tun.c 32.9 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 37 38 39 40 41 42 43 44 45 46
 *
 *  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.
 */

#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>
A
Arnd Bergmann 已提交
47
#include <linux/smp_lock.h>
L
Linus Torvalds 已提交
48 49 50 51 52 53 54 55 56 57 58 59 60 61
#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>
#include <linux/if.h>
#include <linux/if_arp.h>
#include <linux/if_ether.h>
#include <linux/if_tun.h>
#include <linux/crc32.h>
62
#include <linux/nsproxy.h>
63
#include <linux/virtio_net.h>
64
#include <net/net_namespace.h>
65
#include <net/netns/generic.h>
66
#include <net/rtnetlink.h>
67
#include <net/sock.h>
L
Linus Torvalds 已提交
68 69 70 71

#include <asm/system.h>
#include <asm/uaccess.h>

72 73 74
/* Uncomment to enable debugging */
/* #define TUN_DEBUG 1 */

L
Linus Torvalds 已提交
75 76
#ifdef TUN_DEBUG
static int debug;
77 78 79 80 81 82 83 84

#define DBG  if(tun->debug)printk
#define DBG1 if(debug==2)printk
#else
#define DBG( a... )
#define DBG1( a... )
#endif

85 86 87 88 89 90 91
#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
92
struct tun_file {
93
	atomic_t count;
E
Eric W. Biederman 已提交
94
	struct tun_struct *tun;
E
Eric W. Biederman 已提交
95
	struct net *net;
E
Eric W. Biederman 已提交
96 97
};

98 99
struct tun_sock;

100
struct tun_struct {
E
Eric W. Biederman 已提交
101
	struct tun_file		*tfile;
102
	unsigned int 		flags;
103 104 105 106 107 108
	uid_t			owner;
	gid_t			group;

	struct sk_buff_head	readq;

	struct net_device	*dev;
109
	struct fasync_struct	*fasync;
110

111
	struct tap_filter       txflt;
112 113
	struct sock		*sk;
	struct socket		socket;
114 115 116

#ifdef TUN_DEBUG
	int debug;
L
Linus Torvalds 已提交
117
#endif
118
};
L
Linus Torvalds 已提交
119

120 121 122 123 124 125 126 127 128 129
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);
}

130 131
static int tun_attach(struct tun_struct *tun, struct file *file)
{
E
Eric W. Biederman 已提交
132
	struct tun_file *tfile = file->private_data;
133
	const struct cred *cred = current_cred();
134
	int err;
135 136 137 138 139

	ASSERT_RTNL();

	/* Check permissions */
	if (((tun->owner != -1 && cred->euid != tun->owner) ||
140
	     (tun->group != -1 && !in_egroup_p(tun->group))) &&
141 142 143
		!capable(CAP_NET_ADMIN))
		return -EPERM;

144 145 146 147 148 149 150 151 152 153 154
	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
155 156
	tfile->tun = tun;
	tun->tfile = tfile;
157
	dev_hold(tun->dev);
158
	sock_hold(tun->sk);
159
	atomic_inc(&tfile->count);
160

161 162 163
out:
	netif_tx_unlock_bh(tun->dev);
	return err;
164 165
}

E
Eric W. Biederman 已提交
166 167 168
static void __tun_detach(struct tun_struct *tun)
{
	/* Detach from net device */
169
	netif_tx_lock_bh(tun->dev);
E
Eric W. Biederman 已提交
170
	tun->tfile = NULL;
171
	netif_tx_unlock_bh(tun->dev);
E
Eric W. Biederman 已提交
172 173 174

	/* Drop read queue */
	skb_queue_purge(&tun->readq);
175 176 177 178 179 180 181 182 183 184

	/* 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
185 186 187 188
}

static struct tun_struct *__tun_get(struct tun_file *tfile)
{
189 190 191 192 193 194
	struct tun_struct *tun = NULL;

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

	return tun;
E
Eric W. Biederman 已提交
195 196 197 198 199 200 201 202 203
}

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

static void tun_put(struct tun_struct *tun)
{
204 205 206 207
	struct tun_file *tfile = tun->tfile;

	if (atomic_dec_and_test(&tfile->count))
		tun_detach(tfile->tun);
E
Eric W. Biederman 已提交
208 209
}

210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259
/* TAP filterting */
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;

260 261
	/* Remaining multicast addresses are hashed,
	 * unicast will leave the filter disabled. */
262
	memset(filter->mask, 0, sizeof(filter->mask));
263 264 265 266 267
	for (; n < uf.count; n++) {
		if (!is_multicast_ether_addr(addr[n].u)) {
			err = 0; /* no filter */
			goto done;
		}
268
		addr_hash_set(filter->mask, addr[n].u);
269
	}
270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319

	/* 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
320 321
/* Network device part of the driver */

322
static const struct ethtool_ops tun_ethtool_ops;
L
Linus Torvalds 已提交
323

324 325 326 327 328 329 330 331 332
/* 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) {
333
		wake_up_all(&tun->socket.wait);
334 335 336 337 338
		if (atomic_dec_and_test(&tfile->count))
			__tun_detach(tun);
	}
}

339 340 341 342 343 344 345
static void tun_free_netdev(struct net_device *dev)
{
	struct tun_struct *tun = netdev_priv(dev);

	sock_put(tun->sk);
}

L
Linus Torvalds 已提交
346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367
/* 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 */
static int tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
{
	struct tun_struct *tun = netdev_priv(dev);

	DBG(KERN_INFO "%s: tun_net_xmit %d\n", tun->dev->name, skb->len);

	/* Drop packet if interface is not attached */
E
Eric W. Biederman 已提交
368
	if (!tun->tfile)
L
Linus Torvalds 已提交
369 370
		goto drop;

371 372 373 374 375 376
	/* 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;

L
Linus Torvalds 已提交
377 378 379 380 381 382 383 384
	if (skb_queue_len(&tun->readq) >= dev->tx_queue_len) {
		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. */
385
			dev->stats.tx_fifo_errors++;
L
Linus Torvalds 已提交
386 387 388 389 390 391 392
		} else {
			/* Single queue mode.
			 * Driver handles dropping of all packets itself. */
			goto drop;
		}
	}

393
	/* Enqueue packet */
L
Linus Torvalds 已提交
394 395 396 397 398 399
	skb_queue_tail(&tun->readq, skb);
	dev->trans_start = jiffies;

	/* Notify and wake up reader process */
	if (tun->flags & TUN_FASYNC)
		kill_fasync(&tun->fasync, SIGIO, POLL_IN);
400
	wake_up_interruptible(&tun->socket.wait);
L
Linus Torvalds 已提交
401 402 403
	return 0;

drop:
404
	dev->stats.tx_dropped++;
L
Linus Torvalds 已提交
405 406 407 408
	kfree_skb(skb);
	return 0;
}

409
static void tun_net_mclist(struct net_device *dev)
L
Linus Torvalds 已提交
410
{
411 412 413 414 415 416
	/*
	 * 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.
	 */
	return;
L
Linus Torvalds 已提交
417 418
}

E
Ed Swierk 已提交
419 420 421 422 423 424 425 426 427 428 429 430
#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;
}

S
Stephen Hemminger 已提交
431
static const struct net_device_ops tun_netdev_ops = {
432
	.ndo_uninit		= tun_net_uninit,
S
Stephen Hemminger 已提交
433 434
	.ndo_open		= tun_net_open,
	.ndo_stop		= tun_net_close,
435
	.ndo_start_xmit		= tun_net_xmit,
S
Stephen Hemminger 已提交
436 437 438 439
	.ndo_change_mtu		= tun_net_change_mtu,
};

static const struct net_device_ops tap_netdev_ops = {
440
	.ndo_uninit		= tun_net_uninit,
S
Stephen Hemminger 已提交
441 442
	.ndo_open		= tun_net_open,
	.ndo_stop		= tun_net_close,
443
	.ndo_start_xmit		= tun_net_xmit,
S
Stephen Hemminger 已提交
444 445 446 447 448 449
	.ndo_change_mtu		= tun_net_change_mtu,
	.ndo_set_multicast_list	= tun_net_mclist,
	.ndo_set_mac_address	= eth_mac_addr,
	.ndo_validate_addr	= eth_validate_addr,
};

L
Linus Torvalds 已提交
450 451 452 453
/* Initialize net device. */
static void tun_net_init(struct net_device *dev)
{
	struct tun_struct *tun = netdev_priv(dev);
454

L
Linus Torvalds 已提交
455 456
	switch (tun->flags & TUN_TYPE_MASK) {
	case TUN_TUN_DEV:
S
Stephen Hemminger 已提交
457 458
		dev->netdev_ops = &tun_netdev_ops;

L
Linus Torvalds 已提交
459 460 461 462 463 464
		/* Point-to-Point TUN Device */
		dev->hard_header_len = 0;
		dev->addr_len = 0;
		dev->mtu = 1500;

		/* Zero header length */
465
		dev->type = ARPHRD_NONE;
L
Linus Torvalds 已提交
466 467 468 469 470
		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
471
		dev->netdev_ops = &tap_netdev_ops;
L
Linus Torvalds 已提交
472 473
		/* Ethernet TAP Device */
		ether_setup(dev);
474

475
		random_ether_addr(dev->dev_addr);
476

L
Linus Torvalds 已提交
477 478 479 480 481 482 483 484 485
		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)
486
{
487 488
	struct tun_file *tfile = file->private_data;
	struct tun_struct *tun = __tun_get(tfile);
489 490
	struct sock *sk = tun->sk;
	unsigned int mask = 0;
L
Linus Torvalds 已提交
491 492

	if (!tun)
493
		return POLLERR;
L
Linus Torvalds 已提交
494 495 496

	DBG(KERN_INFO "%s: tun_chr_poll\n", tun->dev->name);

497
	poll_wait(file, &tun->socket.wait, wait);
498

499
	if (!skb_queue_empty(&tun->readq))
L
Linus Torvalds 已提交
500 501
		mask |= POLLIN | POLLRDNORM;

502 503 504 505 506
	if (sock_writeable(sk) ||
	    (!test_and_set_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags) &&
	     sock_writeable(sk)))
		mask |= POLLOUT | POLLWRNORM;

507 508 509
	if (tun->dev->reg_state != NETREG_REGISTERED)
		mask = POLLERR;

E
Eric W. Biederman 已提交
510
	tun_put(tun);
L
Linus Torvalds 已提交
511 512 513
	return mask;
}

514 515
/* 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). */
516 517 518
static inline struct sk_buff *tun_alloc_skb(struct tun_struct *tun,
					    size_t prepad, size_t len,
					    size_t linear, int noblock)
519
{
520
	struct sock *sk = tun->sk;
521
	struct sk_buff *skb;
522
	int err;
523 524

	/* Under a page?  Don't bother with paged skb. */
H
Herbert Xu 已提交
525
	if (prepad + len < PAGE_SIZE || !linear)
526
		linear = len;
527

528 529
	skb = sock_alloc_send_pskb(sk, prepad + linear, len - linear, noblock,
				   &err);
530
	if (!skb)
531
		return ERR_PTR(err);
532 533 534

	skb_reserve(skb, prepad);
	skb_put(skb, linear);
535 536
	skb->data_len = len - linear;
	skb->len += len - linear;
537 538 539 540

	return skb;
}

L
Linus Torvalds 已提交
541
/* Get packet from user space buffer */
542
static __inline__ ssize_t tun_get_user(struct tun_struct *tun,
543
				       const struct iovec *iv, size_t count,
544
				       int noblock)
L
Linus Torvalds 已提交
545
{
546
	struct tun_pi pi = { 0, cpu_to_be16(ETH_P_IP) };
L
Linus Torvalds 已提交
547 548
	struct sk_buff *skb;
	size_t len = count, align = 0;
549
	struct virtio_net_hdr gso = { 0 };
550
	int offset = 0;
L
Linus Torvalds 已提交
551 552 553 554 555

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

556
		if (memcpy_fromiovecend((void *)&pi, iv, 0, sizeof(pi)))
L
Linus Torvalds 已提交
557
			return -EFAULT;
558
		offset += sizeof(pi);
L
Linus Torvalds 已提交
559 560
	}

561 562 563 564
	if (tun->flags & TUN_VNET_HDR) {
		if ((len -= sizeof(gso)) > count)
			return -EINVAL;

565
		if (memcpy_fromiovecend((void *)&gso, iv, offset, sizeof(gso)))
566 567
			return -EFAULT;

568 569 570 571
		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;

572 573
		if (gso.hdr_len > len)
			return -EINVAL;
574
		offset += sizeof(gso);
575 576
	}

577
	if ((tun->flags & TUN_TYPE_MASK) == TUN_TAP_DEV) {
L
Linus Torvalds 已提交
578
		align = NET_IP_ALIGN;
H
Herbert Xu 已提交
579 580
		if (unlikely(len < ETH_HLEN ||
			     (gso.hdr_len && gso.hdr_len < ETH_HLEN)))
581 582
			return -EINVAL;
	}
583

584 585 586 587 588
	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
589 590
	}

591
	if (skb_copy_datagram_from_iovec(skb, 0, iv, offset, len)) {
592
		tun->dev->stats.rx_dropped++;
D
Dave Jones 已提交
593
		kfree_skb(skb);
L
Linus Torvalds 已提交
594
		return -EFAULT;
D
Dave Jones 已提交
595
	}
L
Linus Torvalds 已提交
596

597 598 599 600 601 602 603 604 605 606
	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;
		}
	} else if (tun->flags & TUN_NOCHECKSUM)
		skb->ip_summed = CHECKSUM_UNNECESSARY;

L
Linus Torvalds 已提交
607 608
	switch (tun->flags & TUN_TYPE_MASK) {
	case TUN_TUN_DEV:
609 610 611 612 613 614 615 616 617 618 619 620 621 622 623
		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;
			}
		}

624
		skb_reset_mac_header(skb);
L
Linus Torvalds 已提交
625
		skb->protocol = pi.proto;
626
		skb->dev = tun->dev;
L
Linus Torvalds 已提交
627 628 629 630 631 632
		break;
	case TUN_TAP_DEV:
		skb->protocol = eth_type_trans(skb, tun->dev);
		break;
	};

633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661
	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;
		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;
	}
662

L
Linus Torvalds 已提交
663
	netif_rx_ni(skb);
664

665 666
	tun->dev->stats.rx_packets++;
	tun->dev->stats.rx_bytes += len;
L
Linus Torvalds 已提交
667 668

	return count;
669
}
L
Linus Torvalds 已提交
670

671 672
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
673
{
674
	struct file *file = iocb->ki_filp;
H
Herbert Xu 已提交
675
	struct tun_struct *tun = tun_get(file);
E
Eric W. Biederman 已提交
676
	ssize_t result;
L
Linus Torvalds 已提交
677 678 679 680 681 682

	if (!tun)
		return -EBADFD;

	DBG(KERN_INFO "%s: tun_chr_write %ld\n", tun->dev->name, count);

683
	result = tun_get_user(tun, iv, iov_length(iv, count),
684
			      file->f_flags & O_NONBLOCK);
E
Eric W. Biederman 已提交
685 686 687

	tun_put(tun);
	return result;
L
Linus Torvalds 已提交
688 689 690 691 692
}

/* Put packet to the user space buffer */
static __inline__ ssize_t tun_put_user(struct tun_struct *tun,
				       struct sk_buff *skb,
693
				       const struct iovec *iv, int len)
L
Linus Torvalds 已提交
694 695 696 697 698 699 700 701 702 703 704 705
{
	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;
		}
706

707
		if (memcpy_toiovecend(iv, (void *) &pi, 0, sizeof(pi)))
L
Linus Torvalds 已提交
708 709
			return -EFAULT;
		total += sizeof(pi);
710
	}
L
Linus Torvalds 已提交
711

712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739
	if (tun->flags & TUN_VNET_HDR) {
		struct virtio_net_hdr gso = { 0 }; /* no info leak */
		if ((len -= sizeof(gso)) < 0)
			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;
			else
				BUG();
			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;
			gso.csum_start = skb->csum_start - skb_headroom(skb);
			gso.csum_offset = skb->csum_offset;
		} /* else everything is zero */

740 741
		if (unlikely(memcpy_toiovecend(iv, (void *)&gso, total,
					       sizeof(gso))))
742 743 744 745
			return -EFAULT;
		total += sizeof(gso);
	}

L
Linus Torvalds 已提交
746 747
	len = min_t(int, skb->len, len);

748
	skb_copy_datagram_const_iovec(skb, 0, iv, total, len);
L
Linus Torvalds 已提交
749 750
	total += len;

751 752
	tun->dev->stats.tx_packets++;
	tun->dev->stats.tx_bytes += len;
L
Linus Torvalds 已提交
753 754 755 756

	return total;
}

757 758
static ssize_t tun_chr_aio_read(struct kiocb *iocb, const struct iovec *iv,
			    unsigned long count, loff_t pos)
L
Linus Torvalds 已提交
759
{
760
	struct file *file = iocb->ki_filp;
761 762
	struct tun_file *tfile = file->private_data;
	struct tun_struct *tun = __tun_get(tfile);
L
Linus Torvalds 已提交
763 764 765 766 767 768 769 770 771
	DECLARE_WAITQUEUE(wait, current);
	struct sk_buff *skb;
	ssize_t len, ret = 0;

	if (!tun)
		return -EBADFD;

	DBG(KERN_INFO "%s: tun_chr_read\n", tun->dev->name);

A
Akinobu Mita 已提交
772
	len = iov_length(iv, count);
E
Eric W. Biederman 已提交
773 774 775 776
	if (len < 0) {
		ret = -EINVAL;
		goto out;
	}
L
Linus Torvalds 已提交
777

778
	add_wait_queue(&tun->socket.wait, &wait);
L
Linus Torvalds 已提交
779 780 781 782 783 784 785 786 787 788 789 790 791
	while (len) {
		current->state = TASK_INTERRUPTIBLE;

		/* Read frames from the queue */
		if (!(skb=skb_dequeue(&tun->readq))) {
			if (file->f_flags & O_NONBLOCK) {
				ret = -EAGAIN;
				break;
			}
			if (signal_pending(current)) {
				ret = -ERESTARTSYS;
				break;
			}
792 793 794 795
			if (tun->dev->reg_state != NETREG_REGISTERED) {
				ret = -EIO;
				break;
			}
L
Linus Torvalds 已提交
796 797 798 799 800 801 802

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

803
		ret = tun_put_user(tun, skb, iv, len);
804 805
		kfree_skb(skb);
		break;
L
Linus Torvalds 已提交
806 807 808
	}

	current->state = TASK_RUNNING;
809
	remove_wait_queue(&tun->socket.wait, &wait);
L
Linus Torvalds 已提交
810

E
Eric W. Biederman 已提交
811 812
out:
	tun_put(tun);
L
Linus Torvalds 已提交
813 814 815 816 817 818 819 820 821 822
	return ret;
}

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

	skb_queue_head_init(&tun->readq);

	tun->owner = -1;
823
	tun->group = -1;
L
Linus Torvalds 已提交
824 825

	dev->ethtool_ops = &tun_ethtool_ops;
826
	dev->destructor = tun_free_netdev;
L
Linus Torvalds 已提交
827 828
}

829 830 831 832 833 834 835 836 837 838 839 840 841 842 843
/* 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,
};

844 845 846 847 848 849 850 851 852 853
static void tun_sock_write_space(struct sock *sk)
{
	struct tun_struct *tun;

	if (!sock_writeable(sk))
		return;

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

H
Herbert Xu 已提交
854 855 856
	if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
		wake_up_interruptible_sync(sk->sk_sleep);

857 858 859 860 861 862
	tun = container_of(sk, struct tun_sock, sk)->tun;
	kill_fasync(&tun->fasync, SIGIO, POLL_OUT);
}

static void tun_sock_destruct(struct sock *sk)
{
863
	free_netdev(container_of(sk, struct tun_sock, sk)->tun->dev);
864 865 866 867 868 869 870
}

static struct proto tun_proto = {
	.name		= "tun",
	.owner		= THIS_MODULE,
	.obj_size	= sizeof(struct tun_sock),
};
871

872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917
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);

918
static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
L
Linus Torvalds 已提交
919
{
920
	struct sock *sk;
L
Linus Torvalds 已提交
921 922 923 924
	struct tun_struct *tun;
	struct net_device *dev;
	int err;

925 926
	dev = __dev_get_by_name(net, ifr->ifr_name);
	if (dev) {
927 928
		if (ifr->ifr_flags & IFF_TUN_EXCL)
			return -EBUSY;
929 930 931 932 933 934 935
		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;

936 937 938
		err = tun_attach(tun, file);
		if (err < 0)
			return err;
939
	}
L
Linus Torvalds 已提交
940 941 942 943 944 945
	else {
		char *name;
		unsigned long flags = 0;

		err = -EINVAL;

946 947 948
		if (!capable(CAP_NET_ADMIN))
			return -EPERM;

L
Linus Torvalds 已提交
949 950 951 952 953 954 955 956 957
		/* 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";
958
		} else
L
Linus Torvalds 已提交
959
			goto failed;
960

L
Linus Torvalds 已提交
961 962 963 964 965 966 967 968
		if (*ifr->ifr_name)
			name = ifr->ifr_name;

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

969
		dev_net_set(dev, net);
970
		dev->rtnl_link_ops = &tun_link_ops;
S
Stephen Hemminger 已提交
971

L
Linus Torvalds 已提交
972 973 974
		tun = netdev_priv(dev);
		tun->dev = dev;
		tun->flags = flags;
975
		tun->txflt.count = 0;
L
Linus Torvalds 已提交
976

977 978 979 980 981
		err = -ENOMEM;
		sk = sk_alloc(net, AF_UNSPEC, GFP_KERNEL, &tun_proto);
		if (!sk)
			goto err_free_dev;

982
		init_waitqueue_head(&tun->socket.wait);
983 984 985 986 987 988 989
		sock_init_data(&tun->socket, sk);
		sk->sk_write_space = tun_sock_write_space;
		sk->sk_sndbuf = INT_MAX;

		tun->sk = sk;
		container_of(sk, struct tun_sock, sk)->tun = tun;

L
Linus Torvalds 已提交
990 991 992 993 994
		tun_net_init(dev);

		if (strchr(dev->name, '%')) {
			err = dev_alloc_name(dev, dev->name);
			if (err < 0)
995
				goto err_free_sk;
L
Linus Torvalds 已提交
996 997
		}

998
		err = -EINVAL;
L
Linus Torvalds 已提交
999 1000
		err = register_netdevice(tun->dev);
		if (err < 0)
1001 1002
			goto err_free_sk;

1003 1004 1005 1006 1007
		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))
			printk(KERN_ERR "Failed to create tun sysfs files\n");

1008
		sk->sk_destruct = tun_sock_destruct;
1009 1010 1011

		err = tun_attach(tun, file);
		if (err < 0)
1012
			goto failed;
L
Linus Torvalds 已提交
1013 1014 1015 1016 1017 1018
	}

	DBG(KERN_INFO "%s: tun_set_iff\n", tun->dev->name);

	if (ifr->ifr_flags & IFF_NO_PI)
		tun->flags |= TUN_NO_PI;
1019 1020
	else
		tun->flags &= ~TUN_NO_PI;
L
Linus Torvalds 已提交
1021 1022 1023

	if (ifr->ifr_flags & IFF_ONE_QUEUE)
		tun->flags |= TUN_ONE_QUEUE;
1024 1025
	else
		tun->flags &= ~TUN_ONE_QUEUE;
L
Linus Torvalds 已提交
1026

1027 1028 1029 1030 1031
	if (ifr->ifr_flags & IFF_VNET_HDR)
		tun->flags |= TUN_VNET_HDR;
	else
		tun->flags &= ~TUN_VNET_HDR;

1032 1033 1034 1035 1036 1037
	/* 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
1038 1039 1040
	strcpy(ifr->ifr_name, tun->dev->name);
	return 0;

1041 1042
 err_free_sk:
	sock_put(sk);
L
Linus Torvalds 已提交
1043 1044 1045 1046 1047 1048
 err_free_dev:
	free_netdev(dev);
 failed:
	return err;
}

1049 1050
static int tun_get_iff(struct net *net, struct file *file, struct ifreq *ifr)
{
E
Eric W. Biederman 已提交
1051
	struct tun_struct *tun = tun_get(file);
1052 1053 1054 1055 1056 1057 1058 1059

	if (!tun)
		return -EBADFD;

	DBG(KERN_INFO "%s: tun_get_iff\n", tun->dev->name);

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

1060
	ifr->ifr_flags = tun_flags(tun);
1061

E
Eric W. Biederman 已提交
1062
	tun_put(tun);
1063 1064 1065
	return 0;
}

1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105
/* This is like a cut-down ethtool ops, except done via tun fd so no
 * privs required. */
static int set_offload(struct net_device *dev, unsigned long arg)
{
	unsigned int old_features, features;

	old_features = dev->features;
	/* Unset features, set them as we chew on the arg. */
	features = (old_features & ~(NETIF_F_HW_CSUM|NETIF_F_SG|NETIF_F_FRAGLIST
				    |NETIF_F_TSO_ECN|NETIF_F_TSO|NETIF_F_TSO6));

	if (arg & TUN_F_CSUM) {
		features |= NETIF_F_HW_CSUM|NETIF_F_SG|NETIF_F_FRAGLIST;
		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);
		}
	}

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

	dev->features = features;
	if (old_features != dev->features)
		netdev_features_change(dev);

	return 0;
}

1106
static int tun_chr_ioctl(struct inode *inode, struct file *file,
L
Linus Torvalds 已提交
1107 1108
			 unsigned int cmd, unsigned long arg)
{
E
Eric W. Biederman 已提交
1109
	struct tun_file *tfile = file->private_data;
E
Eric W. Biederman 已提交
1110
	struct tun_struct *tun;
L
Linus Torvalds 已提交
1111 1112
	void __user* argp = (void __user*)arg;
	struct ifreq ifr;
1113
	int sndbuf;
1114
	int ret;
L
Linus Torvalds 已提交
1115 1116 1117 1118 1119

	if (cmd == TUNSETIFF || _IOC_TYPE(cmd) == 0x89)
		if (copy_from_user(&ifr, argp, sizeof ifr))
			return -EFAULT;

E
Eric W. Biederman 已提交
1120 1121 1122 1123 1124 1125 1126 1127 1128
	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);
	}

E
Eric W. Biederman 已提交
1129
	tun = __tun_get(tfile);
L
Linus Torvalds 已提交
1130 1131 1132 1133 1134 1135
	if (cmd == TUNSETIFF && !tun) {
		int err;

		ifr.ifr_name[IFNAMSIZ-1] = '\0';

		rtnl_lock();
E
Eric W. Biederman 已提交
1136
		err = tun_set_iff(tfile->net, file, &ifr);
L
Linus Torvalds 已提交
1137 1138 1139 1140 1141 1142 1143 1144 1145 1146
		rtnl_unlock();

		if (err)
			return err;

		if (copy_to_user(argp, &ifr, sizeof(ifr)))
			return -EFAULT;
		return 0;
	}

1147

L
Linus Torvalds 已提交
1148 1149 1150 1151 1152
	if (!tun)
		return -EBADFD;

	DBG(KERN_INFO "%s: tun_chr_ioctl cmd %d\n", tun->dev->name, cmd);

E
Eric W. Biederman 已提交
1153
	ret = 0;
L
Linus Torvalds 已提交
1154
	switch (cmd) {
1155 1156 1157
	case TUNGETIFF:
		ret = tun_get_iff(current->nsproxy->net_ns, file, &ifr);
		if (ret)
E
Eric W. Biederman 已提交
1158
			break;
1159 1160

		if (copy_to_user(argp, &ifr, sizeof(ifr)))
E
Eric W. Biederman 已提交
1161
			ret = -EFAULT;
1162 1163
		break;

L
Linus Torvalds 已提交
1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182
	case TUNSETNOCSUM:
		/* Disable/Enable checksum */
		if (arg)
			tun->flags |= TUN_NOCHECKSUM;
		else
			tun->flags &= ~TUN_NOCHECKSUM;

		DBG(KERN_INFO "%s: checksum %s\n",
		    tun->dev->name, arg ? "disabled" : "enabled");
		break;

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

		DBG(KERN_INFO "%s: persist %s\n",
T
Toyo Abe 已提交
1183
		    tun->dev->name, arg ? "enabled" : "disabled");
L
Linus Torvalds 已提交
1184 1185 1186 1187 1188 1189 1190 1191 1192
		break;

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

		DBG(KERN_INFO "%s: owner set to %d\n", tun->dev->name, tun->owner);
		break;

1193 1194 1195 1196 1197 1198 1199
	case TUNSETGROUP:
		/* Set group of the device */
		tun->group= (gid_t) arg;

		DBG(KERN_INFO "%s: group set to %d\n", tun->dev->name, tun->group);
		break;

1200 1201
	case TUNSETLINK:
		/* Only allow setting the type when the interface is down */
1202
		rtnl_lock();
1203 1204 1205
		if (tun->dev->flags & IFF_UP) {
			DBG(KERN_INFO "%s: Linktype set failed because interface is up\n",
				tun->dev->name);
1206
			ret = -EBUSY;
1207 1208 1209
		} else {
			tun->dev->type = (int) arg;
			DBG(KERN_INFO "%s: linktype set to %d\n", tun->dev->name, tun->dev->type);
1210
			ret = 0;
1211
		}
1212
		rtnl_unlock();
E
Eric W. Biederman 已提交
1213
		break;
1214

L
Linus Torvalds 已提交
1215 1216 1217 1218 1219
#ifdef TUN_DEBUG
	case TUNSETDEBUG:
		tun->debug = arg;
		break;
#endif
1220 1221 1222 1223
	case TUNSETOFFLOAD:
		rtnl_lock();
		ret = set_offload(tun->dev, arg);
		rtnl_unlock();
E
Eric W. Biederman 已提交
1224
		break;
1225

1226 1227
	case TUNSETTXFILTER:
		/* Can be set only for TAPs */
E
Eric W. Biederman 已提交
1228
		ret = -EINVAL;
1229
		if ((tun->flags & TUN_TYPE_MASK) != TUN_TAP_DEV)
E
Eric W. Biederman 已提交
1230
			break;
1231
		rtnl_lock();
H
Harvey Harrison 已提交
1232
		ret = update_filter(&tun->txflt, (void __user *)arg);
1233
		rtnl_unlock();
E
Eric W. Biederman 已提交
1234
		break;
L
Linus Torvalds 已提交
1235 1236

	case SIOCGIFHWADDR:
1237 1238 1239 1240
		/* Get hw addres */
		memcpy(ifr.ifr_hwaddr.sa_data, tun->dev->dev_addr, ETH_ALEN);
		ifr.ifr_hwaddr.sa_family = tun->dev->type;
		if (copy_to_user(argp, &ifr, sizeof ifr))
E
Eric W. Biederman 已提交
1241 1242
			ret = -EFAULT;
		break;
L
Linus Torvalds 已提交
1243 1244

	case SIOCSIFHWADDR:
1245
		/* Set hw address */
J
Johannes Berg 已提交
1246 1247
		DBG(KERN_DEBUG "%s: set hw address: %pM\n",
			tun->dev->name, ifr.ifr_hwaddr.sa_data);
1248 1249 1250 1251

		rtnl_lock();
		ret = dev_set_mac_address(tun->dev, &ifr.ifr_hwaddr);
		rtnl_unlock();
E
Eric W. Biederman 已提交
1252
		break;
1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268

	case TUNGETSNDBUF:
		sndbuf = tun->sk->sk_sndbuf;
		if (copy_to_user(argp, &sndbuf, sizeof(sndbuf)))
			ret = -EFAULT;
		break;

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

		tun->sk->sk_sndbuf = sndbuf;
		break;

L
Linus Torvalds 已提交
1269
	default:
E
Eric W. Biederman 已提交
1270 1271
		ret = -EINVAL;
		break;
L
Linus Torvalds 已提交
1272 1273
	};

E
Eric W. Biederman 已提交
1274 1275
	tun_put(tun);
	return ret;
L
Linus Torvalds 已提交
1276 1277 1278 1279
}

static int tun_chr_fasync(int fd, struct file *file, int on)
{
E
Eric W. Biederman 已提交
1280
	struct tun_struct *tun = tun_get(file);
L
Linus Torvalds 已提交
1281 1282 1283 1284 1285 1286 1287
	int ret;

	if (!tun)
		return -EBADFD;

	DBG(KERN_INFO "%s: tun_chr_fasync %d\n", tun->dev->name, on);

J
Jonathan Corbet 已提交
1288
	lock_kernel();
L
Linus Torvalds 已提交
1289
	if ((ret = fasync_helper(fd, file, on, &tun->fasync)) < 0)
J
Jonathan Corbet 已提交
1290
		goto out;
1291

L
Linus Torvalds 已提交
1292
	if (on) {
1293
		ret = __f_setown(file, task_pid(current), PIDTYPE_PID, 0);
L
Linus Torvalds 已提交
1294
		if (ret)
J
Jonathan Corbet 已提交
1295
			goto out;
L
Linus Torvalds 已提交
1296
		tun->flags |= TUN_FASYNC;
1297
	} else
L
Linus Torvalds 已提交
1298
		tun->flags &= ~TUN_FASYNC;
J
Jonathan Corbet 已提交
1299 1300 1301
	ret = 0;
out:
	unlock_kernel();
E
Eric W. Biederman 已提交
1302
	tun_put(tun);
J
Jonathan Corbet 已提交
1303
	return ret;
L
Linus Torvalds 已提交
1304 1305 1306 1307
}

static int tun_chr_open(struct inode *inode, struct file * file)
{
E
Eric W. Biederman 已提交
1308
	struct tun_file *tfile;
A
Arnd Bergmann 已提交
1309
	cycle_kernel_lock();
L
Linus Torvalds 已提交
1310
	DBG1(KERN_INFO "tunX: tun_chr_open\n");
E
Eric W. Biederman 已提交
1311 1312 1313 1314

	tfile = kmalloc(sizeof(*tfile), GFP_KERNEL);
	if (!tfile)
		return -ENOMEM;
1315
	atomic_set(&tfile->count, 0);
E
Eric W. Biederman 已提交
1316
	tfile->tun = NULL;
E
Eric W. Biederman 已提交
1317
	tfile->net = get_net(current->nsproxy->net_ns);
E
Eric W. Biederman 已提交
1318
	file->private_data = tfile;
L
Linus Torvalds 已提交
1319 1320 1321 1322 1323
	return 0;
}

static int tun_chr_close(struct inode *inode, struct file *file)
{
E
Eric W. Biederman 已提交
1324
	struct tun_file *tfile = file->private_data;
E
Eric W. Biederman 已提交
1325
	struct tun_struct *tun;
L
Linus Torvalds 已提交
1326 1327


E
Eric W. Biederman 已提交
1328 1329
	rtnl_lock();
	tun = __tun_get(tfile);
E
Eric W. Biederman 已提交
1330 1331
	if (tun) {
		DBG(KERN_INFO "%s: tun_chr_close\n", tun->dev->name);
L
Linus Torvalds 已提交
1332

E
Eric W. Biederman 已提交
1333
		__tun_detach(tun);
L
Linus Torvalds 已提交
1334

E
Eric W. Biederman 已提交
1335
		/* If desireable, unregister the netdevice. */
1336
		if (!(tun->flags & TUN_PERSIST))
E
Eric W. Biederman 已提交
1337
			unregister_netdevice(tun->dev);
L
Linus Torvalds 已提交
1338

E
Eric W. Biederman 已提交
1339
	}
E
Eric W. Biederman 已提交
1340
	rtnl_unlock();
L
Linus Torvalds 已提交
1341

1342 1343 1344 1345
	tun = tfile->tun;
	if (tun)
		sock_put(tun->sk);

E
Eric W. Biederman 已提交
1346
	put_net(tfile->net);
E
Eric W. Biederman 已提交
1347
	kfree(tfile);
L
Linus Torvalds 已提交
1348 1349 1350 1351

	return 0;
}

1352
static const struct file_operations tun_fops = {
1353
	.owner	= THIS_MODULE,
L
Linus Torvalds 已提交
1354
	.llseek = no_llseek,
1355 1356 1357 1358
	.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
1359 1360 1361 1362
	.poll	= tun_chr_poll,
	.ioctl	= tun_chr_ioctl,
	.open	= tun_chr_open,
	.release = tun_chr_close,
1363
	.fasync = tun_chr_fasync
L
Linus Torvalds 已提交
1364 1365 1366 1367 1368
};

static struct miscdevice tun_miscdev = {
	.minor = TUN_MINOR,
	.name = "tun",
1369
	.devnode = "net/tun",
L
Linus Torvalds 已提交
1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428
	.fops = &tun_fops,
};

/* ethtool interface */

static int tun_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
	cmd->supported		= 0;
	cmd->advertising	= 0;
	cmd->speed		= SPEED_10;
	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);

	strcpy(info->driver, DRV_NAME);
	strcpy(info->version, DRV_VERSION);
	strcpy(info->fw_version, "N/A");

	switch (tun->flags & TUN_TYPE_MASK) {
	case TUN_TUN_DEV:
		strcpy(info->bus_info, "tun");
		break;
	case TUN_TAP_DEV:
		strcpy(info->bus_info, "tap");
		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
}

static u32 tun_get_link(struct net_device *dev)
{
	struct tun_struct *tun = netdev_priv(dev);
E
Eric W. Biederman 已提交
1429
	return !!tun->tfile;
L
Linus Torvalds 已提交
1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447
}

static u32 tun_get_rx_csum(struct net_device *dev)
{
	struct tun_struct *tun = netdev_priv(dev);
	return (tun->flags & TUN_NOCHECKSUM) == 0;
}

static int tun_set_rx_csum(struct net_device *dev, u32 data)
{
	struct tun_struct *tun = netdev_priv(dev);
	if (data)
		tun->flags &= ~TUN_NOCHECKSUM;
	else
		tun->flags |= TUN_NOCHECKSUM;
	return 0;
}

1448
static const struct ethtool_ops tun_ethtool_ops = {
L
Linus Torvalds 已提交
1449 1450 1451 1452 1453 1454 1455 1456 1457
	.get_settings	= tun_get_settings,
	.get_drvinfo	= tun_get_drvinfo,
	.get_msglevel	= tun_get_msglevel,
	.set_msglevel	= tun_set_msglevel,
	.get_link	= tun_get_link,
	.get_rx_csum	= tun_get_rx_csum,
	.set_rx_csum	= tun_set_rx_csum
};

1458

L
Linus Torvalds 已提交
1459 1460 1461 1462 1463 1464 1465
static int __init tun_init(void)
{
	int ret = 0;

	printk(KERN_INFO "tun: %s, %s\n", DRV_DESCRIPTION, DRV_VERSION);
	printk(KERN_INFO "tun: %s\n", DRV_COPYRIGHT);

1466
	ret = rtnl_link_register(&tun_link_ops);
1467
	if (ret) {
1468 1469
		printk(KERN_ERR "tun: Can't register link_ops\n");
		goto err_linkops;
1470 1471
	}

L
Linus Torvalds 已提交
1472
	ret = misc_register(&tun_miscdev);
1473
	if (ret) {
L
Linus Torvalds 已提交
1474
		printk(KERN_ERR "tun: Can't register misc device %d\n", TUN_MINOR);
1475 1476
		goto err_misc;
	}
1477
	return  0;
1478
err_misc:
1479 1480
	rtnl_link_unregister(&tun_link_ops);
err_linkops:
L
Linus Torvalds 已提交
1481 1482 1483 1484 1485
	return ret;
}

static void tun_cleanup(void)
{
1486
	misc_deregister(&tun_miscdev);
1487
	rtnl_link_unregister(&tun_link_ops);
L
Linus Torvalds 已提交
1488 1489 1490 1491 1492 1493 1494 1495
}

module_init(tun_init);
module_exit(tun_cleanup);
MODULE_DESCRIPTION(DRV_DESCRIPTION);
MODULE_AUTHOR(DRV_COPYRIGHT);
MODULE_LICENSE("GPL");
MODULE_ALIAS_MISCDEV(TUN_MINOR);