tun.c 33.2 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);
401
	return NETDEV_TX_OK;
L
Linus Torvalds 已提交
402 403

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

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
	struct sock *sk;
490
	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
	sk = tun->sk;

L
Linus Torvalds 已提交
497 498
	DBG(KERN_INFO "%s: tun_chr_poll\n", tun->dev->name);

499
	poll_wait(file, &tun->socket.wait, wait);
500

501
	if (!skb_queue_empty(&tun->readq))
L
Linus Torvalds 已提交
502 503
		mask |= POLLIN | POLLRDNORM;

504 505 506 507 508
	if (sock_writeable(sk) ||
	    (!test_and_set_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags) &&
	     sock_writeable(sk)))
		mask |= POLLOUT | POLLWRNORM;

509 510 511
	if (tun->dev->reg_state != NETREG_REGISTERED)
		mask = POLLERR;

E
Eric W. Biederman 已提交
512
	tun_put(tun);
L
Linus Torvalds 已提交
513 514 515
	return mask;
}

516 517
/* 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). */
518 519 520
static inline struct sk_buff *tun_alloc_skb(struct tun_struct *tun,
					    size_t prepad, size_t len,
					    size_t linear, int noblock)
521
{
522
	struct sock *sk = tun->sk;
523
	struct sk_buff *skb;
524
	int err;
525 526

	/* Under a page?  Don't bother with paged skb. */
H
Herbert Xu 已提交
527
	if (prepad + len < PAGE_SIZE || !linear)
528
		linear = len;
529

530 531
	skb = sock_alloc_send_pskb(sk, prepad + linear, len - linear, noblock,
				   &err);
532
	if (!skb)
533
		return ERR_PTR(err);
534 535 536

	skb_reserve(skb, prepad);
	skb_put(skb, linear);
537 538
	skb->data_len = len - linear;
	skb->len += len - linear;
539 540 541 542

	return skb;
}

L
Linus Torvalds 已提交
543
/* Get packet from user space buffer */
544
static __inline__ ssize_t tun_get_user(struct tun_struct *tun,
545
				       const struct iovec *iv, size_t count,
546
				       int noblock)
L
Linus Torvalds 已提交
547
{
548
	struct tun_pi pi = { 0, cpu_to_be16(ETH_P_IP) };
L
Linus Torvalds 已提交
549 550
	struct sk_buff *skb;
	size_t len = count, align = 0;
551
	struct virtio_net_hdr gso = { 0 };
552
	int offset = 0;
L
Linus Torvalds 已提交
553 554 555 556 557

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

558
		if (memcpy_fromiovecend((void *)&pi, iv, 0, sizeof(pi)))
L
Linus Torvalds 已提交
559
			return -EFAULT;
560
		offset += sizeof(pi);
L
Linus Torvalds 已提交
561 562
	}

563 564 565 566
	if (tun->flags & TUN_VNET_HDR) {
		if ((len -= sizeof(gso)) > count)
			return -EINVAL;

567
		if (memcpy_fromiovecend((void *)&gso, iv, offset, sizeof(gso)))
568 569
			return -EFAULT;

570 571 572 573
		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;

574 575
		if (gso.hdr_len > len)
			return -EINVAL;
576
		offset += sizeof(gso);
577 578
	}

579
	if ((tun->flags & TUN_TYPE_MASK) == TUN_TAP_DEV) {
L
Linus Torvalds 已提交
580
		align = NET_IP_ALIGN;
H
Herbert Xu 已提交
581 582
		if (unlikely(len < ETH_HLEN ||
			     (gso.hdr_len && gso.hdr_len < ETH_HLEN)))
583 584
			return -EINVAL;
	}
585

586 587 588 589 590
	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
591 592
	}

593
	if (skb_copy_datagram_from_iovec(skb, 0, iv, offset, len)) {
594
		tun->dev->stats.rx_dropped++;
D
Dave Jones 已提交
595
		kfree_skb(skb);
L
Linus Torvalds 已提交
596
		return -EFAULT;
D
Dave Jones 已提交
597
	}
L
Linus Torvalds 已提交
598

599 600 601 602 603 604 605 606 607 608
	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
609 610
	switch (tun->flags & TUN_TYPE_MASK) {
	case TUN_TUN_DEV:
611 612 613 614 615 616 617 618 619 620 621 622 623 624 625
		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;
			}
		}

626
		skb_reset_mac_header(skb);
L
Linus Torvalds 已提交
627
		skb->protocol = pi.proto;
628
		skb->dev = tun->dev;
L
Linus Torvalds 已提交
629 630 631 632 633 634
		break;
	case TUN_TAP_DEV:
		skb->protocol = eth_type_trans(skb, tun->dev);
		break;
	};

635 636 637 638 639 640 641 642 643
	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;
644 645 646
		case VIRTIO_NET_HDR_GSO_UDP:
			skb_shinfo(skb)->gso_type = SKB_GSO_UDP;
			break;
647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666
		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;
	}
667

L
Linus Torvalds 已提交
668
	netif_rx_ni(skb);
669

670 671
	tun->dev->stats.rx_packets++;
	tun->dev->stats.rx_bytes += len;
L
Linus Torvalds 已提交
672 673

	return count;
674
}
L
Linus Torvalds 已提交
675

676 677
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
678
{
679
	struct file *file = iocb->ki_filp;
H
Herbert Xu 已提交
680
	struct tun_struct *tun = tun_get(file);
E
Eric W. Biederman 已提交
681
	ssize_t result;
L
Linus Torvalds 已提交
682 683 684 685 686 687

	if (!tun)
		return -EBADFD;

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

688
	result = tun_get_user(tun, iv, iov_length(iv, count),
689
			      file->f_flags & O_NONBLOCK);
E
Eric W. Biederman 已提交
690 691 692

	tun_put(tun);
	return result;
L
Linus Torvalds 已提交
693 694 695 696 697
}

/* Put packet to the user space buffer */
static __inline__ ssize_t tun_put_user(struct tun_struct *tun,
				       struct sk_buff *skb,
698
				       const struct iovec *iv, int len)
L
Linus Torvalds 已提交
699 700 701 702 703 704 705 706 707 708 709 710
{
	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;
		}
711

712
		if (memcpy_toiovecend(iv, (void *) &pi, 0, sizeof(pi)))
L
Linus Torvalds 已提交
713 714
			return -EFAULT;
		total += sizeof(pi);
715
	}
L
Linus Torvalds 已提交
716

717 718 719 720 721 722 723 724 725 726 727 728 729 730 731
	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;
732 733
			else if (sinfo->gso_type & SKB_GSO_UDP)
				gso.gso_type = VIRTIO_NET_HDR_GSO_UDP;
734 735 736 737 738 739 740 741 742 743 744 745 746
			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 */

747 748
		if (unlikely(memcpy_toiovecend(iv, (void *)&gso, total,
					       sizeof(gso))))
749 750 751 752
			return -EFAULT;
		total += sizeof(gso);
	}

L
Linus Torvalds 已提交
753 754
	len = min_t(int, skb->len, len);

755
	skb_copy_datagram_const_iovec(skb, 0, iv, total, len);
L
Linus Torvalds 已提交
756 757
	total += len;

758 759
	tun->dev->stats.tx_packets++;
	tun->dev->stats.tx_bytes += len;
L
Linus Torvalds 已提交
760 761 762 763

	return total;
}

764 765
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
766
{
767
	struct file *file = iocb->ki_filp;
768 769
	struct tun_file *tfile = file->private_data;
	struct tun_struct *tun = __tun_get(tfile);
L
Linus Torvalds 已提交
770 771 772 773 774 775 776 777 778
	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
779
	len = iov_length(iv, count);
E
Eric W. Biederman 已提交
780 781 782 783
	if (len < 0) {
		ret = -EINVAL;
		goto out;
	}
L
Linus Torvalds 已提交
784

785
	add_wait_queue(&tun->socket.wait, &wait);
L
Linus Torvalds 已提交
786 787 788 789 790 791 792 793 794 795 796 797 798
	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;
			}
799 800 801 802
			if (tun->dev->reg_state != NETREG_REGISTERED) {
				ret = -EIO;
				break;
			}
L
Linus Torvalds 已提交
803 804 805 806 807 808 809

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

810
		ret = tun_put_user(tun, skb, iv, len);
811 812
		kfree_skb(skb);
		break;
L
Linus Torvalds 已提交
813 814 815
	}

	current->state = TASK_RUNNING;
816
	remove_wait_queue(&tun->socket.wait, &wait);
L
Linus Torvalds 已提交
817

E
Eric W. Biederman 已提交
818 819
out:
	tun_put(tun);
L
Linus Torvalds 已提交
820 821 822 823 824 825 826 827 828 829
	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;
830
	tun->group = -1;
L
Linus Torvalds 已提交
831 832

	dev->ethtool_ops = &tun_ethtool_ops;
833
	dev->destructor = tun_free_netdev;
L
Linus Torvalds 已提交
834 835
}

836 837 838 839 840 841 842 843 844 845 846 847 848 849 850
/* 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,
};

851 852 853 854 855 856 857 858 859 860
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
861 862 863
	if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
		wake_up_interruptible_sync(sk->sk_sleep);

864 865 866 867 868 869
	tun = container_of(sk, struct tun_sock, sk)->tun;
	kill_fasync(&tun->fasync, SIGIO, POLL_OUT);
}

static void tun_sock_destruct(struct sock *sk)
{
870
	free_netdev(container_of(sk, struct tun_sock, sk)->tun->dev);
871 872 873 874 875 876 877
}

static struct proto tun_proto = {
	.name		= "tun",
	.owner		= THIS_MODULE,
	.obj_size	= sizeof(struct tun_sock),
};
878

879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924
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);

925
static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
L
Linus Torvalds 已提交
926
{
927
	struct sock *sk;
L
Linus Torvalds 已提交
928 929 930 931
	struct tun_struct *tun;
	struct net_device *dev;
	int err;

932 933
	dev = __dev_get_by_name(net, ifr->ifr_name);
	if (dev) {
934 935
		if (ifr->ifr_flags & IFF_TUN_EXCL)
			return -EBUSY;
936 937 938 939 940 941 942
		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;

943 944 945
		err = tun_attach(tun, file);
		if (err < 0)
			return err;
946
	}
L
Linus Torvalds 已提交
947 948 949 950 951 952
	else {
		char *name;
		unsigned long flags = 0;

		err = -EINVAL;

953 954 955
		if (!capable(CAP_NET_ADMIN))
			return -EPERM;

L
Linus Torvalds 已提交
956 957 958 959 960 961 962 963 964
		/* 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";
965
		} else
L
Linus Torvalds 已提交
966
			goto failed;
967

L
Linus Torvalds 已提交
968 969 970 971 972 973 974 975
		if (*ifr->ifr_name)
			name = ifr->ifr_name;

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

976
		dev_net_set(dev, net);
977
		dev->rtnl_link_ops = &tun_link_ops;
S
Stephen Hemminger 已提交
978

L
Linus Torvalds 已提交
979 980 981
		tun = netdev_priv(dev);
		tun->dev = dev;
		tun->flags = flags;
982
		tun->txflt.count = 0;
L
Linus Torvalds 已提交
983

984 985 986 987 988
		err = -ENOMEM;
		sk = sk_alloc(net, AF_UNSPEC, GFP_KERNEL, &tun_proto);
		if (!sk)
			goto err_free_dev;

989
		init_waitqueue_head(&tun->socket.wait);
990 991 992 993 994 995 996
		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
997 998 999 1000 1001
		tun_net_init(dev);

		if (strchr(dev->name, '%')) {
			err = dev_alloc_name(dev, dev->name);
			if (err < 0)
1002
				goto err_free_sk;
L
Linus Torvalds 已提交
1003 1004 1005 1006
		}

		err = register_netdevice(tun->dev);
		if (err < 0)
1007 1008
			goto err_free_sk;

1009 1010 1011 1012 1013
		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");

1014
		sk->sk_destruct = tun_sock_destruct;
1015 1016 1017

		err = tun_attach(tun, file);
		if (err < 0)
1018
			goto failed;
L
Linus Torvalds 已提交
1019 1020 1021 1022 1023 1024
	}

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

	if (ifr->ifr_flags & IFF_NO_PI)
		tun->flags |= TUN_NO_PI;
1025 1026
	else
		tun->flags &= ~TUN_NO_PI;
L
Linus Torvalds 已提交
1027 1028 1029

	if (ifr->ifr_flags & IFF_ONE_QUEUE)
		tun->flags |= TUN_ONE_QUEUE;
1030 1031
	else
		tun->flags &= ~TUN_ONE_QUEUE;
L
Linus Torvalds 已提交
1032

1033 1034 1035 1036 1037
	if (ifr->ifr_flags & IFF_VNET_HDR)
		tun->flags |= TUN_VNET_HDR;
	else
		tun->flags &= ~TUN_VNET_HDR;

1038 1039 1040 1041 1042 1043
	/* 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
1044 1045 1046
	strcpy(ifr->ifr_name, tun->dev->name);
	return 0;

1047 1048
 err_free_sk:
	sock_put(sk);
L
Linus Torvalds 已提交
1049 1050 1051 1052 1053 1054
 err_free_dev:
	free_netdev(dev);
 failed:
	return err;
}

1055 1056
static int tun_get_iff(struct net *net, struct file *file, struct ifreq *ifr)
{
E
Eric W. Biederman 已提交
1057
	struct tun_struct *tun = tun_get(file);
1058 1059 1060 1061 1062 1063 1064 1065

	if (!tun)
		return -EBADFD;

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

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

1066
	ifr->ifr_flags = tun_flags(tun);
1067

E
Eric W. Biederman 已提交
1068
	tun_put(tun);
1069 1070 1071
	return 0;
}

1072 1073 1074 1075 1076 1077 1078 1079 1080
/* 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
1081 1082
				    |NETIF_F_TSO_ECN|NETIF_F_TSO|NETIF_F_TSO6
				    |NETIF_F_UFO));
1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098

	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);
		}
1099 1100 1101 1102 1103

		if (arg & TUN_F_UFO) {
			features |= NETIF_F_UFO;
			arg &= ~TUN_F_UFO;
		}
1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117
	}

	/* 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;
}

1118
static int tun_chr_ioctl(struct inode *inode, struct file *file,
L
Linus Torvalds 已提交
1119 1120
			 unsigned int cmd, unsigned long arg)
{
E
Eric W. Biederman 已提交
1121
	struct tun_file *tfile = file->private_data;
E
Eric W. Biederman 已提交
1122
	struct tun_struct *tun;
L
Linus Torvalds 已提交
1123 1124
	void __user* argp = (void __user*)arg;
	struct ifreq ifr;
1125
	int sndbuf;
1126
	int ret;
L
Linus Torvalds 已提交
1127 1128 1129 1130 1131

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

E
Eric W. Biederman 已提交
1132 1133 1134 1135 1136 1137 1138 1139 1140
	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
1141
	tun = __tun_get(tfile);
L
Linus Torvalds 已提交
1142 1143 1144 1145 1146 1147
	if (cmd == TUNSETIFF && !tun) {
		int err;

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

		rtnl_lock();
E
Eric W. Biederman 已提交
1148
		err = tun_set_iff(tfile->net, file, &ifr);
L
Linus Torvalds 已提交
1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
		rtnl_unlock();

		if (err)
			return err;

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

1159

L
Linus Torvalds 已提交
1160 1161 1162 1163 1164
	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
1165
	ret = 0;
L
Linus Torvalds 已提交
1166
	switch (cmd) {
1167 1168 1169
	case TUNGETIFF:
		ret = tun_get_iff(current->nsproxy->net_ns, file, &ifr);
		if (ret)
E
Eric W. Biederman 已提交
1170
			break;
1171 1172

		if (copy_to_user(argp, &ifr, sizeof(ifr)))
E
Eric W. Biederman 已提交
1173
			ret = -EFAULT;
1174 1175
		break;

L
Linus Torvalds 已提交
1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194
	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
1195
		    tun->dev->name, arg ? "enabled" : "disabled");
L
Linus Torvalds 已提交
1196 1197 1198 1199 1200 1201 1202 1203 1204
		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;

1205 1206 1207 1208 1209 1210 1211
	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;

1212 1213
	case TUNSETLINK:
		/* Only allow setting the type when the interface is down */
1214
		rtnl_lock();
1215 1216 1217
		if (tun->dev->flags & IFF_UP) {
			DBG(KERN_INFO "%s: Linktype set failed because interface is up\n",
				tun->dev->name);
1218
			ret = -EBUSY;
1219 1220 1221
		} else {
			tun->dev->type = (int) arg;
			DBG(KERN_INFO "%s: linktype set to %d\n", tun->dev->name, tun->dev->type);
1222
			ret = 0;
1223
		}
1224
		rtnl_unlock();
E
Eric W. Biederman 已提交
1225
		break;
1226

L
Linus Torvalds 已提交
1227 1228 1229 1230 1231
#ifdef TUN_DEBUG
	case TUNSETDEBUG:
		tun->debug = arg;
		break;
#endif
1232 1233 1234 1235
	case TUNSETOFFLOAD:
		rtnl_lock();
		ret = set_offload(tun->dev, arg);
		rtnl_unlock();
E
Eric W. Biederman 已提交
1236
		break;
1237

1238 1239
	case TUNSETTXFILTER:
		/* Can be set only for TAPs */
E
Eric W. Biederman 已提交
1240
		ret = -EINVAL;
1241
		if ((tun->flags & TUN_TYPE_MASK) != TUN_TAP_DEV)
E
Eric W. Biederman 已提交
1242
			break;
1243
		rtnl_lock();
H
Harvey Harrison 已提交
1244
		ret = update_filter(&tun->txflt, (void __user *)arg);
1245
		rtnl_unlock();
E
Eric W. Biederman 已提交
1246
		break;
L
Linus Torvalds 已提交
1247 1248

	case SIOCGIFHWADDR:
1249 1250 1251 1252
		/* 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
1253 1254
			ret = -EFAULT;
		break;
L
Linus Torvalds 已提交
1255 1256

	case SIOCSIFHWADDR:
1257
		/* Set hw address */
J
Johannes Berg 已提交
1258 1259
		DBG(KERN_DEBUG "%s: set hw address: %pM\n",
			tun->dev->name, ifr.ifr_hwaddr.sa_data);
1260 1261 1262 1263

		rtnl_lock();
		ret = dev_set_mac_address(tun->dev, &ifr.ifr_hwaddr);
		rtnl_unlock();
E
Eric W. Biederman 已提交
1264
		break;
1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280

	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
1281
	default:
E
Eric W. Biederman 已提交
1282 1283
		ret = -EINVAL;
		break;
L
Linus Torvalds 已提交
1284 1285
	};

E
Eric W. Biederman 已提交
1286 1287
	tun_put(tun);
	return ret;
L
Linus Torvalds 已提交
1288 1289 1290 1291
}

static int tun_chr_fasync(int fd, struct file *file, int on)
{
E
Eric W. Biederman 已提交
1292
	struct tun_struct *tun = tun_get(file);
L
Linus Torvalds 已提交
1293 1294 1295 1296 1297 1298 1299
	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
1300
	lock_kernel();
L
Linus Torvalds 已提交
1301
	if ((ret = fasync_helper(fd, file, on, &tun->fasync)) < 0)
J
Jonathan Corbet 已提交
1302
		goto out;
1303

L
Linus Torvalds 已提交
1304
	if (on) {
1305
		ret = __f_setown(file, task_pid(current), PIDTYPE_PID, 0);
L
Linus Torvalds 已提交
1306
		if (ret)
J
Jonathan Corbet 已提交
1307
			goto out;
L
Linus Torvalds 已提交
1308
		tun->flags |= TUN_FASYNC;
1309
	} else
L
Linus Torvalds 已提交
1310
		tun->flags &= ~TUN_FASYNC;
J
Jonathan Corbet 已提交
1311 1312 1313
	ret = 0;
out:
	unlock_kernel();
E
Eric W. Biederman 已提交
1314
	tun_put(tun);
J
Jonathan Corbet 已提交
1315
	return ret;
L
Linus Torvalds 已提交
1316 1317 1318 1319
}

static int tun_chr_open(struct inode *inode, struct file * file)
{
E
Eric W. Biederman 已提交
1320
	struct tun_file *tfile;
A
Arnd Bergmann 已提交
1321
	cycle_kernel_lock();
L
Linus Torvalds 已提交
1322
	DBG1(KERN_INFO "tunX: tun_chr_open\n");
E
Eric W. Biederman 已提交
1323 1324 1325 1326

	tfile = kmalloc(sizeof(*tfile), GFP_KERNEL);
	if (!tfile)
		return -ENOMEM;
1327
	atomic_set(&tfile->count, 0);
E
Eric W. Biederman 已提交
1328
	tfile->tun = NULL;
E
Eric W. Biederman 已提交
1329
	tfile->net = get_net(current->nsproxy->net_ns);
E
Eric W. Biederman 已提交
1330
	file->private_data = tfile;
L
Linus Torvalds 已提交
1331 1332 1333 1334 1335
	return 0;
}

static int tun_chr_close(struct inode *inode, struct file *file)
{
E
Eric W. Biederman 已提交
1336
	struct tun_file *tfile = file->private_data;
E
Eric W. Biederman 已提交
1337
	struct tun_struct *tun;
L
Linus Torvalds 已提交
1338

E
Eric W. Biederman 已提交
1339
	tun = __tun_get(tfile);
E
Eric W. Biederman 已提交
1340
	if (tun) {
H
Herbert Xu 已提交
1341 1342 1343
		struct net_device *dev = tun->dev;

		DBG(KERN_INFO "%s: tun_chr_close\n", dev->name);
L
Linus Torvalds 已提交
1344

E
Eric W. Biederman 已提交
1345
		__tun_detach(tun);
L
Linus Torvalds 已提交
1346

E
Eric W. Biederman 已提交
1347
		/* If desireable, unregister the netdevice. */
H
Herbert Xu 已提交
1348 1349 1350 1351 1352 1353
		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
1354
	}
L
Linus Torvalds 已提交
1355

1356 1357 1358 1359
	tun = tfile->tun;
	if (tun)
		sock_put(tun->sk);

E
Eric W. Biederman 已提交
1360
	put_net(tfile->net);
E
Eric W. Biederman 已提交
1361
	kfree(tfile);
L
Linus Torvalds 已提交
1362 1363 1364 1365

	return 0;
}

1366
static const struct file_operations tun_fops = {
1367
	.owner	= THIS_MODULE,
L
Linus Torvalds 已提交
1368
	.llseek = no_llseek,
1369 1370 1371 1372
	.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
1373 1374 1375 1376
	.poll	= tun_chr_poll,
	.ioctl	= tun_chr_ioctl,
	.open	= tun_chr_open,
	.release = tun_chr_close,
1377
	.fasync = tun_chr_fasync
L
Linus Torvalds 已提交
1378 1379 1380 1381 1382
};

static struct miscdevice tun_miscdev = {
	.minor = TUN_MINOR,
	.name = "tun",
1383
	.devnode = "net/tun",
L
Linus Torvalds 已提交
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 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442
	.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
1443
	return !!tun->tfile;
L
Linus Torvalds 已提交
1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461
}

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;
}

1462
static const struct ethtool_ops tun_ethtool_ops = {
L
Linus Torvalds 已提交
1463 1464 1465 1466 1467 1468 1469 1470 1471
	.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
};

1472

L
Linus Torvalds 已提交
1473 1474 1475 1476 1477 1478 1479
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);

1480
	ret = rtnl_link_register(&tun_link_ops);
1481
	if (ret) {
1482 1483
		printk(KERN_ERR "tun: Can't register link_ops\n");
		goto err_linkops;
1484 1485
	}

L
Linus Torvalds 已提交
1486
	ret = misc_register(&tun_miscdev);
1487
	if (ret) {
L
Linus Torvalds 已提交
1488
		printk(KERN_ERR "tun: Can't register misc device %d\n", TUN_MINOR);
1489 1490
		goto err_misc;
	}
1491
	return  0;
1492
err_misc:
1493 1494
	rtnl_link_unregister(&tun_link_ops);
err_linkops:
L
Linus Torvalds 已提交
1495 1496 1497 1498 1499
	return ret;
}

static void tun_cleanup(void)
{
1500
	misc_deregister(&tun_miscdev);
1501
	rtnl_link_unregister(&tun_link_ops);
L
Linus Torvalds 已提交
1502 1503 1504 1505 1506 1507 1508 1509
}

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