tun.c 35.9 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 47 48 49 50 51 52 53 54 55
 *
 *  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>
#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>
56
#include <linux/compat.h>
L
Linus Torvalds 已提交
57 58 59 60 61
#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
	uid_t			owner;
	gid_t			group;

	struct net_device	*dev;
107
	struct fasync_struct	*fasync;
108

109
	struct tap_filter       txflt;
110
	struct socket		socket;
111 112 113

#ifdef TUN_DEBUG
	int debug;
L
Linus Torvalds 已提交
114
#endif
115
};
L
Linus Torvalds 已提交
116

117 118 119 120 121 122 123 124 125 126
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);
}

127 128
static int tun_attach(struct tun_struct *tun, struct file *file)
{
E
Eric W. Biederman 已提交
129
	struct tun_file *tfile = file->private_data;
130
	int err;
131 132 133

	ASSERT_RTNL();

134 135 136 137 138 139 140 141 142 143 144
	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
145 146
	tfile->tun = tun;
	tun->tfile = tfile;
147
	tun->socket.file = file;
148
	dev_hold(tun->dev);
149
	sock_hold(tun->socket.sk);
150
	atomic_inc(&tfile->count);
151

152 153 154
out:
	netif_tx_unlock_bh(tun->dev);
	return err;
155 156
}

E
Eric W. Biederman 已提交
157 158 159
static void __tun_detach(struct tun_struct *tun)
{
	/* Detach from net device */
160
	netif_tx_lock_bh(tun->dev);
E
Eric W. Biederman 已提交
161
	tun->tfile = NULL;
162
	tun->socket.file = NULL;
163
	netif_tx_unlock_bh(tun->dev);
E
Eric W. Biederman 已提交
164 165

	/* Drop read queue */
166
	skb_queue_purge(&tun->socket.sk->sk_receive_queue);
167 168 169 170 171 172 173 174 175 176

	/* 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
177 178 179 180
}

static struct tun_struct *__tun_get(struct tun_file *tfile)
{
181 182 183 184 185 186
	struct tun_struct *tun = NULL;

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

	return tun;
E
Eric W. Biederman 已提交
187 188 189 190 191 192 193 194 195
}

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

static void tun_put(struct tun_struct *tun)
{
196 197 198 199
	struct tun_file *tfile = tun->tfile;

	if (atomic_dec_and_test(&tfile->count))
		tun_detach(tfile->tun);
E
Eric W. Biederman 已提交
200 201
}

202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251
/* 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;

252 253
	/* Remaining multicast addresses are hashed,
	 * unicast will leave the filter disabled. */
254
	memset(filter->mask, 0, sizeof(filter->mask));
255 256 257 258 259
	for (; n < uf.count; n++) {
		if (!is_multicast_ether_addr(addr[n].u)) {
			err = 0; /* no filter */
			goto done;
		}
260
		addr_hash_set(filter->mask, addr[n].u);
261
	}
262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311

	/* 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
312 313
/* Network device part of the driver */

314
static const struct ethtool_ops tun_ethtool_ops;
L
Linus Torvalds 已提交
315

316 317 318 319 320 321 322 323 324
/* 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) {
325
		wake_up_all(&tun->socket.wait);
326 327 328 329 330
		if (atomic_dec_and_test(&tfile->count))
			__tun_detach(tun);
	}
}

331 332 333 334
static void tun_free_netdev(struct net_device *dev)
{
	struct tun_struct *tun = netdev_priv(dev);

335
	sock_put(tun->socket.sk);
336 337
}

L
Linus Torvalds 已提交
338 339 340 341 342 343 344 345 346 347 348 349 350 351 352
/* 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 */
353
static netdev_tx_t tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
L
Linus Torvalds 已提交
354 355 356 357 358 359
{
	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
360
	if (!tun->tfile)
L
Linus Torvalds 已提交
361 362
		goto drop;

363 364 365 366 367 368
	/* 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;

369
	if (skb_queue_len(&tun->socket.sk->sk_receive_queue) >= dev->tx_queue_len) {
L
Linus Torvalds 已提交
370 371 372 373 374 375 376
		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. */
377
			dev->stats.tx_fifo_errors++;
L
Linus Torvalds 已提交
378 379 380 381 382 383 384
		} else {
			/* Single queue mode.
			 * Driver handles dropping of all packets itself. */
			goto drop;
		}
	}

385
	/* Enqueue packet */
386
	skb_queue_tail(&tun->socket.sk->sk_receive_queue, skb);
L
Linus Torvalds 已提交
387 388 389 390 391
	dev->trans_start = jiffies;

	/* Notify and wake up reader process */
	if (tun->flags & TUN_FASYNC)
		kill_fasync(&tun->fasync, SIGIO, POLL_IN);
392 393
	wake_up_interruptible_poll(&tun->socket.wait, POLLIN |
				   POLLRDNORM | POLLRDBAND);
394
	return NETDEV_TX_OK;
L
Linus Torvalds 已提交
395 396

drop:
397
	dev->stats.tx_dropped++;
L
Linus Torvalds 已提交
398
	kfree_skb(skb);
399
	return NETDEV_TX_OK;
L
Linus Torvalds 已提交
400 401
}

402
static void tun_net_mclist(struct net_device *dev)
L
Linus Torvalds 已提交
403
{
404 405 406 407 408 409
	/*
	 * 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
410 411
}

E
Ed Swierk 已提交
412 413 414 415 416 417 418 419 420 421 422 423
#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
424
static const struct net_device_ops tun_netdev_ops = {
425
	.ndo_uninit		= tun_net_uninit,
S
Stephen Hemminger 已提交
426 427
	.ndo_open		= tun_net_open,
	.ndo_stop		= tun_net_close,
428
	.ndo_start_xmit		= tun_net_xmit,
S
Stephen Hemminger 已提交
429 430 431 432
	.ndo_change_mtu		= tun_net_change_mtu,
};

static const struct net_device_ops tap_netdev_ops = {
433
	.ndo_uninit		= tun_net_uninit,
S
Stephen Hemminger 已提交
434 435
	.ndo_open		= tun_net_open,
	.ndo_stop		= tun_net_close,
436
	.ndo_start_xmit		= tun_net_xmit,
S
Stephen Hemminger 已提交
437 438 439 440 441 442
	.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
443 444 445 446
/* Initialize net device. */
static void tun_net_init(struct net_device *dev)
{
	struct tun_struct *tun = netdev_priv(dev);
447

L
Linus Torvalds 已提交
448 449
	switch (tun->flags & TUN_TYPE_MASK) {
	case TUN_TUN_DEV:
S
Stephen Hemminger 已提交
450 451
		dev->netdev_ops = &tun_netdev_ops;

L
Linus Torvalds 已提交
452 453 454 455 456 457
		/* Point-to-Point TUN Device */
		dev->hard_header_len = 0;
		dev->addr_len = 0;
		dev->mtu = 1500;

		/* Zero header length */
458
		dev->type = ARPHRD_NONE;
L
Linus Torvalds 已提交
459 460 461 462 463
		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
464
		dev->netdev_ops = &tap_netdev_ops;
L
Linus Torvalds 已提交
465 466
		/* Ethernet TAP Device */
		ether_setup(dev);
467

468
		random_ether_addr(dev->dev_addr);
469

L
Linus Torvalds 已提交
470 471 472 473 474 475 476 477 478
		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)
479
{
480 481
	struct tun_file *tfile = file->private_data;
	struct tun_struct *tun = __tun_get(tfile);
482
	struct sock *sk;
483
	unsigned int mask = 0;
L
Linus Torvalds 已提交
484 485

	if (!tun)
486
		return POLLERR;
L
Linus Torvalds 已提交
487

488
	sk = tun->socket.sk;
489

L
Linus Torvalds 已提交
490 491
	DBG(KERN_INFO "%s: tun_chr_poll\n", tun->dev->name);

492
	poll_wait(file, &tun->socket.wait, wait);
493

494
	if (!skb_queue_empty(&sk->sk_receive_queue))
L
Linus Torvalds 已提交
495 496
		mask |= POLLIN | POLLRDNORM;

497 498 499 500 501
	if (sock_writeable(sk) ||
	    (!test_and_set_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags) &&
	     sock_writeable(sk)))
		mask |= POLLOUT | POLLWRNORM;

502 503 504
	if (tun->dev->reg_state != NETREG_REGISTERED)
		mask = POLLERR;

E
Eric W. Biederman 已提交
505
	tun_put(tun);
L
Linus Torvalds 已提交
506 507 508
	return mask;
}

509 510
/* 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). */
511 512 513
static inline struct sk_buff *tun_alloc_skb(struct tun_struct *tun,
					    size_t prepad, size_t len,
					    size_t linear, int noblock)
514
{
515
	struct sock *sk = tun->socket.sk;
516
	struct sk_buff *skb;
517
	int err;
518 519

	/* Under a page?  Don't bother with paged skb. */
H
Herbert Xu 已提交
520
	if (prepad + len < PAGE_SIZE || !linear)
521
		linear = len;
522

523 524
	skb = sock_alloc_send_pskb(sk, prepad + linear, len - linear, noblock,
				   &err);
525
	if (!skb)
526
		return ERR_PTR(err);
527 528 529

	skb_reserve(skb, prepad);
	skb_put(skb, linear);
530 531
	skb->data_len = len - linear;
	skb->len += len - linear;
532 533 534 535

	return skb;
}

L
Linus Torvalds 已提交
536
/* Get packet from user space buffer */
537
static __inline__ ssize_t tun_get_user(struct tun_struct *tun,
538
				       const struct iovec *iv, size_t count,
539
				       int noblock)
L
Linus Torvalds 已提交
540
{
541
	struct tun_pi pi = { 0, cpu_to_be16(ETH_P_IP) };
L
Linus Torvalds 已提交
542 543
	struct sk_buff *skb;
	size_t len = count, align = 0;
544
	struct virtio_net_hdr gso = { 0 };
545
	int offset = 0;
L
Linus Torvalds 已提交
546 547 548 549 550

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

551
		if (memcpy_fromiovecend((void *)&pi, iv, 0, sizeof(pi)))
L
Linus Torvalds 已提交
552
			return -EFAULT;
553
		offset += sizeof(pi);
L
Linus Torvalds 已提交
554 555
	}

556 557 558 559
	if (tun->flags & TUN_VNET_HDR) {
		if ((len -= sizeof(gso)) > count)
			return -EINVAL;

560
		if (memcpy_fromiovecend((void *)&gso, iv, offset, sizeof(gso)))
561 562
			return -EFAULT;

563 564 565 566
		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;

567 568
		if (gso.hdr_len > len)
			return -EINVAL;
569
		offset += sizeof(gso);
570 571
	}

572
	if ((tun->flags & TUN_TYPE_MASK) == TUN_TAP_DEV) {
L
Linus Torvalds 已提交
573
		align = NET_IP_ALIGN;
H
Herbert Xu 已提交
574 575
		if (unlikely(len < ETH_HLEN ||
			     (gso.hdr_len && gso.hdr_len < ETH_HLEN)))
576 577
			return -EINVAL;
	}
578

579 580 581 582 583
	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
584 585
	}

586
	if (skb_copy_datagram_from_iovec(skb, 0, iv, offset, len)) {
587
		tun->dev->stats.rx_dropped++;
D
Dave Jones 已提交
588
		kfree_skb(skb);
L
Linus Torvalds 已提交
589
		return -EFAULT;
D
Dave Jones 已提交
590
	}
L
Linus Torvalds 已提交
591

592 593 594 595 596 597 598 599 600 601
	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
602 603
	switch (tun->flags & TUN_TYPE_MASK) {
	case TUN_TUN_DEV:
604 605 606 607 608 609 610 611 612 613 614 615 616 617 618
		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;
			}
		}

619
		skb_reset_mac_header(skb);
L
Linus Torvalds 已提交
620
		skb->protocol = pi.proto;
621
		skb->dev = tun->dev;
L
Linus Torvalds 已提交
622 623 624 625 626 627
		break;
	case TUN_TAP_DEV:
		skb->protocol = eth_type_trans(skb, tun->dev);
		break;
	};

628 629 630 631 632 633 634 635 636
	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;
637 638 639
		case VIRTIO_NET_HDR_GSO_UDP:
			skb_shinfo(skb)->gso_type = SKB_GSO_UDP;
			break;
640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659
		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;
	}
660

L
Linus Torvalds 已提交
661
	netif_rx_ni(skb);
662

663 664
	tun->dev->stats.rx_packets++;
	tun->dev->stats.rx_bytes += len;
L
Linus Torvalds 已提交
665 666

	return count;
667
}
L
Linus Torvalds 已提交
668

669 670
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
671
{
672
	struct file *file = iocb->ki_filp;
H
Herbert Xu 已提交
673
	struct tun_struct *tun = tun_get(file);
E
Eric W. Biederman 已提交
674
	ssize_t result;
L
Linus Torvalds 已提交
675 676 677 678 679 680

	if (!tun)
		return -EBADFD;

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

681
	result = tun_get_user(tun, iv, iov_length(iv, count),
682
			      file->f_flags & O_NONBLOCK);
E
Eric W. Biederman 已提交
683 684 685

	tun_put(tun);
	return result;
L
Linus Torvalds 已提交
686 687 688 689 690
}

/* Put packet to the user space buffer */
static __inline__ ssize_t tun_put_user(struct tun_struct *tun,
				       struct sk_buff *skb,
691
				       const struct iovec *iv, int len)
L
Linus Torvalds 已提交
692 693 694 695 696 697 698 699 700 701 702 703
{
	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;
		}
704

705
		if (memcpy_toiovecend(iv, (void *) &pi, 0, sizeof(pi)))
L
Linus Torvalds 已提交
706 707
			return -EFAULT;
		total += sizeof(pi);
708
	}
L
Linus Torvalds 已提交
709

710 711 712 713 714 715 716 717 718 719 720 721 722 723 724
	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;
725 726
			else if (sinfo->gso_type & SKB_GSO_UDP)
				gso.gso_type = VIRTIO_NET_HDR_GSO_UDP;
727 728 729 730 731 732 733 734 735 736 737 738 739
			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);
749
	total += skb->len;
L
Linus Torvalds 已提交
750

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;
}

757 758 759
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
760 761 762
{
	DECLARE_WAITQUEUE(wait, current);
	struct sk_buff *skb;
763
	ssize_t ret = 0;
L
Linus Torvalds 已提交
764 765 766

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

767
	add_wait_queue(&tun->socket.wait, &wait);
L
Linus Torvalds 已提交
768 769 770 771
	while (len) {
		current->state = TASK_INTERRUPTIBLE;

		/* Read frames from the queue */
772
		if (!(skb=skb_dequeue(&tun->socket.sk->sk_receive_queue))) {
773
			if (noblock) {
L
Linus Torvalds 已提交
774 775 776 777 778 779 780
				ret = -EAGAIN;
				break;
			}
			if (signal_pending(current)) {
				ret = -ERESTARTSYS;
				break;
			}
781 782 783 784
			if (tun->dev->reg_state != NETREG_REGISTERED) {
				ret = -EIO;
				break;
			}
L
Linus Torvalds 已提交
785 786 787 788 789 790 791

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

792
		ret = tun_put_user(tun, skb, iv, len);
793 794
		kfree_skb(skb);
		break;
L
Linus Torvalds 已提交
795 796 797
	}

	current->state = TASK_RUNNING;
798
	remove_wait_queue(&tun->socket.wait, &wait);
L
Linus Torvalds 已提交
799

800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820
	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
821 822
out:
	tun_put(tun);
L
Linus Torvalds 已提交
823 824 825 826 827 828 829 830
	return ret;
}

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

	tun->owner = -1;
831
	tun->group = -1;
L
Linus Torvalds 已提交
832 833

	dev->ethtool_ops = &tun_ethtool_ops;
834
	dev->destructor = tun_free_netdev;
L
Linus Torvalds 已提交
835 836
}

837 838 839 840 841 842 843 844 845 846 847 848 849 850 851
/* 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,
};

852 853 854 855 856 857 858 859 860 861
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
862
	if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
863 864
		wake_up_interruptible_sync_poll(sk->sk_sleep, POLLOUT |
						POLLWRNORM | POLLWRBAND);
H
Herbert Xu 已提交
865

866
	tun = tun_sk(sk)->tun;
867 868 869 870 871
	kill_fasync(&tun->fasync, SIGIO, POLL_OUT);
}

static void tun_sock_destruct(struct sock *sk)
{
872
	free_netdev(tun_sk(sk)->tun->dev);
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
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;
}

/* Ops structure to mimic raw sockets with tun */
static const struct proto_ops tun_socket_ops = {
	.sendmsg = tun_sendmsg,
	.recvmsg = tun_recvmsg,
};

906 907 908 909 910
static struct proto tun_proto = {
	.name		= "tun",
	.owner		= THIS_MODULE,
	.obj_size	= sizeof(struct tun_sock),
};
911

912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957
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);

958
static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
L
Linus Torvalds 已提交
959
{
960
	struct sock *sk;
L
Linus Torvalds 已提交
961 962 963 964
	struct tun_struct *tun;
	struct net_device *dev;
	int err;

965 966
	dev = __dev_get_by_name(net, ifr->ifr_name);
	if (dev) {
P
Paul Moore 已提交
967 968
		const struct cred *cred = current_cred();

969 970
		if (ifr->ifr_flags & IFF_TUN_EXCL)
			return -EBUSY;
971 972 973 974 975 976 977
		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
978 979 980 981
		if (((tun->owner != -1 && cred->euid != tun->owner) ||
		     (tun->group != -1 && !in_egroup_p(tun->group))) &&
		    !capable(CAP_NET_ADMIN))
			return -EPERM;
982
		err = security_tun_dev_attach(tun->socket.sk);
P
Paul Moore 已提交
983 984 985
		if (err < 0)
			return err;

986 987 988
		err = tun_attach(tun, file);
		if (err < 0)
			return err;
989
	}
L
Linus Torvalds 已提交
990 991 992 993
	else {
		char *name;
		unsigned long flags = 0;

994 995
		if (!capable(CAP_NET_ADMIN))
			return -EPERM;
P
Paul Moore 已提交
996 997 998
		err = security_tun_dev_create();
		if (err < 0)
			return err;
999

L
Linus Torvalds 已提交
1000 1001 1002 1003 1004 1005 1006 1007 1008
		/* 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";
1009
		} else
1010
			return -EINVAL;
1011

L
Linus Torvalds 已提交
1012 1013 1014 1015 1016 1017 1018 1019
		if (*ifr->ifr_name)
			name = ifr->ifr_name;

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

1020
		dev_net_set(dev, net);
1021
		dev->rtnl_link_ops = &tun_link_ops;
S
Stephen Hemminger 已提交
1022

L
Linus Torvalds 已提交
1023 1024 1025
		tun = netdev_priv(dev);
		tun->dev = dev;
		tun->flags = flags;
1026
		tun->txflt.count = 0;
L
Linus Torvalds 已提交
1027

1028 1029 1030 1031 1032
		err = -ENOMEM;
		sk = sk_alloc(net, AF_UNSPEC, GFP_KERNEL, &tun_proto);
		if (!sk)
			goto err_free_dev;

1033
		init_waitqueue_head(&tun->socket.wait);
1034
		tun->socket.ops = &tun_socket_ops;
1035 1036 1037 1038
		sock_init_data(&tun->socket, sk);
		sk->sk_write_space = tun_sock_write_space;
		sk->sk_sndbuf = INT_MAX;

1039
		tun_sk(sk)->tun = tun;
1040

P
Paul Moore 已提交
1041 1042
		security_tun_dev_post_create(sk);

L
Linus Torvalds 已提交
1043 1044 1045 1046 1047
		tun_net_init(dev);

		if (strchr(dev->name, '%')) {
			err = dev_alloc_name(dev, dev->name);
			if (err < 0)
1048
				goto err_free_sk;
L
Linus Torvalds 已提交
1049 1050 1051 1052
		}

		err = register_netdevice(tun->dev);
		if (err < 0)
1053 1054
			goto err_free_sk;

1055 1056 1057 1058 1059
		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");

1060
		sk->sk_destruct = tun_sock_destruct;
1061 1062 1063

		err = tun_attach(tun, file);
		if (err < 0)
1064
			goto failed;
L
Linus Torvalds 已提交
1065 1066 1067 1068 1069 1070
	}

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

	if (ifr->ifr_flags & IFF_NO_PI)
		tun->flags |= TUN_NO_PI;
1071 1072
	else
		tun->flags &= ~TUN_NO_PI;
L
Linus Torvalds 已提交
1073 1074 1075

	if (ifr->ifr_flags & IFF_ONE_QUEUE)
		tun->flags |= TUN_ONE_QUEUE;
1076 1077
	else
		tun->flags &= ~TUN_ONE_QUEUE;
L
Linus Torvalds 已提交
1078

1079 1080 1081 1082 1083
	if (ifr->ifr_flags & IFF_VNET_HDR)
		tun->flags |= TUN_VNET_HDR;
	else
		tun->flags &= ~TUN_VNET_HDR;

1084 1085 1086 1087 1088 1089
	/* 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
1090 1091 1092
	strcpy(ifr->ifr_name, tun->dev->name);
	return 0;

1093 1094
 err_free_sk:
	sock_put(sk);
L
Linus Torvalds 已提交
1095 1096 1097 1098 1099 1100
 err_free_dev:
	free_netdev(dev);
 failed:
	return err;
}

1101 1102
static int tun_get_iff(struct net *net, struct tun_struct *tun,
		       struct ifreq *ifr)
1103 1104 1105 1106 1107
{
	DBG(KERN_INFO "%s: tun_get_iff\n", tun->dev->name);

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

1108
	ifr->ifr_flags = tun_flags(tun);
1109 1110 1111 1112

	return 0;
}

1113 1114 1115 1116 1117 1118 1119 1120 1121
/* 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
1122 1123
				    |NETIF_F_TSO_ECN|NETIF_F_TSO|NETIF_F_TSO6
				    |NETIF_F_UFO));
1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139

	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);
		}
1140 1141 1142 1143 1144

		if (arg & TUN_F_UFO) {
			features |= NETIF_F_UFO;
			arg &= ~TUN_F_UFO;
		}
1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
	}

	/* 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;
}

1159 1160
static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
			    unsigned long arg, int ifreq_len)
L
Linus Torvalds 已提交
1161
{
E
Eric W. Biederman 已提交
1162
	struct tun_file *tfile = file->private_data;
E
Eric W. Biederman 已提交
1163
	struct tun_struct *tun;
L
Linus Torvalds 已提交
1164 1165
	void __user* argp = (void __user*)arg;
	struct ifreq ifr;
1166
	int sndbuf;
1167
	int ret;
L
Linus Torvalds 已提交
1168 1169

	if (cmd == TUNSETIFF || _IOC_TYPE(cmd) == 0x89)
1170
		if (copy_from_user(&ifr, argp, ifreq_len))
L
Linus Torvalds 已提交
1171 1172
			return -EFAULT;

E
Eric W. Biederman 已提交
1173 1174 1175 1176 1177 1178 1179 1180 1181
	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);
	}

1182 1183
	rtnl_lock();

E
Eric W. Biederman 已提交
1184
	tun = __tun_get(tfile);
L
Linus Torvalds 已提交
1185 1186 1187
	if (cmd == TUNSETIFF && !tun) {
		ifr.ifr_name[IFNAMSIZ-1] = '\0';

1188
		ret = tun_set_iff(tfile->net, file, &ifr);
L
Linus Torvalds 已提交
1189

1190 1191
		if (ret)
			goto unlock;
L
Linus Torvalds 已提交
1192

1193
		if (copy_to_user(argp, &ifr, ifreq_len))
1194 1195
			ret = -EFAULT;
		goto unlock;
L
Linus Torvalds 已提交
1196 1197
	}

1198
	ret = -EBADFD;
L
Linus Torvalds 已提交
1199
	if (!tun)
1200
		goto unlock;
L
Linus Torvalds 已提交
1201 1202 1203

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

E
Eric W. Biederman 已提交
1204
	ret = 0;
L
Linus Torvalds 已提交
1205
	switch (cmd) {
1206
	case TUNGETIFF:
1207
		ret = tun_get_iff(current->nsproxy->net_ns, tun, &ifr);
1208
		if (ret)
E
Eric W. Biederman 已提交
1209
			break;
1210

1211
		if (copy_to_user(argp, &ifr, ifreq_len))
E
Eric W. Biederman 已提交
1212
			ret = -EFAULT;
1213 1214
		break;

L
Linus Torvalds 已提交
1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233
	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
1234
		    tun->dev->name, arg ? "enabled" : "disabled");
L
Linus Torvalds 已提交
1235 1236 1237 1238 1239 1240 1241 1242 1243
		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;

1244 1245 1246 1247 1248 1249 1250
	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;

1251 1252 1253 1254 1255
	case TUNSETLINK:
		/* Only allow setting the type when the interface is down */
		if (tun->dev->flags & IFF_UP) {
			DBG(KERN_INFO "%s: Linktype set failed because interface is up\n",
				tun->dev->name);
1256
			ret = -EBUSY;
1257 1258 1259
		} else {
			tun->dev->type = (int) arg;
			DBG(KERN_INFO "%s: linktype set to %d\n", tun->dev->name, tun->dev->type);
1260
			ret = 0;
1261
		}
E
Eric W. Biederman 已提交
1262
		break;
1263

L
Linus Torvalds 已提交
1264 1265 1266 1267 1268
#ifdef TUN_DEBUG
	case TUNSETDEBUG:
		tun->debug = arg;
		break;
#endif
1269 1270
	case TUNSETOFFLOAD:
		ret = set_offload(tun->dev, arg);
E
Eric W. Biederman 已提交
1271
		break;
1272

1273 1274
	case TUNSETTXFILTER:
		/* Can be set only for TAPs */
E
Eric W. Biederman 已提交
1275
		ret = -EINVAL;
1276
		if ((tun->flags & TUN_TYPE_MASK) != TUN_TAP_DEV)
E
Eric W. Biederman 已提交
1277
			break;
H
Harvey Harrison 已提交
1278
		ret = update_filter(&tun->txflt, (void __user *)arg);
E
Eric W. Biederman 已提交
1279
		break;
L
Linus Torvalds 已提交
1280 1281

	case SIOCGIFHWADDR:
1282 1283 1284
		/* Get hw addres */
		memcpy(ifr.ifr_hwaddr.sa_data, tun->dev->dev_addr, ETH_ALEN);
		ifr.ifr_hwaddr.sa_family = tun->dev->type;
1285
		if (copy_to_user(argp, &ifr, ifreq_len))
E
Eric W. Biederman 已提交
1286 1287
			ret = -EFAULT;
		break;
L
Linus Torvalds 已提交
1288 1289

	case SIOCSIFHWADDR:
1290
		/* Set hw address */
J
Johannes Berg 已提交
1291 1292
		DBG(KERN_DEBUG "%s: set hw address: %pM\n",
			tun->dev->name, ifr.ifr_hwaddr.sa_data);
1293 1294

		ret = dev_set_mac_address(tun->dev, &ifr.ifr_hwaddr);
E
Eric W. Biederman 已提交
1295
		break;
1296 1297

	case TUNGETSNDBUF:
1298
		sndbuf = tun->socket.sk->sk_sndbuf;
1299 1300 1301 1302 1303 1304 1305 1306 1307 1308
		if (copy_to_user(argp, &sndbuf, sizeof(sndbuf)))
			ret = -EFAULT;
		break;

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

1309
		tun->socket.sk->sk_sndbuf = sndbuf;
1310 1311
		break;

L
Linus Torvalds 已提交
1312
	default:
E
Eric W. Biederman 已提交
1313 1314
		ret = -EINVAL;
		break;
L
Linus Torvalds 已提交
1315 1316
	};

1317 1318 1319 1320
unlock:
	rtnl_unlock();
	if (tun)
		tun_put(tun);
E
Eric W. Biederman 已提交
1321
	return ret;
L
Linus Torvalds 已提交
1322 1323
}

1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358
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
1359 1360
static int tun_chr_fasync(int fd, struct file *file, int on)
{
E
Eric W. Biederman 已提交
1361
	struct tun_struct *tun = tun_get(file);
L
Linus Torvalds 已提交
1362 1363 1364 1365 1366 1367 1368 1369
	int ret;

	if (!tun)
		return -EBADFD;

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

	if ((ret = fasync_helper(fd, file, on, &tun->fasync)) < 0)
J
Jonathan Corbet 已提交
1370
		goto out;
1371

L
Linus Torvalds 已提交
1372
	if (on) {
1373
		ret = __f_setown(file, task_pid(current), PIDTYPE_PID, 0);
L
Linus Torvalds 已提交
1374
		if (ret)
J
Jonathan Corbet 已提交
1375
			goto out;
L
Linus Torvalds 已提交
1376
		tun->flags |= TUN_FASYNC;
1377
	} else
L
Linus Torvalds 已提交
1378
		tun->flags &= ~TUN_FASYNC;
J
Jonathan Corbet 已提交
1379 1380
	ret = 0;
out:
E
Eric W. Biederman 已提交
1381
	tun_put(tun);
J
Jonathan Corbet 已提交
1382
	return ret;
L
Linus Torvalds 已提交
1383 1384 1385 1386
}

static int tun_chr_open(struct inode *inode, struct file * file)
{
E
Eric W. Biederman 已提交
1387
	struct tun_file *tfile;
T
Thomas Gleixner 已提交
1388

L
Linus Torvalds 已提交
1389
	DBG1(KERN_INFO "tunX: tun_chr_open\n");
E
Eric W. Biederman 已提交
1390 1391 1392 1393

	tfile = kmalloc(sizeof(*tfile), GFP_KERNEL);
	if (!tfile)
		return -ENOMEM;
1394
	atomic_set(&tfile->count, 0);
E
Eric W. Biederman 已提交
1395
	tfile->tun = NULL;
E
Eric W. Biederman 已提交
1396
	tfile->net = get_net(current->nsproxy->net_ns);
E
Eric W. Biederman 已提交
1397
	file->private_data = tfile;
L
Linus Torvalds 已提交
1398 1399 1400 1401 1402
	return 0;
}

static int tun_chr_close(struct inode *inode, struct file *file)
{
E
Eric W. Biederman 已提交
1403
	struct tun_file *tfile = file->private_data;
E
Eric W. Biederman 已提交
1404
	struct tun_struct *tun;
L
Linus Torvalds 已提交
1405

E
Eric W. Biederman 已提交
1406
	tun = __tun_get(tfile);
E
Eric W. Biederman 已提交
1407
	if (tun) {
H
Herbert Xu 已提交
1408 1409 1410
		struct net_device *dev = tun->dev;

		DBG(KERN_INFO "%s: tun_chr_close\n", dev->name);
L
Linus Torvalds 已提交
1411

E
Eric W. Biederman 已提交
1412
		__tun_detach(tun);
L
Linus Torvalds 已提交
1413

E
Eric W. Biederman 已提交
1414
		/* If desireable, unregister the netdevice. */
H
Herbert Xu 已提交
1415 1416 1417 1418 1419 1420
		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
1421
	}
L
Linus Torvalds 已提交
1422

1423 1424
	tun = tfile->tun;
	if (tun)
1425
		sock_put(tun->socket.sk);
1426

E
Eric W. Biederman 已提交
1427
	put_net(tfile->net);
E
Eric W. Biederman 已提交
1428
	kfree(tfile);
L
Linus Torvalds 已提交
1429 1430 1431 1432

	return 0;
}

1433
static const struct file_operations tun_fops = {
1434
	.owner	= THIS_MODULE,
L
Linus Torvalds 已提交
1435
	.llseek = no_llseek,
1436 1437 1438 1439
	.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
1440
	.poll	= tun_chr_poll,
1441 1442 1443 1444
	.unlocked_ioctl	= tun_chr_ioctl,
#ifdef CONFIG_COMPAT
	.compat_ioctl = tun_chr_compat_ioctl,
#endif
L
Linus Torvalds 已提交
1445 1446
	.open	= tun_chr_open,
	.release = tun_chr_close,
1447
	.fasync = tun_chr_fasync
L
Linus Torvalds 已提交
1448 1449 1450 1451 1452
};

static struct miscdevice tun_miscdev = {
	.minor = TUN_MINOR,
	.name = "tun",
1453
	.nodename = "net/tun",
L
Linus Torvalds 已提交
1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512
	.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
1513
	return !!tun->tfile;
L
Linus Torvalds 已提交
1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531
}

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;
}

1532
static const struct ethtool_ops tun_ethtool_ops = {
L
Linus Torvalds 已提交
1533 1534 1535 1536 1537 1538 1539 1540 1541
	.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
};

1542

L
Linus Torvalds 已提交
1543 1544 1545 1546 1547 1548 1549
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);

1550
	ret = rtnl_link_register(&tun_link_ops);
1551
	if (ret) {
1552 1553
		printk(KERN_ERR "tun: Can't register link_ops\n");
		goto err_linkops;
1554 1555
	}

L
Linus Torvalds 已提交
1556
	ret = misc_register(&tun_miscdev);
1557
	if (ret) {
L
Linus Torvalds 已提交
1558
		printk(KERN_ERR "tun: Can't register misc device %d\n", TUN_MINOR);
1559 1560
		goto err_misc;
	}
1561
	return  0;
1562
err_misc:
1563 1564
	rtnl_link_unregister(&tun_link_ops);
err_linkops:
L
Linus Torvalds 已提交
1565 1566 1567 1568 1569
	return ret;
}

static void tun_cleanup(void)
{
1570
	misc_deregister(&tun_miscdev);
1571
	rtnl_link_unregister(&tun_link_ops);
L
Linus Torvalds 已提交
1572 1573
}

1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590
/* 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
1591 1592 1593 1594 1595 1596
module_init(tun_init);
module_exit(tun_cleanup);
MODULE_DESCRIPTION(DRV_DESCRIPTION);
MODULE_AUTHOR(DRV_COPYRIGHT);
MODULE_LICENSE("GPL");
MODULE_ALIAS_MISCDEV(TUN_MINOR);