register_netdevice(tun->dev);
		if (err < 0)
1144 1145
			goto err_free_sk;

1146 1147 1148
		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))
1149
			pr_err("Failed to create tun sysfs files\n");
1150

1151
		sk->sk_destruct = tun_sock_destruct;
1152 1153 1154

		err = tun_attach(tun, file);
		if (err < 0)
1155
			goto failed;
L
Linus Torvalds 已提交
1156 1157
	}

1158
	tun_debug(KERN_INFO, tun, "tun_set_iff\n");
L
Linus Torvalds 已提交
1159 1160 1161

	if (ifr->ifr_flags & IFF_NO_PI)
		tun->flags |= TUN_NO_PI;
1162 1163
	else
		tun->flags &= ~TUN_NO_PI;
L
Linus Torvalds 已提交
1164 1165 1166

	if (ifr->ifr_flags & IFF_ONE_QUEUE)
		tun->flags |= TUN_ONE_QUEUE;
1167 1168
	else
		tun->flags &= ~TUN_ONE_QUEUE;
L
Linus Torvalds 已提交
1169

1170 1171 1172 1173 1174
	if (ifr->ifr_flags & IFF_VNET_HDR)
		tun->flags |= TUN_VNET_HDR;
	else
		tun->flags &= ~TUN_VNET_HDR;

1175 1176 1177 1178 1179 1180
	/* 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
1181 1182 1183
	strcpy(ifr->ifr_name, tun->dev->name);
	return 0;

1184
 err_free_sk:
1185
	tun_free_netdev(dev);
L
Linus Torvalds 已提交
1186 1187 1188 1189 1190 1191
 err_free_dev:
	free_netdev(dev);
 failed:
	return err;
}

1192 1193
static int tun_get_iff(struct net *net, struct tun_struct *tun,
		       struct ifreq *ifr)
1194
{
1195
	tun_debug(KERN_INFO, tun, "tun_get_iff\n");
1196 1197 1198

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

1199
	ifr->ifr_flags = tun_flags(tun);
1200 1201 1202 1203

	return 0;
}

1204 1205
/* This is like a cut-down ethtool ops, except done via tun fd so no
 * privs required. */
1206
static int set_offload(struct tun_struct *tun, unsigned long arg)
1207
{
1208
	netdev_features_t features = 0;
1209 1210

	if (arg & TUN_F_CSUM) {
1211
		features |= NETIF_F_HW_CSUM;
1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224
		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);
		}
1225 1226 1227 1228 1229

		if (arg & TUN_F_UFO) {
			features |= NETIF_F_UFO;
			arg &= ~TUN_F_UFO;
		}
1230 1231 1232 1233 1234 1235 1236
	}

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

1237 1238
	tun->set_features = features;
	netdev_update_features(tun->dev);
1239 1240 1241 1242

	return 0;
}

1243 1244
static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
			    unsigned long arg, int ifreq_len)
L
Linus Torvalds 已提交
1245
{
E
Eric W. Biederman 已提交
1246
	struct tun_file *tfile = file->private_data;
E
Eric W. Biederman 已提交
1247
	struct tun_struct *tun;
L
Linus Torvalds 已提交
1248
	void __user* argp = (void __user*)arg;
M
Michael S. Tsirkin 已提交
1249
	struct sock_fprog fprog;
L
Linus Torvalds 已提交
1250
	struct ifreq ifr;
1251
	int sndbuf;
1252
	int vnet_hdr_sz;
1253
	int ret;
L
Linus Torvalds 已提交
1254 1255

	if (cmd == TUNSETIFF || _IOC_TYPE(cmd) == 0x89)
1256
		if (copy_from_user(&ifr, argp, ifreq_len))
L
Linus Torvalds 已提交
1257 1258
			return -EFAULT;

E
Eric W. Biederman 已提交
1259 1260 1261 1262 1263 1264 1265 1266 1267
	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);
	}

1268 1269
	rtnl_lock();

E
Eric W. Biederman 已提交
1270
	tun = __tun_get(tfile);
L
Linus Torvalds 已提交
1271 1272 1273
	if (cmd == TUNSETIFF && !tun) {
		ifr.ifr_name[IFNAMSIZ-1] = '\0';

1274
		ret = tun_set_iff(tfile->net, file, &ifr);
L
Linus Torvalds 已提交
1275

1276 1277
		if (ret)
			goto unlock;
L
Linus Torvalds 已提交
1278

1279
		if (copy_to_user(argp, &ifr, ifreq_len))
1280 1281
			ret = -EFAULT;
		goto unlock;
L
Linus Torvalds 已提交
1282 1283
	}

1284
	ret = -EBADFD;
L
Linus Torvalds 已提交
1285
	if (!tun)
1286
		goto unlock;
L
Linus Torvalds 已提交
1287

1288
	tun_debug(KERN_INFO, tun, "tun_chr_ioctl cmd %d\n", cmd);
L
Linus Torvalds 已提交
1289

E
Eric W. Biederman 已提交
1290
	ret = 0;
L
Linus Torvalds 已提交
1291
	switch (cmd) {
1292
	case TUNGETIFF:
1293
		ret = tun_get_iff(current->nsproxy->net_ns, tun, &ifr);
1294
		if (ret)
E
Eric W. Biederman 已提交
1295
			break;
1296

1297
		if (copy_to_user(argp, &ifr, ifreq_len))
E
Eric W. Biederman 已提交
1298
			ret = -EFAULT;
1299 1300
		break;

L
Linus Torvalds 已提交
1301 1302 1303
	case TUNSETNOCSUM:
		/* Disable/Enable checksum */

1304 1305
		/* [unimplemented] */
		tun_debug(KERN_INFO, tun, "ignored: set checksum %s\n",
1306
			  arg ? "disabled" : "enabled");
L
Linus Torvalds 已提交
1307 1308 1309 1310 1311 1312 1313 1314 1315
		break;

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

1316 1317
		tun_debug(KERN_INFO, tun, "persist %s\n",
			  arg ? "enabled" : "disabled");
L
Linus Torvalds 已提交
1318 1319 1320 1321 1322 1323
		break;

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

1324
		tun_debug(KERN_INFO, tun, "owner set to %d\n", tun->owner);
L
Linus Torvalds 已提交
1325 1326
		break;

1327 1328 1329 1330
	case TUNSETGROUP:
		/* Set group of the device */
		tun->group= (gid_t) arg;

1331
		tun_debug(KERN_INFO, tun, "group set to %d\n", tun->group);
1332 1333
		break;

1334 1335 1336
	case TUNSETLINK:
		/* Only allow setting the type when the interface is down */
		if (tun->dev->flags & IFF_UP) {
1337 1338
			tun_debug(KERN_INFO, tun,
				  "Linktype set failed because interface is up\n");
1339
			ret = -EBUSY;
1340 1341
		} else {
			tun->dev->type = (int) arg;
1342 1343
			tun_debug(KERN_INFO, tun, "linktype set to %d\n",
				  tun->dev->type);
1344
			ret = 0;
1345
		}
E
Eric W. Biederman 已提交
1346
		break;
1347

L
Linus Torvalds 已提交
1348 1349 1350 1351 1352
#ifdef TUN_DEBUG
	case TUNSETDEBUG:
		tun->debug = arg;
		break;
#endif
1353
	case TUNSETOFFLOAD:
1354
		ret = set_offload(tun, arg);
E
Eric W. Biederman 已提交
1355
		break;
1356

1357 1358
	case TUNSETTXFILTER:
		/* Can be set only for TAPs */
E
Eric W. Biederman 已提交
1359
		ret = -EINVAL;
1360
		if ((tun->flags & TUN_TYPE_MASK) != TUN_TAP_DEV)
E
Eric W. Biederman 已提交
1361
			break;
H
Harvey Harrison 已提交
1362
		ret = update_filter(&tun->txflt, (void __user *)arg);
E
Eric W. Biederman 已提交
1363
		break;
L
Linus Torvalds 已提交
1364 1365

	case SIOCGIFHWADDR:
1366
		/* Get hw address */
1367 1368
		memcpy(ifr.ifr_hwaddr.sa_data, tun->dev->dev_addr, ETH_ALEN);
		ifr.ifr_hwaddr.sa_family = tun->dev->type;
1369
		if (copy_to_user(argp, &ifr, ifreq_len))
E
Eric W. Biederman 已提交
1370 1371
			ret = -EFAULT;
		break;
L
Linus Torvalds 已提交
1372 1373

	case SIOCSIFHWADDR:
1374
		/* Set hw address */
1375 1376
		tun_debug(KERN_DEBUG, tun, "set hw address: %pM\n",
			  ifr.ifr_hwaddr.sa_data);
1377 1378

		ret = dev_set_mac_address(tun->dev, &ifr.ifr_hwaddr);
E
Eric W. Biederman 已提交
1379
		break;
1380 1381

	case TUNGETSNDBUF:
1382
		sndbuf = tun->socket.sk->sk_sndbuf;
1383 1384 1385 1386 1387 1388 1389 1390 1391 1392
		if (copy_to_user(argp, &sndbuf, sizeof(sndbuf)))
			ret = -EFAULT;
		break;

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

1393
		tun->socket.sk->sk_sndbuf = sndbuf;
1394 1395
		break;

1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414
	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
1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434
	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
1435
	default:
E
Eric W. Biederman 已提交
1436 1437
		ret = -EINVAL;
		break;
1438
	}
L
Linus Torvalds 已提交
1439

1440 1441 1442 1443
unlock:
	rtnl_unlock();
	if (tun)
		tun_put(tun);
E
Eric W. Biederman 已提交
1444
	return ret;
L
Linus Torvalds 已提交
1445 1446
}

1447 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
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
1482 1483
static int tun_chr_fasync(int fd, struct file *file, int on)
{
E
Eric W. Biederman 已提交
1484
	struct tun_struct *tun = tun_get(file);
L
Linus Torvalds 已提交
1485 1486 1487 1488 1489
	int ret;

	if (!tun)
		return -EBADFD;

1490
	tun_debug(KERN_INFO, tun, "tun_chr_fasync %d\n", on);
L
Linus Torvalds 已提交
1491 1492

	if ((ret = fasync_helper(fd, file, on, &tun->fasync)) < 0)
J
Jonathan Corbet 已提交
1493
		goto out;
1494

L
Linus Torvalds 已提交
1495
	if (on) {
1496
		ret = __f_setown(file, task_pid(current), PIDTYPE_PID, 0);
L
Linus Torvalds 已提交
1497
		if (ret)
J
Jonathan Corbet 已提交
1498
			goto out;
L
Linus Torvalds 已提交
1499
		tun->flags |= TUN_FASYNC;
1500
	} else
L
Linus Torvalds 已提交
1501
		tun->flags &= ~TUN_FASYNC;
J
Jonathan Corbet 已提交
1502 1503
	ret = 0;
out:
E
Eric W. Biederman 已提交
1504
	tun_put(tun);
J
Jonathan Corbet 已提交
1505
	return ret;
L
Linus Torvalds 已提交
1506 1507 1508 1509
}

static int tun_chr_open(struct inode *inode, struct file * file)
{
E
Eric W. Biederman 已提交
1510
	struct tun_file *tfile;
T
Thomas Gleixner 已提交
1511

1512
	DBG1(KERN_INFO, "tunX: tun_chr_open\n");
E
Eric W. Biederman 已提交
1513 1514 1515 1516

	tfile = kmalloc(sizeof(*tfile), GFP_KERNEL);
	if (!tfile)
		return -ENOMEM;
1517
	atomic_set(&tfile->count, 0);
E
Eric W. Biederman 已提交
1518
	tfile->tun = NULL;
E
Eric W. Biederman 已提交
1519
	tfile->net = get_net(current->nsproxy->net_ns);
E
Eric W. Biederman 已提交
1520
	file->private_data = tfile;
L
Linus Torvalds 已提交
1521 1522 1523 1524 1525
	return 0;
}

static int tun_chr_close(struct inode *inode, struct file *file)
{
E
Eric W. Biederman 已提交
1526
	struct tun_file *tfile = file->private_data;
E
Eric W. Biederman 已提交
1527
	struct tun_struct *tun;
L
Linus Torvalds 已提交
1528

E
Eric W. Biederman 已提交
1529
	tun = __tun_get(tfile);
E
Eric W. Biederman 已提交
1530
	if (tun) {
H
Herbert Xu 已提交
1531 1532
		struct net_device *dev = tun->dev;

1533
		tun_debug(KERN_INFO, tun, "tun_chr_close\n");
L
Linus Torvalds 已提交
1534

E
Eric W. Biederman 已提交
1535
		__tun_detach(tun);
L
Linus Torvalds 已提交
1536

D
Daniel Mack 已提交
1537
		/* If desirable, unregister the netdevice. */
H
Herbert Xu 已提交
1538 1539 1540 1541 1542 1543
		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
1544
	}
L
Linus Torvalds 已提交
1545

1546 1547
	tun = tfile->tun;
	if (tun)
1548
		sock_put(tun->socket.sk);
1549

E
Eric W. Biederman 已提交
1550
	put_net(tfile->net);
E
Eric W. Biederman 已提交
1551
	kfree(tfile);
L
Linus Torvalds 已提交
1552 1553 1554 1555

	return 0;
}

1556
static const struct file_operations tun_fops = {
1557
	.owner	= THIS_MODULE,
L
Linus Torvalds 已提交
1558
	.llseek = no_llseek,
1559 1560 1561 1562
	.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
1563
	.poll	= tun_chr_poll,
1564 1565 1566 1567
	.unlocked_ioctl	= tun_chr_ioctl,
#ifdef CONFIG_COMPAT
	.compat_ioctl = tun_chr_compat_ioctl,
#endif
L
Linus Torvalds 已提交
1568 1569
	.open	= tun_chr_open,
	.release = tun_chr_close,
1570
	.fasync = tun_chr_fasync
L
Linus Torvalds 已提交
1571 1572 1573 1574 1575
};

static struct miscdevice tun_miscdev = {
	.minor = TUN_MINOR,
	.name = "tun",
1576
	.nodename = "net/tun",
L
Linus Torvalds 已提交
1577 1578 1579 1580 1581 1582 1583 1584 1585
	.fops = &tun_fops,
};

/* ethtool interface */

static int tun_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
	cmd->supported		= 0;
	cmd->advertising	= 0;
1586
	ethtool_cmd_speed_set(cmd, SPEED_10);
L
Linus Torvalds 已提交
1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600
	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);

1601 1602
	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
L
Linus Torvalds 已提交
1603 1604 1605

	switch (tun->flags & TUN_TYPE_MASK) {
	case TUN_TUN_DEV:
1606
		strlcpy(info->bus_info, "tun", sizeof(info->bus_info));
L
Linus Torvalds 已提交
1607 1608
		break;
	case TUN_TAP_DEV:
1609
		strlcpy(info->bus_info, "tap", sizeof(info->bus_info));
L
Linus Torvalds 已提交
1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631
		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
}

1632
static const struct ethtool_ops tun_ethtool_ops = {
L
Linus Torvalds 已提交
1633 1634 1635 1636
	.get_settings	= tun_get_settings,
	.get_drvinfo	= tun_get_drvinfo,
	.get_msglevel	= tun_get_msglevel,
	.set_msglevel	= tun_set_msglevel,
1637
	.get_link	= ethtool_op_get_link,
L
Linus Torvalds 已提交
1638 1639
};

1640

L
Linus Torvalds 已提交
1641 1642 1643 1644
static int __init tun_init(void)
{
	int ret = 0;

1645 1646
	pr_info("%s, %s\n", DRV_DESCRIPTION, DRV_VERSION);
	pr_info("%s\n", DRV_COPYRIGHT);
L
Linus Torvalds 已提交
1647

1648
	ret = rtnl_link_register(&tun_link_ops);
1649
	if (ret) {
1650
		pr_err("Can't register link_ops\n");
1651
		goto err_linkops;
1652 1653
	}

L
Linus Torvalds 已提交
1654
	ret = misc_register(&tun_miscdev);
1655
	if (ret) {
1656
		pr_err("Can't register misc device %d\n", TUN_MINOR);
1657 1658
		goto err_misc;
	}
1659
	return  0;
1660
err_misc:
1661 1662
	rtnl_link_unregister(&tun_link_ops);
err_linkops:
L
Linus Torvalds 已提交
1663 1664 1665 1666 1667
	return ret;
}

static void tun_cleanup(void)
{
1668
	misc_deregister(&tun_miscdev);
1669
	rtnl_link_unregister(&tun_link_ops);
L
Linus Torvalds 已提交
1670 1671
}

1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688
/* 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
1689 1690 1691 1692 1693 1694
module_init(tun_init);
module_exit(tun_cleanup);
MODULE_DESCRIPTION(DRV_DESCRIPTION);
MODULE_AUTHOR(DRV_COPYRIGHT);
MODULE_LICENSE("GPL");
MODULE_ALIAS_MISCDEV(TUN_MINOR);
1695
MODULE_ALIAS("devname:net/tun");