macvtap.c 26.4 KB
Newer Older
A
Arnd Bergmann 已提交
1 2 3 4 5 6 7 8 9 10 11
#include <linux/etherdevice.h>
#include <linux/if_macvlan.h>
#include <linux/interrupt.h>
#include <linux/nsproxy.h>
#include <linux/compat.h>
#include <linux/if_tun.h>
#include <linux/module.h>
#include <linux/skbuff.h>
#include <linux/cache.h>
#include <linux/sched.h>
#include <linux/types.h>
12
#include <linux/slab.h>
A
Arnd Bergmann 已提交
13 14 15
#include <linux/init.h>
#include <linux/wait.h>
#include <linux/cdev.h>
A
Al Viro 已提交
16
#include <linux/idr.h>
A
Arnd Bergmann 已提交
17 18 19 20 21
#include <linux/fs.h>

#include <net/net_namespace.h>
#include <net/rtnetlink.h>
#include <net/sock.h>
22
#include <linux/virtio_net.h>
A
Arnd Bergmann 已提交
23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

/*
 * A macvtap queue is the central object of this driver, it connects
 * an open character device to a macvlan interface. There can be
 * multiple queues on one interface, which map back to queues
 * implemented in hardware on the underlying device.
 *
 * macvtap_proto is used to allocate queues through the sock allocation
 * mechanism.
 *
 * TODO: multiqueue support is currently not implemented, even though
 * macvtap is basically prepared for that. We will need to add this
 * here as well as in virtio-net and qemu to get line rate on 10gbit
 * adapters from a guest.
 */
struct macvtap_queue {
	struct sock sk;
	struct socket sock;
41
	struct socket_wq wq;
42
	int vnet_hdr_sz;
43
	struct macvlan_dev __rcu *vlan;
A
Arnd Bergmann 已提交
44
	struct file *file;
45
	unsigned int flags;
A
Arnd Bergmann 已提交
46 47 48 49 50 51 52 53 54
};

static struct proto macvtap_proto = {
	.name = "macvtap",
	.owner = THIS_MODULE,
	.obj_size = sizeof (struct macvtap_queue),
};

/*
55
 * Variables for dealing with macvtaps device numbers.
A
Arnd Bergmann 已提交
56
 */
57
static dev_t macvtap_major;
58 59 60 61
#define MACVTAP_NUM_DEVS (1U << MINORBITS)
static DEFINE_MUTEX(minor_lock);
static DEFINE_IDR(minor_idr);

62
#define GOODCOPY_LEN 128
A
Arnd Bergmann 已提交
63 64 65
static struct class *macvtap_class;
static struct cdev macvtap_cdev;

A
Arnd Bergmann 已提交
66 67
static const struct proto_ops macvtap_socket_ops;

A
Arnd Bergmann 已提交
68 69
/*
 * RCU usage:
70 71 72
 * The macvtap_queue and the macvlan_dev are loosely coupled, the
 * pointers from one to the other can only be read while rcu_read_lock
 * or macvtap_lock is held.
A
Arnd Bergmann 已提交
73
 *
74 75 76 77
 * Both the file and the macvlan_dev hold a reference on the macvtap_queue
 * through sock_hold(&q->sk). When the macvlan_dev goes away first,
 * q->vlan becomes inaccessible. When the files gets closed,
 * macvtap_get_queue() fails.
A
Arnd Bergmann 已提交
78
 *
79 80 81 82
 * There may still be references to the struct sock inside of the
 * queue from outbound SKBs, but these never reference back to the
 * file or the dev. The data structure is freed through __sk_free
 * when both our references and any pending SKBs are gone.
A
Arnd Bergmann 已提交
83 84 85 86
 */
static DEFINE_SPINLOCK(macvtap_lock);

/*
87 88 89
 * get_slot: return a [unused/occupied] slot in vlan->taps[]:
 *	- if 'q' is NULL, return the first empty slot;
 *	- otherwise, return the slot this pointer occupies.
A
Arnd Bergmann 已提交
90
 */
91 92 93 94 95 96 97 98 99 100 101 102 103
static int get_slot(struct macvlan_dev *vlan, struct macvtap_queue *q)
{
	int i;

	for (i = 0; i < MAX_MACVTAP_QUEUES; i++) {
		if (rcu_dereference(vlan->taps[i]) == q)
			return i;
	}

	/* Should never happen */
	BUG_ON(1);
}

A
Arnd Bergmann 已提交
104 105 106 107
static int macvtap_set_queue(struct net_device *dev, struct file *file,
				struct macvtap_queue *q)
{
	struct macvlan_dev *vlan = netdev_priv(dev);
108
	int index;
A
Arnd Bergmann 已提交
109 110 111
	int err = -EBUSY;

	spin_lock(&macvtap_lock);
112
	if (vlan->numvtaps == MAX_MACVTAP_QUEUES)
A
Arnd Bergmann 已提交
113 114 115
		goto out;

	err = 0;
116
	index = get_slot(vlan, NULL);
117
	rcu_assign_pointer(q->vlan, vlan);
118
	rcu_assign_pointer(vlan->taps[index], q);
119
	sock_hold(&q->sk);
A
Arnd Bergmann 已提交
120 121

	q->file = file;
122
	file->private_data = q;
A
Arnd Bergmann 已提交
123

124 125
	vlan->numvtaps++;

A
Arnd Bergmann 已提交
126 127 128 129 130 131
out:
	spin_unlock(&macvtap_lock);
	return err;
}

/*
132 133 134
 * The file owning the queue got closed, give up both
 * the reference that the files holds as well as the
 * one from the macvlan_dev if that still exists.
A
Arnd Bergmann 已提交
135 136 137 138
 *
 * Using the spinlock makes sure that we don't get
 * to the queue again after destroying it.
 */
139
static void macvtap_put_queue(struct macvtap_queue *q)
A
Arnd Bergmann 已提交
140
{
141
	struct macvlan_dev *vlan;
A
Arnd Bergmann 已提交
142 143

	spin_lock(&macvtap_lock);
144 145
	vlan = rcu_dereference_protected(q->vlan,
					 lockdep_is_held(&macvtap_lock));
146
	if (vlan) {
147 148
		int index = get_slot(vlan, q);

149 150
		RCU_INIT_POINTER(vlan->taps[index], NULL);
		RCU_INIT_POINTER(q->vlan, NULL);
151
		sock_put(&q->sk);
152
		--vlan->numvtaps;
A
Arnd Bergmann 已提交
153 154 155 156 157 158 159 160 161
	}

	spin_unlock(&macvtap_lock);

	synchronize_rcu();
	sock_put(&q->sk);
}

/*
162 163 164 165 166
 * Select a queue based on the rxq of the device on which this packet
 * arrived. If the incoming device is not mq, calculate a flow hash
 * to select a queue. If all fails, find the first available queue.
 * Cache vlan->numvtaps since it can become zero during the execution
 * of this function.
A
Arnd Bergmann 已提交
167 168 169 170 171
 */
static struct macvtap_queue *macvtap_get_queue(struct net_device *dev,
					       struct sk_buff *skb)
{
	struct macvlan_dev *vlan = netdev_priv(dev);
172 173 174 175 176 177 178
	struct macvtap_queue *tap = NULL;
	int numvtaps = vlan->numvtaps;
	__u32 rxq;

	if (!numvtaps)
		goto out;

179 180 181 182 183 184 185 186
	/* Check if we can use flow to select a queue */
	rxq = skb_get_rxhash(skb);
	if (rxq) {
		tap = rcu_dereference(vlan->taps[rxq % numvtaps]);
		if (tap)
			goto out;
	}

187 188
	if (likely(skb_rx_queue_recorded(skb))) {
		rxq = skb_get_rx_queue(skb);
A
Arnd Bergmann 已提交
189

190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206
		while (unlikely(rxq >= numvtaps))
			rxq -= numvtaps;

		tap = rcu_dereference(vlan->taps[rxq]);
		if (tap)
			goto out;
	}

	/* Everything failed - find first available queue */
	for (rxq = 0; rxq < MAX_MACVTAP_QUEUES; rxq++) {
		tap = rcu_dereference(vlan->taps[rxq]);
		if (tap)
			break;
	}

out:
	return tap;
A
Arnd Bergmann 已提交
207 208
}

209 210
/*
 * The net_device is going away, give up the reference
211 212
 * that it holds on all queues and safely set the pointer
 * from the queues to NULL.
213
 */
A
Arnd Bergmann 已提交
214 215 216
static void macvtap_del_queues(struct net_device *dev)
{
	struct macvlan_dev *vlan = netdev_priv(dev);
217 218
	struct macvtap_queue *q, *qlist[MAX_MACVTAP_QUEUES];
	int i, j = 0;
219

220
	/* macvtap_put_queue can free some slots, so go through all slots */
221
	spin_lock(&macvtap_lock);
222
	for (i = 0; i < MAX_MACVTAP_QUEUES && vlan->numvtaps; i++) {
223 224
		q = rcu_dereference_protected(vlan->taps[i],
					      lockdep_is_held(&macvtap_lock));
225 226
		if (q) {
			qlist[j++] = q;
227 228
			RCU_INIT_POINTER(vlan->taps[i], NULL);
			RCU_INIT_POINTER(q->vlan, NULL);
229 230
			vlan->numvtaps--;
		}
231
	}
232
	BUG_ON(vlan->numvtaps != 0);
233 234
	/* guarantee that any future macvtap_set_queue will fail */
	vlan->numvtaps = MAX_MACVTAP_QUEUES;
235 236 237
	spin_unlock(&macvtap_lock);

	synchronize_rcu();
238 239 240

	for (--j; j >= 0; j--)
		sock_put(&qlist[j]->sk);
A
Arnd Bergmann 已提交
241 242 243 244 245 246 247 248 249 250 251
}

/*
 * Forward happens for data that gets sent from one macvlan
 * endpoint to another one in bridge mode. We just take
 * the skb and put it into the receive queue.
 */
static int macvtap_forward(struct net_device *dev, struct sk_buff *skb)
{
	struct macvtap_queue *q = macvtap_get_queue(dev, skb);
	if (!q)
H
Herbert Xu 已提交
252 253 254 255
		goto drop;

	if (skb_queue_len(&q->sk.sk_receive_queue) >= dev->tx_queue_len)
		goto drop;
A
Arnd Bergmann 已提交
256 257

	skb_queue_tail(&q->sk.sk_receive_queue, skb);
E
Eric Dumazet 已提交
258
	wake_up_interruptible_poll(sk_sleep(&q->sk), POLLIN | POLLRDNORM | POLLRDBAND);
H
Herbert Xu 已提交
259 260 261 262 263
	return NET_RX_SUCCESS;

drop:
	kfree_skb(skb);
	return NET_RX_DROP;
A
Arnd Bergmann 已提交
264 265 266 267 268 269 270 271 272 273 274 275 276
}

/*
 * Receive is for data from the external interface (lowerdev),
 * in case of macvtap, we can treat that the same way as
 * forward, which macvlan cannot.
 */
static int macvtap_receive(struct sk_buff *skb)
{
	skb_push(skb, ETH_HLEN);
	return macvtap_forward(skb->dev, skb);
}

277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328
static int macvtap_get_minor(struct macvlan_dev *vlan)
{
	int retval = -ENOMEM;
	int id;

	mutex_lock(&minor_lock);
	if (idr_pre_get(&minor_idr, GFP_KERNEL) == 0)
		goto exit;

	retval = idr_get_new_above(&minor_idr, vlan, 1, &id);
	if (retval < 0) {
		if (retval == -EAGAIN)
			retval = -ENOMEM;
		goto exit;
	}
	if (id < MACVTAP_NUM_DEVS) {
		vlan->minor = id;
	} else {
		printk(KERN_ERR "too many macvtap devices\n");
		retval = -EINVAL;
		idr_remove(&minor_idr, id);
	}
exit:
	mutex_unlock(&minor_lock);
	return retval;
}

static void macvtap_free_minor(struct macvlan_dev *vlan)
{
	mutex_lock(&minor_lock);
	if (vlan->minor) {
		idr_remove(&minor_idr, vlan->minor);
		vlan->minor = 0;
	}
	mutex_unlock(&minor_lock);
}

static struct net_device *dev_get_by_macvtap_minor(int minor)
{
	struct net_device *dev = NULL;
	struct macvlan_dev *vlan;

	mutex_lock(&minor_lock);
	vlan = idr_find(&minor_idr, minor);
	if (vlan) {
		dev = vlan->dev;
		dev_hold(dev);
	}
	mutex_unlock(&minor_lock);
	return dev;
}

A
Arnd Bergmann 已提交
329 330 331 332 333
static int macvtap_newlink(struct net *src_net,
			   struct net_device *dev,
			   struct nlattr *tb[],
			   struct nlattr *data[])
{
334 335 336 337 338
	/* Don't put anything that may fail after macvlan_common_newlink
	 * because we can't undo what it does.
	 */
	return macvlan_common_newlink(src_net, dev, tb, data,
				      macvtap_receive, macvtap_forward);
A
Arnd Bergmann 已提交
339 340 341 342 343 344 345 346 347
}

static void macvtap_dellink(struct net_device *dev,
			    struct list_head *head)
{
	macvtap_del_queues(dev);
	macvlan_dellink(dev, head);
}

H
Herbert Xu 已提交
348 349 350 351 352 353
static void macvtap_setup(struct net_device *dev)
{
	macvlan_common_setup(dev);
	dev->tx_queue_len = TUN_READQ_SIZE;
}

A
Arnd Bergmann 已提交
354 355
static struct rtnl_link_ops macvtap_link_ops __read_mostly = {
	.kind		= "macvtap",
H
Herbert Xu 已提交
356
	.setup		= macvtap_setup,
A
Arnd Bergmann 已提交
357 358 359 360 361 362 363
	.newlink	= macvtap_newlink,
	.dellink	= macvtap_dellink,
};


static void macvtap_sock_write_space(struct sock *sk)
{
364 365
	wait_queue_head_t *wqueue;

A
Arnd Bergmann 已提交
366 367 368 369
	if (!sock_writeable(sk) ||
	    !test_and_clear_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags))
		return;

370 371 372
	wqueue = sk_sleep(sk);
	if (wqueue && waitqueue_active(wqueue))
		wake_up_interruptible_poll(wqueue, POLLOUT | POLLWRNORM | POLLWRBAND);
A
Arnd Bergmann 已提交
373 374
}

375 376 377 378 379
static void macvtap_sock_destruct(struct sock *sk)
{
	skb_queue_purge(&sk->sk_receive_queue);
}

A
Arnd Bergmann 已提交
380 381 382
static int macvtap_open(struct inode *inode, struct file *file)
{
	struct net *net = current->nsproxy->net_ns;
383
	struct net_device *dev = dev_get_by_macvtap_minor(iminor(inode));
A
Arnd Bergmann 已提交
384 385 386 387 388 389 390 391 392 393 394 395 396
	struct macvtap_queue *q;
	int err;

	err = -ENODEV;
	if (!dev)
		goto out;

	err = -ENOMEM;
	q = (struct macvtap_queue *)sk_alloc(net, AF_UNSPEC, GFP_KERNEL,
					     &macvtap_proto);
	if (!q)
		goto out;

397 398
	q->sock.wq = &q->wq;
	init_waitqueue_head(&q->wq.wait);
A
Arnd Bergmann 已提交
399 400
	q->sock.type = SOCK_RAW;
	q->sock.state = SS_CONNECTED;
A
Arnd Bergmann 已提交
401 402
	q->sock.file = file;
	q->sock.ops = &macvtap_socket_ops;
A
Arnd Bergmann 已提交
403 404
	sock_init_data(&q->sock, &q->sk);
	q->sk.sk_write_space = macvtap_sock_write_space;
405
	q->sk.sk_destruct = macvtap_sock_destruct;
406
	q->flags = IFF_VNET_HDR | IFF_NO_PI | IFF_TAP;
407
	q->vnet_hdr_sz = sizeof(struct virtio_net_hdr);
A
Arnd Bergmann 已提交
408

409 410 411
	/*
	 * so far only KVM virtio_net uses macvtap, enable zero copy between
	 * guest kernel and host kernel when lower device supports zerocopy
412 413 414
	 *
	 * The macvlan supports zerocopy iff the lower device supports zero
	 * copy so we don't have to look at the lower device directly.
415
	 */
416 417
	if ((dev->features & NETIF_F_HIGHDMA) && (dev->features & NETIF_F_SG))
		sock_set_flag(&q->sk, SOCK_ZEROCOPY);
418

A
Arnd Bergmann 已提交
419 420 421 422 423 424 425 426 427 428 429 430 431
	err = macvtap_set_queue(dev, file, q);
	if (err)
		sock_put(&q->sk);

out:
	if (dev)
		dev_put(dev);

	return err;
}

static int macvtap_release(struct inode *inode, struct file *file)
{
432 433
	struct macvtap_queue *q = file->private_data;
	macvtap_put_queue(q);
A
Arnd Bergmann 已提交
434 435 436 437 438
	return 0;
}

static unsigned int macvtap_poll(struct file *file, poll_table * wait)
{
439
	struct macvtap_queue *q = file->private_data;
A
Arnd Bergmann 已提交
440 441 442 443 444 445
	unsigned int mask = POLLERR;

	if (!q)
		goto out;

	mask = 0;
446
	poll_wait(file, &q->wq.wait, wait);
A
Arnd Bergmann 已提交
447 448 449 450 451 452 453 454 455 456 457 458 459

	if (!skb_queue_empty(&q->sk.sk_receive_queue))
		mask |= POLLIN | POLLRDNORM;

	if (sock_writeable(&q->sk) ||
	    (!test_and_set_bit(SOCK_ASYNC_NOSPACE, &q->sock.flags) &&
	     sock_writeable(&q->sk)))
		mask |= POLLOUT | POLLWRNORM;

out:
	return mask;
}

460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482
static inline struct sk_buff *macvtap_alloc_skb(struct sock *sk, size_t prepad,
						size_t len, size_t linear,
						int noblock, int *err)
{
	struct sk_buff *skb;

	/* Under a page?  Don't bother with paged skb. */
	if (prepad + len < PAGE_SIZE || !linear)
		linear = len;

	skb = sock_alloc_send_pskb(sk, prepad + linear, len - linear, noblock,
				   err);
	if (!skb)
		return NULL;

	skb_reserve(skb, prepad);
	skb_put(skb, linear);
	skb->data_len = len - linear;
	skb->len += len - linear;

	return skb;
}

483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507
/* set skb frags from iovec, this can move to core network code for reuse */
static int zerocopy_sg_from_iovec(struct sk_buff *skb, const struct iovec *from,
				  int offset, size_t count)
{
	int len = iov_length(from, count) - offset;
	int copy = skb_headlen(skb);
	int size, offset1 = 0;
	int i = 0;

	/* Skip over from offset */
	while (count && (offset >= from->iov_len)) {
		offset -= from->iov_len;
		++from;
		--count;
	}

	/* copy up to skb headlen */
	while (count && (copy > 0)) {
		size = min_t(unsigned int, copy, from->iov_len - offset);
		if (copy_from_user(skb->data + offset1, from->iov_base + offset,
				   size))
			return -EFAULT;
		if (copy > size) {
			++from;
			--count;
508 509 510
			offset = 0;
		} else
			offset += size;
511 512 513 514 515 516 517 518 519 520 521
		copy -= size;
		offset1 += size;
	}

	if (len == offset1)
		return 0;

	while (count--) {
		struct page *page[MAX_SKB_FRAGS];
		int num_pages;
		unsigned long base;
522
		unsigned long truesize;
523

524
		len = from->iov_len - offset;
525
		if (!len) {
526
			offset = 0;
527 528 529
			++from;
			continue;
		}
530
		base = (unsigned long)from->iov_base + offset;
531 532 533
		size = ((base & ~PAGE_MASK) + len + ~PAGE_MASK) >> PAGE_SHIFT;
		num_pages = get_user_pages_fast(base, size, 0, &page[i]);
		if ((num_pages != size) ||
534 535 536
		    (num_pages > MAX_SKB_FRAGS - skb_shinfo(skb)->nr_frags)) {
			for (i = 0; i < num_pages; i++)
				put_page(page[i]);
537
			return -EFAULT;
538
		}
539
		truesize = size * PAGE_SIZE;
540 541
		skb->data_len += len;
		skb->len += len;
542 543
		skb->truesize += truesize;
		atomic_add(truesize, &skb->sk->sk_wmem_alloc);
544
		while (len) {
545 546 547
			int off = base & ~PAGE_MASK;
			int size = min_t(int, len, PAGE_SIZE - off);
			__skb_fill_page_desc(skb, i, page[i], off, size);
548 549
			skb_shinfo(skb)->nr_frags++;
			/* increase sk_wmem_alloc */
550 551
			base += size;
			len -= size;
552 553
			i++;
		}
554
		offset = 0;
555 556 557 558 559
		++from;
	}
	return 0;
}

560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632
/*
 * macvtap_skb_from_vnet_hdr and macvtap_skb_to_vnet_hdr should
 * be shared with the tun/tap driver.
 */
static int macvtap_skb_from_vnet_hdr(struct sk_buff *skb,
				     struct virtio_net_hdr *vnet_hdr)
{
	unsigned short gso_type = 0;
	if (vnet_hdr->gso_type != VIRTIO_NET_HDR_GSO_NONE) {
		switch (vnet_hdr->gso_type & ~VIRTIO_NET_HDR_GSO_ECN) {
		case VIRTIO_NET_HDR_GSO_TCPV4:
			gso_type = SKB_GSO_TCPV4;
			break;
		case VIRTIO_NET_HDR_GSO_TCPV6:
			gso_type = SKB_GSO_TCPV6;
			break;
		case VIRTIO_NET_HDR_GSO_UDP:
			gso_type = SKB_GSO_UDP;
			break;
		default:
			return -EINVAL;
		}

		if (vnet_hdr->gso_type & VIRTIO_NET_HDR_GSO_ECN)
			gso_type |= SKB_GSO_TCP_ECN;

		if (vnet_hdr->gso_size == 0)
			return -EINVAL;
	}

	if (vnet_hdr->flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) {
		if (!skb_partial_csum_set(skb, vnet_hdr->csum_start,
					  vnet_hdr->csum_offset))
			return -EINVAL;
	}

	if (vnet_hdr->gso_type != VIRTIO_NET_HDR_GSO_NONE) {
		skb_shinfo(skb)->gso_size = vnet_hdr->gso_size;
		skb_shinfo(skb)->gso_type = gso_type;

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

static int macvtap_skb_to_vnet_hdr(const struct sk_buff *skb,
				   struct virtio_net_hdr *vnet_hdr)
{
	memset(vnet_hdr, 0, sizeof(*vnet_hdr));

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

		/* This is a hint as to how much should be linear. */
		vnet_hdr->hdr_len = skb_headlen(skb);
		vnet_hdr->gso_size = sinfo->gso_size;
		if (sinfo->gso_type & SKB_GSO_TCPV4)
			vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_TCPV4;
		else if (sinfo->gso_type & SKB_GSO_TCPV6)
			vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_TCPV6;
		else if (sinfo->gso_type & SKB_GSO_UDP)
			vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_UDP;
		else
			BUG();
		if (sinfo->gso_type & SKB_GSO_TCP_ECN)
			vnet_hdr->gso_type |= VIRTIO_NET_HDR_GSO_ECN;
	} else
		vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_NONE;

	if (skb->ip_summed == CHECKSUM_PARTIAL) {
		vnet_hdr->flags = VIRTIO_NET_HDR_F_NEEDS_CSUM;
633
		vnet_hdr->csum_start = skb_checksum_start_offset(skb);
634
		vnet_hdr->csum_offset = skb->csum_offset;
635 636
	} else if (skb->ip_summed == CHECKSUM_UNNECESSARY) {
		vnet_hdr->flags = VIRTIO_NET_HDR_F_DATA_VALID;
637 638 639 640 641 642
	} /* else everything is zero */

	return 0;
}


A
Arnd Bergmann 已提交
643
/* Get packet from user space buffer */
644 645 646
static ssize_t macvtap_get_user(struct macvtap_queue *q, struct msghdr *m,
				const struct iovec *iv, unsigned long total_len,
				size_t count, int noblock)
A
Arnd Bergmann 已提交
647 648
{
	struct sk_buff *skb;
649
	struct macvlan_dev *vlan;
650
	unsigned long len = total_len;
A
Arnd Bergmann 已提交
651
	int err;
652 653
	struct virtio_net_hdr vnet_hdr = { 0 };
	int vnet_hdr_len = 0;
654 655
	int copylen;
	bool zerocopy = false;
656 657

	if (q->flags & IFF_VNET_HDR) {
658
		vnet_hdr_len = q->vnet_hdr_sz;
659 660

		err = -EINVAL;
661
		if (len < vnet_hdr_len)
662
			goto err;
663
		len -= vnet_hdr_len;
664 665

		err = memcpy_fromiovecend((void *)&vnet_hdr, iv, 0,
666
					   sizeof(vnet_hdr));
667 668 669 670 671 672 673 674 675 676 677
		if (err < 0)
			goto err;
		if ((vnet_hdr.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) &&
		     vnet_hdr.csum_start + vnet_hdr.csum_offset + 2 >
							vnet_hdr.hdr_len)
			vnet_hdr.hdr_len = vnet_hdr.csum_start +
						vnet_hdr.csum_offset + 2;
		err = -EINVAL;
		if (vnet_hdr.hdr_len > len)
			goto err;
	}
A
Arnd Bergmann 已提交
678

679
	err = -EINVAL;
A
Arnd Bergmann 已提交
680
	if (unlikely(len < ETH_HLEN))
681
		goto err;
A
Arnd Bergmann 已提交
682

683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698
	if (m && m->msg_control && sock_flag(&q->sk, SOCK_ZEROCOPY))
		zerocopy = true;

	if (zerocopy) {
		/* There are 256 bytes to be copied in skb, so there is enough
		 * room for skb expand head in case it is used.
		 * The rest buffer is mapped from userspace.
		 */
		copylen = vnet_hdr.hdr_len;
		if (!copylen)
			copylen = GOODCOPY_LEN;
	} else
		copylen = len;

	skb = macvtap_alloc_skb(&q->sk, NET_IP_ALIGN, copylen,
				vnet_hdr.hdr_len, noblock, &err);
699 700
	if (!skb)
		goto err;
A
Arnd Bergmann 已提交
701

702 703 704 705 706 707
	if (zerocopy) {
		err = zerocopy_sg_from_iovec(skb, iv, vnet_hdr_len, count);
		skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
	} else
		err = skb_copy_datagram_from_iovec(skb, 0, iv, vnet_hdr_len,
						   len);
708
	if (err)
709
		goto err_kfree;
A
Arnd Bergmann 已提交
710 711

	skb_set_network_header(skb, ETH_HLEN);
712 713 714 715 716 717 718 719 720
	skb_reset_mac_header(skb);
	skb->protocol = eth_hdr(skb)->h_proto;

	if (vnet_hdr_len) {
		err = macvtap_skb_from_vnet_hdr(skb, &vnet_hdr);
		if (err)
			goto err_kfree;
	}

721
	rcu_read_lock_bh();
722
	vlan = rcu_dereference_bh(q->vlan);
723 724 725
	/* copy skb_ubuf_info for callback when skb has no error */
	if (zerocopy)
		skb_shinfo(skb)->destructor_arg = m->msg_control;
726 727 728 729 730
	if (vlan)
		macvlan_start_xmit(skb, vlan->dev);
	else
		kfree_skb(skb);
	rcu_read_unlock_bh();
A
Arnd Bergmann 已提交
731

732
	return total_len;
733

734 735 736
err_kfree:
	kfree_skb(skb);

737 738
err:
	rcu_read_lock_bh();
739
	vlan = rcu_dereference_bh(q->vlan);
740
	if (vlan)
E
Eric Dumazet 已提交
741
		vlan->dev->stats.tx_dropped++;
742 743 744
	rcu_read_unlock_bh();

	return err;
A
Arnd Bergmann 已提交
745 746 747 748 749 750 751
}

static ssize_t macvtap_aio_write(struct kiocb *iocb, const struct iovec *iv,
				 unsigned long count, loff_t pos)
{
	struct file *file = iocb->ki_filp;
	ssize_t result = -ENOLINK;
752
	struct macvtap_queue *q = file->private_data;
A
Arnd Bergmann 已提交
753

754 755
	result = macvtap_get_user(q, NULL, iv, iov_length(iv, count), count,
				  file->f_flags & O_NONBLOCK);
A
Arnd Bergmann 已提交
756 757 758 759 760 761 762 763
	return result;
}

/* Put packet to the user space buffer */
static ssize_t macvtap_put_user(struct macvtap_queue *q,
				const struct sk_buff *skb,
				const struct iovec *iv, int len)
{
764
	struct macvlan_dev *vlan;
A
Arnd Bergmann 已提交
765
	int ret;
766 767 768 769
	int vnet_hdr_len = 0;

	if (q->flags & IFF_VNET_HDR) {
		struct virtio_net_hdr vnet_hdr;
770
		vnet_hdr_len = q->vnet_hdr_sz;
771 772 773 774 775 776 777
		if ((len -= vnet_hdr_len) < 0)
			return -EINVAL;

		ret = macvtap_skb_to_vnet_hdr(skb, &vnet_hdr);
		if (ret)
			return ret;

778
		if (memcpy_toiovecend(iv, (void *)&vnet_hdr, 0, sizeof(vnet_hdr)))
779 780
			return -EFAULT;
	}
A
Arnd Bergmann 已提交
781 782 783

	len = min_t(int, skb->len, len);

784
	ret = skb_copy_datagram_const_iovec(skb, 0, iv, vnet_hdr_len, len);
A
Arnd Bergmann 已提交
785

786
	rcu_read_lock_bh();
787
	vlan = rcu_dereference_bh(q->vlan);
A
Arnd Bergmann 已提交
788 789
	if (vlan)
		macvlan_count_rx(vlan, len, ret == 0, 0);
790
	rcu_read_unlock_bh();
A
Arnd Bergmann 已提交
791

792
	return ret ? ret : (len + vnet_hdr_len);
A
Arnd Bergmann 已提交
793 794
}

A
Arnd Bergmann 已提交
795 796 797
static ssize_t macvtap_do_read(struct macvtap_queue *q, struct kiocb *iocb,
			       const struct iovec *iv, unsigned long len,
			       int noblock)
A
Arnd Bergmann 已提交
798 799 800
{
	DECLARE_WAITQUEUE(wait, current);
	struct sk_buff *skb;
A
Arnd Bergmann 已提交
801
	ssize_t ret = 0;
A
Arnd Bergmann 已提交
802

E
Eric Dumazet 已提交
803
	add_wait_queue(sk_sleep(&q->sk), &wait);
A
Arnd Bergmann 已提交
804 805 806 807 808 809
	while (len) {
		current->state = TASK_INTERRUPTIBLE;

		/* Read frames from the queue */
		skb = skb_dequeue(&q->sk.sk_receive_queue);
		if (!skb) {
A
Arnd Bergmann 已提交
810
			if (noblock) {
A
Arnd Bergmann 已提交
811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827
				ret = -EAGAIN;
				break;
			}
			if (signal_pending(current)) {
				ret = -ERESTARTSYS;
				break;
			}
			/* Nothing to read, let's sleep */
			schedule();
			continue;
		}
		ret = macvtap_put_user(q, skb, iv, len);
		kfree_skb(skb);
		break;
	}

	current->state = TASK_RUNNING;
E
Eric Dumazet 已提交
828
	remove_wait_queue(sk_sleep(&q->sk), &wait);
A
Arnd Bergmann 已提交
829 830 831 832 833 834 835 836 837
	return ret;
}

static ssize_t macvtap_aio_read(struct kiocb *iocb, const struct iovec *iv,
				unsigned long count, loff_t pos)
{
	struct file *file = iocb->ki_filp;
	struct macvtap_queue *q = file->private_data;
	ssize_t len, ret = 0;
A
Arnd Bergmann 已提交
838

A
Arnd Bergmann 已提交
839 840 841 842 843 844 845 846
	len = iov_length(iv, count);
	if (len < 0) {
		ret = -EINVAL;
		goto out;
	}

	ret = macvtap_do_read(q, iocb, iv, len, file->f_flags & O_NONBLOCK);
	ret = min_t(ssize_t, ret, len); /* XXX copied from tun.c. Why? */
A
Arnd Bergmann 已提交
847 848 849 850 851 852 853 854 855 856
out:
	return ret;
}

/*
 * provide compatibility with generic tun/tap interface
 */
static long macvtap_ioctl(struct file *file, unsigned int cmd,
			  unsigned long arg)
{
857 858
	struct macvtap_queue *q = file->private_data;
	struct macvlan_dev *vlan;
A
Arnd Bergmann 已提交
859 860 861 862
	void __user *argp = (void __user *)arg;
	struct ifreq __user *ifr = argp;
	unsigned int __user *up = argp;
	unsigned int u;
863 864
	int __user *sp = argp;
	int s;
865
	int ret;
A
Arnd Bergmann 已提交
866 867 868 869 870 871

	switch (cmd) {
	case TUNSETIFF:
		/* ignore the name, just look at flags */
		if (get_user(u, &ifr->ifr_flags))
			return -EFAULT;
872 873 874 875 876 877 878 879

		ret = 0;
		if ((u & ~IFF_VNET_HDR) != (IFF_NO_PI | IFF_TAP))
			ret = -EINVAL;
		else
			q->flags = u;

		return ret;
A
Arnd Bergmann 已提交
880 881

	case TUNGETIFF:
882
		rcu_read_lock_bh();
883
		vlan = rcu_dereference_bh(q->vlan);
884 885 886 887 888
		if (vlan)
			dev_hold(vlan->dev);
		rcu_read_unlock_bh();

		if (!vlan)
A
Arnd Bergmann 已提交
889 890
			return -ENOLINK;

891
		ret = 0;
892
		if (copy_to_user(&ifr->ifr_name, vlan->dev->name, IFNAMSIZ) ||
893
		    put_user(q->flags, &ifr->ifr_flags))
894 895 896
			ret = -EFAULT;
		dev_put(vlan->dev);
		return ret;
A
Arnd Bergmann 已提交
897 898

	case TUNGETFEATURES:
899
		if (put_user(IFF_TAP | IFF_NO_PI | IFF_VNET_HDR, up))
A
Arnd Bergmann 已提交
900 901 902 903 904 905 906 907 908 909
			return -EFAULT;
		return 0;

	case TUNSETSNDBUF:
		if (get_user(u, up))
			return -EFAULT;

		q->sk.sk_sndbuf = u;
		return 0;

910 911 912 913 914 915 916 917 918 919 920 921 922 923 924
	case TUNGETVNETHDRSZ:
		s = q->vnet_hdr_sz;
		if (put_user(s, sp))
			return -EFAULT;
		return 0;

	case TUNSETVNETHDRSZ:
		if (get_user(s, sp))
			return -EFAULT;
		if (s < (int)sizeof(struct virtio_net_hdr))
			return -EINVAL;

		q->vnet_hdr_sz = s;
		return 0;

A
Arnd Bergmann 已提交
925 926 927
	case TUNSETOFFLOAD:
		/* let the user check for future flags */
		if (arg & ~(TUN_F_CSUM | TUN_F_TSO4 | TUN_F_TSO6 |
928
			    TUN_F_TSO_ECN | TUN_F_UFO))
A
Arnd Bergmann 已提交
929 930
			return -EINVAL;

931 932 933 934
		/* TODO: only accept frames with the features that
			 got enabled for forwarded frames */
		if (!(q->flags & IFF_VNET_HDR))
			return  -EINVAL;
A
Arnd Bergmann 已提交
935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963
		return 0;

	default:
		return -EINVAL;
	}
}

#ifdef CONFIG_COMPAT
static long macvtap_compat_ioctl(struct file *file, unsigned int cmd,
				 unsigned long arg)
{
	return macvtap_ioctl(file, cmd, (unsigned long)compat_ptr(arg));
}
#endif

static const struct file_operations macvtap_fops = {
	.owner		= THIS_MODULE,
	.open		= macvtap_open,
	.release	= macvtap_release,
	.aio_read	= macvtap_aio_read,
	.aio_write	= macvtap_aio_write,
	.poll		= macvtap_poll,
	.llseek		= no_llseek,
	.unlocked_ioctl	= macvtap_ioctl,
#ifdef CONFIG_COMPAT
	.compat_ioctl	= macvtap_compat_ioctl,
#endif
};

A
Arnd Bergmann 已提交
964 965 966 967
static int macvtap_sendmsg(struct kiocb *iocb, struct socket *sock,
			   struct msghdr *m, size_t total_len)
{
	struct macvtap_queue *q = container_of(sock, struct macvtap_queue, sock);
968
	return macvtap_get_user(q, m, m->msg_iov, total_len, m->msg_iovlen,
A
Arnd Bergmann 已提交
969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010
			    m->msg_flags & MSG_DONTWAIT);
}

static int macvtap_recvmsg(struct kiocb *iocb, struct socket *sock,
			   struct msghdr *m, size_t total_len,
			   int flags)
{
	struct macvtap_queue *q = container_of(sock, struct macvtap_queue, sock);
	int ret;
	if (flags & ~(MSG_DONTWAIT|MSG_TRUNC))
		return -EINVAL;
	ret = macvtap_do_read(q, 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 macvtap_socket_ops = {
	.sendmsg = macvtap_sendmsg,
	.recvmsg = macvtap_recvmsg,
};

/* 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 *macvtap_get_socket(struct file *file)
{
	struct macvtap_queue *q;
	if (file->f_op != &macvtap_fops)
		return ERR_PTR(-EINVAL);
	q = file->private_data;
	if (!q)
		return ERR_PTR(-EBADFD);
	return &q->sock;
}
EXPORT_SYMBOL_GPL(macvtap_get_socket);

1011 1012 1013 1014
static int macvtap_device_event(struct notifier_block *unused,
				unsigned long event, void *ptr)
{
	struct net_device *dev = ptr;
1015
	struct macvlan_dev *vlan;
1016 1017
	struct device *classdev;
	dev_t devt;
1018
	int err;
1019 1020 1021 1022

	if (dev->rtnl_link_ops != &macvtap_link_ops)
		return NOTIFY_DONE;

1023
	vlan = netdev_priv(dev);
1024 1025 1026 1027 1028 1029 1030

	switch (event) {
	case NETDEV_REGISTER:
		/* Create the device node here after the network device has
		 * been registered but before register_netdevice has
		 * finished running.
		 */
1031 1032 1033 1034 1035
		err = macvtap_get_minor(vlan);
		if (err)
			return notifier_from_errno(err);

		devt = MKDEV(MAJOR(macvtap_major), vlan->minor);
1036 1037
		classdev = device_create(macvtap_class, &dev->dev, devt,
					 dev, "tap%d", dev->ifindex);
1038 1039
		if (IS_ERR(classdev)) {
			macvtap_free_minor(vlan);
1040
			return notifier_from_errno(PTR_ERR(classdev));
1041
		}
1042 1043
		break;
	case NETDEV_UNREGISTER:
1044
		devt = MKDEV(MAJOR(macvtap_major), vlan->minor);
1045
		device_destroy(macvtap_class, devt);
1046
		macvtap_free_minor(vlan);
1047 1048 1049 1050 1051 1052 1053 1054 1055 1056
		break;
	}

	return NOTIFY_DONE;
}

static struct notifier_block macvtap_notifier_block __read_mostly = {
	.notifier_call	= macvtap_device_event,
};

A
Arnd Bergmann 已提交
1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076
static int macvtap_init(void)
{
	int err;

	err = alloc_chrdev_region(&macvtap_major, 0,
				MACVTAP_NUM_DEVS, "macvtap");
	if (err)
		goto out1;

	cdev_init(&macvtap_cdev, &macvtap_fops);
	err = cdev_add(&macvtap_cdev, macvtap_major, MACVTAP_NUM_DEVS);
	if (err)
		goto out2;

	macvtap_class = class_create(THIS_MODULE, "macvtap");
	if (IS_ERR(macvtap_class)) {
		err = PTR_ERR(macvtap_class);
		goto out3;
	}

1077
	err = register_netdevice_notifier(&macvtap_notifier_block);
A
Arnd Bergmann 已提交
1078 1079 1080
	if (err)
		goto out4;

1081 1082 1083 1084
	err = macvlan_link_register(&macvtap_link_ops);
	if (err)
		goto out5;

A
Arnd Bergmann 已提交
1085 1086
	return 0;

1087 1088
out5:
	unregister_netdevice_notifier(&macvtap_notifier_block);
A
Arnd Bergmann 已提交
1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102
out4:
	class_unregister(macvtap_class);
out3:
	cdev_del(&macvtap_cdev);
out2:
	unregister_chrdev_region(macvtap_major, MACVTAP_NUM_DEVS);
out1:
	return err;
}
module_init(macvtap_init);

static void macvtap_exit(void)
{
	rtnl_link_unregister(&macvtap_link_ops);
1103
	unregister_netdevice_notifier(&macvtap_notifier_block);
A
Arnd Bergmann 已提交
1104 1105 1106 1107 1108 1109 1110 1111 1112
	class_unregister(macvtap_class);
	cdev_del(&macvtap_cdev);
	unregister_chrdev_region(macvtap_major, MACVTAP_NUM_DEVS);
}
module_exit(macvtap_exit);

MODULE_ALIAS_RTNL_LINK("macvtap");
MODULE_AUTHOR("Arnd Bergmann <arnd@arndb.de>");
MODULE_LICENSE("GPL");