vlan_dev.c 27.6 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5
/* -*- linux-c -*-
 * INET		802.1Q VLAN
 *		Ethernet-type device handling.
 *
 * Authors:	Ben Greear <greearb@candelatech.com>
P
Patrick McHardy 已提交
6
 *              Please send support related email to: netdev@vger.kernel.org
L
Linus Torvalds 已提交
7
 *              VLAN Home Page: http://www.candelatech.com/~greear/vlan.html
8
 *
L
Linus Torvalds 已提交
9 10 11 12 13 14
 * Fixes:       Mar 22 2001: Martin Bokaemper <mbokaemper@unispherenetworks.com>
 *                - reset skb->pkt_type on incoming packets when MAC was changed
 *                - see that changed MAC is saddr for outgoing packets
 *              Oct 20, 2001:  Ard van Breeman:
 *                - Fix MC-list, finally.
 *                - Flush MC-list on VLAN destroy.
15
 *
L
Linus Torvalds 已提交
16 17 18 19 20 21 22 23
 *
 *		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.
 */

#include <linux/module.h>
24
#include <linux/slab.h>
L
Linus Torvalds 已提交
25 26 27
#include <linux/skbuff.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
P
Patrick McHardy 已提交
28
#include <linux/ethtool.h>
L
Linus Torvalds 已提交
29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44
#include <net/arp.h>

#include "vlan.h"
#include "vlanproc.h"
#include <linux/if_vlan.h>

/*
 *	Rebuild the Ethernet MAC header. This is called after an ARP
 *	(or in future other address resolution) has completed on this
 *	sk_buff. We now let ARP fill in the other fields.
 *
 *	This routine CANNOT use cached dst->neigh!
 *	Really, it is used only when dst->neigh is wrong.
 *
 * TODO:  This needs a checkup, I'm ignorant here. --BLG
 */
45
static int vlan_dev_rebuild_header(struct sk_buff *skb)
L
Linus Torvalds 已提交
46 47 48 49 50 51
{
	struct net_device *dev = skb->dev;
	struct vlan_ethhdr *veth = (struct vlan_ethhdr *)(skb->data);

	switch (veth->h_vlan_encapsulated_proto) {
#ifdef CONFIG_INET
52
	case htons(ETH_P_IP):
L
Linus Torvalds 已提交
53 54 55

		/* TODO:  Confirm this will work with VLAN headers... */
		return arp_find(veth->h_dest, skb);
56
#endif
L
Linus Torvalds 已提交
57
	default:
58 59
		pr_debug("%s: unable to resolve type %X addresses.\n",
			 dev->name, ntohs(veth->h_vlan_encapsulated_proto));
60

L
Linus Torvalds 已提交
61 62
		memcpy(veth->h_source, dev->dev_addr, ETH_ALEN);
		break;
63
	}
L
Linus Torvalds 已提交
64 65 66 67 68 69

	return 0;
}

static inline struct sk_buff *vlan_check_reorder_header(struct sk_buff *skb)
{
70
	if (vlan_dev_info(skb->dev)->flags & VLAN_FLAG_REORDER_HDR) {
71 72
		if (skb_cow(skb, skb_headroom(skb)) < 0)
			skb = NULL;
L
Linus Torvalds 已提交
73 74 75 76
		if (skb) {
			/* Lifted from Gleb's VLAN code... */
			memmove(skb->data - ETH_HLEN,
				skb->data - VLAN_ETH_HLEN, 12);
77
			skb->mac_header += VLAN_HLEN;
L
Linus Torvalds 已提交
78 79 80 81 82 83
		}
	}

	return skb;
}

84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117
static inline void vlan_set_encap_proto(struct sk_buff *skb,
		struct vlan_hdr *vhdr)
{
	__be16 proto;
	unsigned char *rawp;

	/*
	 * Was a VLAN packet, grab the encapsulated protocol, which the layer
	 * three protocols care about.
	 */

	proto = vhdr->h_vlan_encapsulated_proto;
	if (ntohs(proto) >= 1536) {
		skb->protocol = proto;
		return;
	}

	rawp = skb->data;
	if (*(unsigned short *)rawp == 0xFFFF)
		/*
		 * This is a magic hack to spot IPX packets. Older Novell
		 * breaks the protocol design and runs IPX over 802.3 without
		 * an 802.2 LLC layer. We look for FFFF which isn't a used
		 * 802.2 SSAP/DSAP. This won't work for fault tolerant netware
		 * but does for the rest.
		 */
		skb->protocol = htons(ETH_P_802_3);
	else
		/*
		 * Real 802.2 LLC
		 */
		skb->protocol = htons(ETH_P_802_2);
}

L
Linus Torvalds 已提交
118
/*
119
 *	Determine the packet's protocol ID. The rule here is that we
L
Linus Torvalds 已提交
120 121 122 123 124 125 126 127 128 129 130 131 132 133 134
 *	assume 802.3 if the type field is short enough to be a length.
 *	This is normal practice and works for any 'now in use' protocol.
 *
 *  Also, at this point we assume that we ARE dealing exclusively with
 *  VLAN packets, or packets that should be made into VLAN packets based
 *  on a default VLAN ID.
 *
 *  NOTE:  Should be similar to ethernet/eth.c.
 *
 *  SANITY NOTE:  This method is called when a packet is moving up the stack
 *                towards userland.  To get here, it would have already passed
 *                through the ethernet/eth.c eth_type_trans() method.
 *  SANITY NOTE 2: We are referencing to the VLAN_HDR frields, which MAY be
 *                 stored UNALIGNED in the memory.  RISC systems don't like
 *                 such cases very much...
P
Patrick McHardy 已提交
135 136 137
 *  SANITY NOTE 2a: According to Dave Miller & Alexey, it will always be
 *  		    aligned, so there doesn't need to be any of the unaligned
 *  		    stuff.  It has been commented out now...  --Ben
L
Linus Torvalds 已提交
138 139 140
 *
 */
int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev,
P
Patrick McHardy 已提交
141
		  struct packet_type *ptype, struct net_device *orig_dev)
L
Linus Torvalds 已提交
142
{
143
	struct vlan_hdr *vhdr;
E
Eric Dumazet 已提交
144
	struct vlan_rx_stats *rx_stats;
145
	struct net_device *vlan_dev;
146 147
	u16 vlan_id;
	u16 vlan_tci;
L
Linus Torvalds 已提交
148

P
Patrick McHardy 已提交
149 150
	skb = skb_share_check(skb, GFP_ATOMIC);
	if (skb == NULL)
151
		goto err_free;
152

153 154
	if (unlikely(!pskb_may_pull(skb, VLAN_HLEN)))
		goto err_free;
155

156
	vhdr = (struct vlan_hdr *)skb->data;
157 158
	vlan_tci = ntohs(vhdr->h_vlan_TCI);
	vlan_id = vlan_tci & VLAN_VID_MASK;
L
Linus Torvalds 已提交
159 160

	rcu_read_lock();
161
	vlan_dev = __find_vlan_dev(dev, vlan_id);
L
Linus Torvalds 已提交
162

163 164 165 166 167 168
	/* If the VLAN device is defined, we use it.
	 * If not, and the VID is 0, it is a 802.1p packet (not
	 * really a VLAN), so we will just netif_rx it later to the
	 * original interface, but with the skb->proto set to the
	 * wrapped proto: we do nothing here.
	 */
L
Linus Torvalds 已提交
169

170 171 172 173 174 175 176 177 178 179
	if (!vlan_dev) {
		if (vlan_id) {
			pr_debug("%s: ERROR: No net_device for VID: %u on dev: %s\n",
				 __func__, vlan_id, dev->name);
			goto err_unlock;
		}
		rx_stats = NULL;
	} else {
		skb->dev = vlan_dev;

180 181
		rx_stats = this_cpu_ptr(vlan_dev_info(skb->dev)->vlan_rx_stats);

182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214
		u64_stats_update_begin(&rx_stats->syncp);
		rx_stats->rx_packets++;
		rx_stats->rx_bytes += skb->len;

		skb->priority = vlan_get_ingress_priority(skb->dev, vlan_tci);

		pr_debug("%s: priority: %u for TCI: %hu\n",
			 __func__, skb->priority, vlan_tci);

		switch (skb->pkt_type) {
		case PACKET_BROADCAST:
			/* Yeah, stats collect these together.. */
			/* stats->broadcast ++; // no such counter :-( */
			break;

		case PACKET_MULTICAST:
			rx_stats->rx_multicast++;
			break;

		case PACKET_OTHERHOST:
			/* Our lower layer thinks this is not local, let's make
			 * sure.
			 * This allows the VLAN to have a different MAC than the
			 * underlying device, and still route correctly.
			 */
			if (!compare_ether_addr(eth_hdr(skb)->h_dest,
						skb->dev->dev_addr))
				skb->pkt_type = PACKET_HOST;
			break;
		default:
			break;
		}
		u64_stats_update_end(&rx_stats->syncp);
215
	}
L
Linus Torvalds 已提交
216

217
	skb_pull_rcsum(skb, VLAN_HLEN);
218
	vlan_set_encap_proto(skb, vhdr);
L
Linus Torvalds 已提交
219

220 221 222 223 224 225
	if (vlan_dev) {
		skb = vlan_check_reorder_header(skb);
		if (!skb) {
			rx_stats->rx_errors++;
			goto err_unlock;
		}
L
Linus Torvalds 已提交
226
	}
227

228 229
	netif_rx(skb);

L
Linus Torvalds 已提交
230
	rcu_read_unlock();
231 232 233 234 235
	return NET_RX_SUCCESS;

err_unlock:
	rcu_read_unlock();
err_free:
236
	atomic_long_inc(&dev->rx_dropped);
237 238
	kfree_skb(skb);
	return NET_RX_DROP;
L
Linus Torvalds 已提交
239 240
}

241
static inline u16
P
Patrick McHardy 已提交
242
vlan_dev_get_egress_qos_mask(struct net_device *dev, struct sk_buff *skb)
L
Linus Torvalds 已提交
243
{
P
Patrick McHardy 已提交
244
	struct vlan_priority_tci_mapping *mp;
L
Linus Torvalds 已提交
245

P
Patrick McHardy 已提交
246
	mp = vlan_dev_info(dev)->egress_priority_map[(skb->priority & 0xF)];
L
Linus Torvalds 已提交
247 248
	while (mp) {
		if (mp->priority == skb->priority) {
P
Patrick McHardy 已提交
249 250 251
			return mp->vlan_qos; /* This should already be shifted
					      * to mask correctly with the
					      * VLAN's TCI */
L
Linus Torvalds 已提交
252 253 254 255 256 257 258
		}
		mp = mp->next;
	}
	return 0;
}

/*
259
 *	Create the VLAN header for an arbitrary protocol layer
L
Linus Torvalds 已提交
260 261 262 263 264 265 266
 *
 *	saddr=NULL	means use device source address
 *	daddr=NULL	means leave destination address (eg unresolved arp)
 *
 *  This is called when the SKB is moving down the stack towards the
 *  physical devices.
 */
267 268 269 270
static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
				unsigned short type,
				const void *daddr, const void *saddr,
				unsigned int len)
L
Linus Torvalds 已提交
271 272
{
	struct vlan_hdr *vhdr;
273
	unsigned int vhdrlen = 0;
274
	u16 vlan_tci = 0;
275
	int rc;
L
Linus Torvalds 已提交
276

277 278 279
	if (WARN_ON(skb_headroom(skb) < dev->hard_header_len))
		return -ENOSPC;

280
	if (!(vlan_dev_info(dev)->flags & VLAN_FLAG_REORDER_HDR)) {
L
Linus Torvalds 已提交
281 282
		vhdr = (struct vlan_hdr *) skb_push(skb, VLAN_HLEN);

283 284 285
		vlan_tci = vlan_dev_info(dev)->vlan_id;
		vlan_tci |= vlan_dev_get_egress_qos_mask(dev, skb);
		vhdr->h_vlan_TCI = htons(vlan_tci);
L
Linus Torvalds 已提交
286 287

		/*
O
Octavian Purdila 已提交
288 289
		 *  Set the protocol type. For a packet of type ETH_P_802_3/2 we
		 *  put the length in here instead.
L
Linus Torvalds 已提交
290
		 */
O
Octavian Purdila 已提交
291
		if (type != ETH_P_802_3 && type != ETH_P_802_2)
L
Linus Torvalds 已提交
292
			vhdr->h_vlan_encapsulated_proto = htons(type);
P
Patrick McHardy 已提交
293
		else
L
Linus Torvalds 已提交
294
			vhdr->h_vlan_encapsulated_proto = htons(len);
295 296

		skb->protocol = htons(ETH_P_8021Q);
297 298
		type = ETH_P_8021Q;
		vhdrlen = VLAN_HLEN;
L
Linus Torvalds 已提交
299 300 301 302 303 304
	}

	/* Before delegating work to the lower layer, enter our MAC-address */
	if (saddr == NULL)
		saddr = dev->dev_addr;

305
	/* Now make the underlying real hard header */
306
	dev = vlan_dev_info(dev)->real_dev;
307 308 309
	rc = dev_hard_header(skb, dev, type, daddr, saddr, len + vhdrlen);
	if (rc > 0)
		rc += vhdrlen;
L
Linus Torvalds 已提交
310 311 312
	return rc;
}

313 314
static netdev_tx_t vlan_dev_hard_start_xmit(struct sk_buff *skb,
					    struct net_device *dev)
L
Linus Torvalds 已提交
315
{
E
Eric Dumazet 已提交
316 317
	int i = skb_get_queue_mapping(skb);
	struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
L
Linus Torvalds 已提交
318
	struct vlan_ethhdr *veth = (struct vlan_ethhdr *)(skb->data);
E
Eric Dumazet 已提交
319 320
	unsigned int len;
	int ret;
321

L
Linus Torvalds 已提交
322 323 324 325 326
	/* Handle non-VLAN frames if they are sent to us, for example by DHCP.
	 *
	 * NOTE: THIS ASSUMES DIX ETHERNET, SPECIFICALLY NOT SUPPORTING
	 * OTHER THINGS LIKE FDDI/TokenRing/802.3 SNAPs...
	 */
327
	if (veth->h_vlan_proto != htons(ETH_P_8021Q) ||
328 329
	    vlan_dev_info(dev)->flags & VLAN_FLAG_REORDER_HDR) {
		unsigned int orig_headroom = skb_headroom(skb);
330
		u16 vlan_tci;
L
Linus Torvalds 已提交
331

332
		vlan_dev_info(dev)->cnt_encap_on_xmit++;
L
Linus Torvalds 已提交
333

334 335 336
		vlan_tci = vlan_dev_info(dev)->vlan_id;
		vlan_tci |= vlan_dev_get_egress_qos_mask(dev, skb);
		skb = __vlan_put_tag(skb, vlan_tci);
L
Linus Torvalds 已提交
337
		if (!skb) {
338
			txq->tx_dropped++;
339
			return NETDEV_TX_OK;
L
Linus Torvalds 已提交
340 341
		}

P
Patrick McHardy 已提交
342
		if (orig_headroom < VLAN_HLEN)
343
			vlan_dev_info(dev)->cnt_inc_headroom_on_tx++;
L
Linus Torvalds 已提交
344 345 346
	}


347
	skb_set_dev(skb, vlan_dev_info(dev)->real_dev);
E
Eric Dumazet 已提交
348 349 350
	len = skb->len;
	ret = dev_queue_xmit(skb);

351
	if (likely(ret == NET_XMIT_SUCCESS || ret == NET_XMIT_CN)) {
E
Eric Dumazet 已提交
352 353 354 355 356
		txq->tx_packets++;
		txq->tx_bytes += len;
	} else
		txq->tx_dropped++;

357
	return ret;
L
Linus Torvalds 已提交
358 359
}

360 361
static netdev_tx_t vlan_dev_hwaccel_hard_start_xmit(struct sk_buff *skb,
						    struct net_device *dev)
L
Linus Torvalds 已提交
362
{
E
Eric Dumazet 已提交
363 364
	int i = skb_get_queue_mapping(skb);
	struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
365
	u16 vlan_tci;
E
Eric Dumazet 已提交
366 367
	unsigned int len;
	int ret;
L
Linus Torvalds 已提交
368

369 370 371
	vlan_tci = vlan_dev_info(dev)->vlan_id;
	vlan_tci |= vlan_dev_get_egress_qos_mask(dev, skb);
	skb = __vlan_hwaccel_put_tag(skb, vlan_tci);
L
Linus Torvalds 已提交
372

373
	skb->dev = vlan_dev_info(dev)->real_dev;
E
Eric Dumazet 已提交
374 375 376
	len = skb->len;
	ret = dev_queue_xmit(skb);

377
	if (likely(ret == NET_XMIT_SUCCESS || ret == NET_XMIT_CN)) {
E
Eric Dumazet 已提交
378 379 380 381 382
		txq->tx_packets++;
		txq->tx_bytes += len;
	} else
		txq->tx_dropped++;

383
	return ret;
L
Linus Torvalds 已提交
384 385
}

V
Vasu Dev 已提交
386 387 388 389 390 391 392 393
static u16 vlan_dev_select_queue(struct net_device *dev, struct sk_buff *skb)
{
	struct net_device *rdev = vlan_dev_info(dev)->real_dev;
	const struct net_device_ops *ops = rdev->netdev_ops;

	return ops->ndo_select_queue(rdev, skb);
}

394
static int vlan_dev_change_mtu(struct net_device *dev, int new_mtu)
L
Linus Torvalds 已提交
395 396 397 398
{
	/* TODO: gotta make sure the underlying layer can handle it,
	 * maybe an IFF_VLAN_CAPABLE flag for devices?
	 */
399
	if (vlan_dev_info(dev)->real_dev->mtu < new_mtu)
L
Linus Torvalds 已提交
400 401 402 403 404 405 406
		return -ERANGE;

	dev->mtu = new_mtu;

	return 0;
}

407
void vlan_dev_set_ingress_priority(const struct net_device *dev,
408
				   u32 skb_prio, u16 vlan_prio)
L
Linus Torvalds 已提交
409
{
410
	struct vlan_dev_info *vlan = vlan_dev_info(dev);
411 412 413 414 415 416 417

	if (vlan->ingress_priority_map[vlan_prio & 0x7] && !skb_prio)
		vlan->nr_ingress_mappings--;
	else if (!vlan->ingress_priority_map[vlan_prio & 0x7] && skb_prio)
		vlan->nr_ingress_mappings++;

	vlan->ingress_priority_map[vlan_prio & 0x7] = skb_prio;
L
Linus Torvalds 已提交
418 419
}

420
int vlan_dev_set_egress_priority(const struct net_device *dev,
421
				 u32 skb_prio, u16 vlan_prio)
L
Linus Torvalds 已提交
422
{
423
	struct vlan_dev_info *vlan = vlan_dev_info(dev);
L
Linus Torvalds 已提交
424 425
	struct vlan_priority_tci_mapping *mp = NULL;
	struct vlan_priority_tci_mapping *np;
E
Eric Dumazet 已提交
426
	u32 vlan_qos = (vlan_prio << VLAN_PRIO_SHIFT) & VLAN_PRIO_MASK;
427

428
	/* See if a priority mapping exists.. */
429
	mp = vlan->egress_priority_map[skb_prio & 0xF];
430 431
	while (mp) {
		if (mp->priority == skb_prio) {
432 433 434 435 436
			if (mp->vlan_qos && !vlan_qos)
				vlan->nr_egress_mappings--;
			else if (!mp->vlan_qos && vlan_qos)
				vlan->nr_egress_mappings++;
			mp->vlan_qos = vlan_qos;
437
			return 0;
L
Linus Torvalds 已提交
438
		}
439
		mp = mp->next;
L
Linus Torvalds 已提交
440
	}
441 442

	/* Create a new mapping then. */
443
	mp = vlan->egress_priority_map[skb_prio & 0xF];
444 445 446 447 448 449
	np = kmalloc(sizeof(struct vlan_priority_tci_mapping), GFP_KERNEL);
	if (!np)
		return -ENOBUFS;

	np->next = mp;
	np->priority = skb_prio;
450 451 452 453
	np->vlan_qos = vlan_qos;
	vlan->egress_priority_map[skb_prio & 0xF] = np;
	if (vlan_qos)
		vlan->nr_egress_mappings++;
454
	return 0;
L
Linus Torvalds 已提交
455 456
}

457
/* Flags are defined in the vlan_flags enum in include/linux/if_vlan.h file. */
458
int vlan_dev_change_flags(const struct net_device *dev, u32 flags, u32 mask)
L
Linus Torvalds 已提交
459
{
460 461 462
	struct vlan_dev_info *vlan = vlan_dev_info(dev);
	u32 old_flags = vlan->flags;

463 464
	if (mask & ~(VLAN_FLAG_REORDER_HDR | VLAN_FLAG_GVRP |
		     VLAN_FLAG_LOOSE_BINDING))
465 466 467
		return -EINVAL;

	vlan->flags = (old_flags & ~mask) | (flags & mask);
P
Patrick McHardy 已提交
468 469 470 471 472 473 474

	if (netif_running(dev) && (vlan->flags ^ old_flags) & VLAN_FLAG_GVRP) {
		if (vlan->flags & VLAN_FLAG_GVRP)
			vlan_gvrp_request_join(dev);
		else
			vlan_gvrp_request_leave(dev);
	}
475
	return 0;
L
Linus Torvalds 已提交
476 477
}

478
void vlan_dev_get_realdev_name(const struct net_device *dev, char *result)
L
Linus Torvalds 已提交
479
{
480
	strncpy(result, vlan_dev_info(dev)->real_dev->name, 23);
L
Linus Torvalds 已提交
481 482
}

483
static int vlan_dev_open(struct net_device *dev)
L
Linus Torvalds 已提交
484
{
485
	struct vlan_dev_info *vlan = vlan_dev_info(dev);
486 487 488
	struct net_device *real_dev = vlan->real_dev;
	int err;

489 490
	if (!(real_dev->flags & IFF_UP) &&
	    !(vlan->flags & VLAN_FLAG_LOOSE_BINDING))
L
Linus Torvalds 已提交
491 492
		return -ENETDOWN;

493
	if (compare_ether_addr(dev->dev_addr, real_dev->dev_addr)) {
494
		err = dev_uc_add(real_dev, dev->dev_addr);
495
		if (err < 0)
496
			goto out;
497 498
	}

499 500 501 502 503 504 505 506 507 508 509 510
	if (dev->flags & IFF_ALLMULTI) {
		err = dev_set_allmulti(real_dev, 1);
		if (err < 0)
			goto del_unicast;
	}
	if (dev->flags & IFF_PROMISC) {
		err = dev_set_promiscuity(real_dev, 1);
		if (err < 0)
			goto clear_allmulti;
	}

	memcpy(vlan->real_dev_addr, real_dev->dev_addr, ETH_ALEN);
511

P
Patrick McHardy 已提交
512 513 514
	if (vlan->flags & VLAN_FLAG_GVRP)
		vlan_gvrp_request_join(dev);

515 516
	if (netif_carrier_ok(real_dev))
		netif_carrier_on(dev);
L
Linus Torvalds 已提交
517
	return 0;
518 519 520 521 522 523

clear_allmulti:
	if (dev->flags & IFF_ALLMULTI)
		dev_set_allmulti(real_dev, -1);
del_unicast:
	if (compare_ether_addr(dev->dev_addr, real_dev->dev_addr))
524
		dev_uc_del(real_dev, dev->dev_addr);
525
out:
526
	netif_carrier_off(dev);
527
	return err;
L
Linus Torvalds 已提交
528 529
}

530
static int vlan_dev_stop(struct net_device *dev)
L
Linus Torvalds 已提交
531
{
P
Patrick McHardy 已提交
532 533 534 535 536
	struct vlan_dev_info *vlan = vlan_dev_info(dev);
	struct net_device *real_dev = vlan->real_dev;

	if (vlan->flags & VLAN_FLAG_GVRP)
		vlan_gvrp_request_leave(dev);
537

538
	dev_mc_unsync(real_dev, dev);
539
	dev_uc_unsync(real_dev, dev);
540 541 542 543 544
	if (dev->flags & IFF_ALLMULTI)
		dev_set_allmulti(real_dev, -1);
	if (dev->flags & IFF_PROMISC)
		dev_set_promiscuity(real_dev, -1);

545
	if (compare_ether_addr(dev->dev_addr, real_dev->dev_addr))
546
		dev_uc_del(real_dev, dev->dev_addr);
547

548
	netif_carrier_off(dev);
L
Linus Torvalds 已提交
549 550 551
	return 0;
}

552
static int vlan_dev_set_mac_address(struct net_device *dev, void *p)
553
{
554
	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
555 556 557 558 559 560 561 562 563 564
	struct sockaddr *addr = p;
	int err;

	if (!is_valid_ether_addr(addr->sa_data))
		return -EADDRNOTAVAIL;

	if (!(dev->flags & IFF_UP))
		goto out;

	if (compare_ether_addr(addr->sa_data, real_dev->dev_addr)) {
565
		err = dev_uc_add(real_dev, addr->sa_data);
566 567 568 569 570
		if (err < 0)
			return err;
	}

	if (compare_ether_addr(dev->dev_addr, real_dev->dev_addr))
571
		dev_uc_del(real_dev, dev->dev_addr);
572 573 574 575 576 577

out:
	memcpy(dev->dev_addr, addr->sa_data, ETH_ALEN);
	return 0;
}

578
static int vlan_dev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
L
Linus Torvalds 已提交
579
{
580
	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
581
	const struct net_device_ops *ops = real_dev->netdev_ops;
L
Linus Torvalds 已提交
582 583 584 585 586 587
	struct ifreq ifrr;
	int err = -EOPNOTSUPP;

	strncpy(ifrr.ifr_name, real_dev->name, IFNAMSIZ);
	ifrr.ifr_ifru = ifr->ifr_ifru;

P
Patrick McHardy 已提交
588
	switch (cmd) {
L
Linus Torvalds 已提交
589 590 591
	case SIOCGMIIPHY:
	case SIOCGMIIREG:
	case SIOCSMIIREG:
592 593
		if (netif_device_present(real_dev) && ops->ndo_do_ioctl)
			err = ops->ndo_do_ioctl(real_dev, &ifrr, cmd);
L
Linus Torvalds 已提交
594 595 596
		break;
	}

597
	if (!err)
L
Linus Torvalds 已提交
598 599 600 601 602
		ifr->ifr_ifru = ifrr.ifr_ifru;

	return err;
}

F
Frank Blaschka 已提交
603 604 605 606 607 608 609
static int vlan_dev_neigh_setup(struct net_device *dev, struct neigh_parms *pa)
{
	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
	const struct net_device_ops *ops = real_dev->netdev_ops;
	int err = 0;

	if (netif_device_present(real_dev) && ops->ndo_neigh_setup)
610
		err = ops->ndo_neigh_setup(real_dev, pa);
F
Frank Blaschka 已提交
611 612 613 614

	return err;
}

615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639
#if defined(CONFIG_FCOE) || defined(CONFIG_FCOE_MODULE)
static int vlan_dev_fcoe_ddp_setup(struct net_device *dev, u16 xid,
				   struct scatterlist *sgl, unsigned int sgc)
{
	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
	const struct net_device_ops *ops = real_dev->netdev_ops;
	int rc = 0;

	if (ops->ndo_fcoe_ddp_setup)
		rc = ops->ndo_fcoe_ddp_setup(real_dev, xid, sgl, sgc);

	return rc;
}

static int vlan_dev_fcoe_ddp_done(struct net_device *dev, u16 xid)
{
	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
	const struct net_device_ops *ops = real_dev->netdev_ops;
	int len = 0;

	if (ops->ndo_fcoe_ddp_done)
		len = ops->ndo_fcoe_ddp_done(real_dev, xid);

	return len;
}
640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661

static int vlan_dev_fcoe_enable(struct net_device *dev)
{
	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
	const struct net_device_ops *ops = real_dev->netdev_ops;
	int rc = -EINVAL;

	if (ops->ndo_fcoe_enable)
		rc = ops->ndo_fcoe_enable(real_dev);
	return rc;
}

static int vlan_dev_fcoe_disable(struct net_device *dev)
{
	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
	const struct net_device_ops *ops = real_dev->netdev_ops;
	int rc = -EINVAL;

	if (ops->ndo_fcoe_disable)
		rc = ops->ndo_fcoe_disable(real_dev);
	return rc;
}
662 663 664 665 666 667 668 669 670 671 672

static int vlan_dev_fcoe_get_wwn(struct net_device *dev, u64 *wwn, int type)
{
	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
	const struct net_device_ops *ops = real_dev->netdev_ops;
	int rc = -EINVAL;

	if (ops->ndo_fcoe_get_wwn)
		rc = ops->ndo_fcoe_get_wwn(real_dev, wwn, type);
	return rc;
}
673 674
#endif

675
static void vlan_dev_change_rx_flags(struct net_device *dev, int change)
676
{
677
	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
678 679 680 681 682 683 684

	if (change & IFF_ALLMULTI)
		dev_set_allmulti(real_dev, dev->flags & IFF_ALLMULTI ? 1 : -1);
	if (change & IFF_PROMISC)
		dev_set_promiscuity(real_dev, dev->flags & IFF_PROMISC ? 1 : -1);
}

685
static void vlan_dev_set_rx_mode(struct net_device *vlan_dev)
L
Linus Torvalds 已提交
686
{
687
	dev_mc_sync(vlan_dev_info(vlan_dev)->real_dev, vlan_dev);
688
	dev_uc_sync(vlan_dev_info(vlan_dev)->real_dev, vlan_dev);
L
Linus Torvalds 已提交
689
}
690 691 692 693 694 695 696

/*
 * vlan network devices have devices nesting below it, and are a special
 * "super class" of normal network devices; split their locks off into a
 * separate class since they always nest.
 */
static struct lock_class_key vlan_netdev_xmit_lock_key;
697
static struct lock_class_key vlan_netdev_addr_lock_key;
698

699 700 701
static void vlan_dev_set_lockdep_one(struct net_device *dev,
				     struct netdev_queue *txq,
				     void *_subclass)
702 703
{
	lockdep_set_class_and_subclass(&txq->_xmit_lock,
704 705
				       &vlan_netdev_xmit_lock_key,
				       *(int *)_subclass);
706 707 708 709
}

static void vlan_dev_set_lockdep_class(struct net_device *dev, int subclass)
{
710 711 712
	lockdep_set_class_and_subclass(&dev->addr_list_lock,
				       &vlan_netdev_addr_lock_key,
				       subclass);
713
	netdev_for_each_tx_queue(dev, vlan_dev_set_lockdep_one, &subclass);
714 715
}

716 717 718 719 720 721
static const struct header_ops vlan_header_ops = {
	.create	 = vlan_dev_hard_header,
	.rebuild = vlan_dev_rebuild_header,
	.parse	 = eth_header_parse,
};

V
Vasu Dev 已提交
722 723
static const struct net_device_ops vlan_netdev_ops, vlan_netdev_accel_ops,
		    vlan_netdev_ops_sq, vlan_netdev_accel_ops_sq;
724

725 726
static int vlan_dev_init(struct net_device *dev)
{
727
	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
728 729
	int subclass = 0;

730 731
	netif_carrier_off(dev);

732
	/* IFF_BROADCAST|IFF_MULTICAST; ??? */
733 734
	dev->flags  = real_dev->flags & ~(IFF_UP | IFF_PROMISC | IFF_ALLMULTI |
					  IFF_MASTER | IFF_SLAVE);
735 736 737 738 739
	dev->iflink = real_dev->ifindex;
	dev->state  = (real_dev->state & ((1<<__LINK_STATE_NOCARRIER) |
					  (1<<__LINK_STATE_DORMANT))) |
		      (1<<__LINK_STATE_PRESENT);

740
	dev->features |= real_dev->features & real_dev->vlan_features;
741
	dev->gso_max_size = real_dev->gso_max_size;
742

743 744 745 746 747 748 749 750
	/* ipv6 shared card related stuff */
	dev->dev_id = real_dev->dev_id;

	if (is_zero_ether_addr(dev->dev_addr))
		memcpy(dev->dev_addr, real_dev->dev_addr, dev->addr_len);
	if (is_zero_ether_addr(dev->broadcast))
		memcpy(dev->broadcast, real_dev->broadcast, dev->addr_len);

751 752 753 754
#if defined(CONFIG_FCOE) || defined(CONFIG_FCOE_MODULE)
	dev->fcoe_ddp_xid = real_dev->fcoe_ddp_xid;
#endif

755 756 757
	if (real_dev->features & NETIF_F_HW_VLAN_TX) {
		dev->header_ops      = real_dev->header_ops;
		dev->hard_header_len = real_dev->hard_header_len;
V
Vasu Dev 已提交
758 759 760 761
		if (real_dev->netdev_ops->ndo_select_queue)
			dev->netdev_ops = &vlan_netdev_accel_ops_sq;
		else
			dev->netdev_ops = &vlan_netdev_accel_ops;
762 763 764
	} else {
		dev->header_ops      = &vlan_header_ops;
		dev->hard_header_len = real_dev->hard_header_len + VLAN_HLEN;
V
Vasu Dev 已提交
765 766 767 768
		if (real_dev->netdev_ops->ndo_select_queue)
			dev->netdev_ops = &vlan_netdev_ops_sq;
		else
			dev->netdev_ops = &vlan_netdev_ops;
769 770
	}

J
Joonwoo Park 已提交
771
	if (is_vlan_dev(real_dev))
772 773
		subclass = 1;

774
	vlan_dev_set_lockdep_class(dev, subclass);
E
Eric Dumazet 已提交
775 776 777 778 779

	vlan_dev_info(dev)->vlan_rx_stats = alloc_percpu(struct vlan_rx_stats);
	if (!vlan_dev_info(dev)->vlan_rx_stats)
		return -ENOMEM;

780 781 782
	return 0;
}

783 784 785 786 787 788
static void vlan_dev_uninit(struct net_device *dev)
{
	struct vlan_priority_tci_mapping *pm;
	struct vlan_dev_info *vlan = vlan_dev_info(dev);
	int i;

E
Eric Dumazet 已提交
789 790
	free_percpu(vlan->vlan_rx_stats);
	vlan->vlan_rx_stats = NULL;
791 792 793 794 795 796 797 798
	for (i = 0; i < ARRAY_SIZE(vlan->egress_priority_map); i++) {
		while ((pm = vlan->egress_priority_map[i]) != NULL) {
			vlan->egress_priority_map[i] = pm->next;
			kfree(pm);
		}
	}
}

799 800 801 802
static int vlan_ethtool_get_settings(struct net_device *dev,
				     struct ethtool_cmd *cmd)
{
	const struct vlan_dev_info *vlan = vlan_dev_info(dev);
803
	return dev_ethtool_get_settings(vlan->real_dev, cmd);
804 805 806 807 808 809 810 811 812 813
}

static void vlan_ethtool_get_drvinfo(struct net_device *dev,
				     struct ethtool_drvinfo *info)
{
	strcpy(info->driver, vlan_fullname);
	strcpy(info->version, vlan_version);
	strcpy(info->fw_version, "N/A");
}

P
Patrick McHardy 已提交
814 815 816
static u32 vlan_ethtool_get_rx_csum(struct net_device *dev)
{
	const struct vlan_dev_info *vlan = vlan_dev_info(dev);
817
	return dev_ethtool_get_rx_csum(vlan->real_dev);
P
Patrick McHardy 已提交
818 819
}

820 821 822
static u32 vlan_ethtool_get_flags(struct net_device *dev)
{
	const struct vlan_dev_info *vlan = vlan_dev_info(dev);
823
	return dev_ethtool_get_flags(vlan->real_dev);
824 825
}

826
static struct rtnl_link_stats64 *vlan_dev_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
E
Eric Dumazet 已提交
827
{
828
	dev_txq_stats_fold(dev, stats);
E
Eric Dumazet 已提交
829 830

	if (vlan_dev_info(dev)->vlan_rx_stats) {
E
Eric Dumazet 已提交
831
		struct vlan_rx_stats *p, accum = {0};
E
Eric Dumazet 已提交
832 833 834
		int i;

		for_each_possible_cpu(i) {
E
Eric Dumazet 已提交
835 836 837
			u64 rxpackets, rxbytes, rxmulticast;
			unsigned int start;

E
Eric Dumazet 已提交
838
			p = per_cpu_ptr(vlan_dev_info(dev)->vlan_rx_stats, i);
E
Eric Dumazet 已提交
839 840 841 842 843 844 845 846 847
			do {
				start = u64_stats_fetch_begin_bh(&p->syncp);
				rxpackets	= p->rx_packets;
				rxbytes		= p->rx_bytes;
				rxmulticast	= p->rx_multicast;
			} while (u64_stats_fetch_retry_bh(&p->syncp, start));
			accum.rx_packets += rxpackets;
			accum.rx_bytes   += rxbytes;
			accum.rx_multicast += rxmulticast;
848
			/* rx_errors is ulong, not protected by syncp */
E
Eric Dumazet 已提交
849
			accum.rx_errors  += p->rx_errors;
E
Eric Dumazet 已提交
850
		}
E
Eric Dumazet 已提交
851 852 853 854
		stats->rx_packets = accum.rx_packets;
		stats->rx_bytes   = accum.rx_bytes;
		stats->rx_errors  = accum.rx_errors;
		stats->multicast  = accum.rx_multicast;
E
Eric Dumazet 已提交
855 856 857 858
	}
	return stats;
}

859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877
static int vlan_ethtool_set_tso(struct net_device *dev, u32 data)
{
       if (data) {
		struct net_device *real_dev = vlan_dev_info(dev)->real_dev;

		/* Underlying device must support TSO for VLAN-tagged packets
		 * and must have TSO enabled now.
		 */
		if (!(real_dev->vlan_features & NETIF_F_TSO))
			return -EOPNOTSUPP;
		if (!(real_dev->features & NETIF_F_TSO))
			return -EINVAL;
		dev->features |= NETIF_F_TSO;
	} else {
		dev->features &= ~NETIF_F_TSO;
	}
	return 0;
}

P
Patrick McHardy 已提交
878
static const struct ethtool_ops vlan_ethtool_ops = {
879 880
	.get_settings	        = vlan_ethtool_get_settings,
	.get_drvinfo	        = vlan_ethtool_get_drvinfo,
P
Patrick McHardy 已提交
881 882
	.get_link		= ethtool_op_get_link,
	.get_rx_csum		= vlan_ethtool_get_rx_csum,
883
	.get_flags		= vlan_ethtool_get_flags,
884
	.set_tso                = vlan_ethtool_set_tso,
P
Patrick McHardy 已提交
885 886
};

887 888 889 890 891 892
static const struct net_device_ops vlan_netdev_ops = {
	.ndo_change_mtu		= vlan_dev_change_mtu,
	.ndo_init		= vlan_dev_init,
	.ndo_uninit		= vlan_dev_uninit,
	.ndo_open		= vlan_dev_open,
	.ndo_stop		= vlan_dev_stop,
893 894 895 896 897 898 899
	.ndo_start_xmit =  vlan_dev_hard_start_xmit,
	.ndo_validate_addr	= eth_validate_addr,
	.ndo_set_mac_address	= vlan_dev_set_mac_address,
	.ndo_set_rx_mode	= vlan_dev_set_rx_mode,
	.ndo_set_multicast_list	= vlan_dev_set_rx_mode,
	.ndo_change_rx_flags	= vlan_dev_change_rx_flags,
	.ndo_do_ioctl		= vlan_dev_ioctl,
F
Frank Blaschka 已提交
900
	.ndo_neigh_setup	= vlan_dev_neigh_setup,
E
Eric Dumazet 已提交
901
	.ndo_get_stats64	= vlan_dev_get_stats64,
902 903 904
#if defined(CONFIG_FCOE) || defined(CONFIG_FCOE_MODULE)
	.ndo_fcoe_ddp_setup	= vlan_dev_fcoe_ddp_setup,
	.ndo_fcoe_ddp_done	= vlan_dev_fcoe_ddp_done,
905 906
	.ndo_fcoe_enable	= vlan_dev_fcoe_enable,
	.ndo_fcoe_disable	= vlan_dev_fcoe_disable,
907
	.ndo_fcoe_get_wwn	= vlan_dev_fcoe_get_wwn,
908
#endif
909 910 911 912 913 914 915 916
};

static const struct net_device_ops vlan_netdev_accel_ops = {
	.ndo_change_mtu		= vlan_dev_change_mtu,
	.ndo_init		= vlan_dev_init,
	.ndo_uninit		= vlan_dev_uninit,
	.ndo_open		= vlan_dev_open,
	.ndo_stop		= vlan_dev_stop,
V
Vasu Dev 已提交
917 918 919 920 921 922 923 924
	.ndo_start_xmit =  vlan_dev_hwaccel_hard_start_xmit,
	.ndo_validate_addr	= eth_validate_addr,
	.ndo_set_mac_address	= vlan_dev_set_mac_address,
	.ndo_set_rx_mode	= vlan_dev_set_rx_mode,
	.ndo_set_multicast_list	= vlan_dev_set_rx_mode,
	.ndo_change_rx_flags	= vlan_dev_change_rx_flags,
	.ndo_do_ioctl		= vlan_dev_ioctl,
	.ndo_neigh_setup	= vlan_dev_neigh_setup,
E
Eric Dumazet 已提交
925
	.ndo_get_stats64	= vlan_dev_get_stats64,
V
Vasu Dev 已提交
926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949
#if defined(CONFIG_FCOE) || defined(CONFIG_FCOE_MODULE)
	.ndo_fcoe_ddp_setup	= vlan_dev_fcoe_ddp_setup,
	.ndo_fcoe_ddp_done	= vlan_dev_fcoe_ddp_done,
	.ndo_fcoe_enable	= vlan_dev_fcoe_enable,
	.ndo_fcoe_disable	= vlan_dev_fcoe_disable,
	.ndo_fcoe_get_wwn	= vlan_dev_fcoe_get_wwn,
#endif
};

static const struct net_device_ops vlan_netdev_ops_sq = {
	.ndo_select_queue	= vlan_dev_select_queue,
	.ndo_change_mtu		= vlan_dev_change_mtu,
	.ndo_init		= vlan_dev_init,
	.ndo_uninit		= vlan_dev_uninit,
	.ndo_open		= vlan_dev_open,
	.ndo_stop		= vlan_dev_stop,
	.ndo_start_xmit =  vlan_dev_hard_start_xmit,
	.ndo_validate_addr	= eth_validate_addr,
	.ndo_set_mac_address	= vlan_dev_set_mac_address,
	.ndo_set_rx_mode	= vlan_dev_set_rx_mode,
	.ndo_set_multicast_list	= vlan_dev_set_rx_mode,
	.ndo_change_rx_flags	= vlan_dev_change_rx_flags,
	.ndo_do_ioctl		= vlan_dev_ioctl,
	.ndo_neigh_setup	= vlan_dev_neigh_setup,
E
Eric Dumazet 已提交
950
	.ndo_get_stats64	= vlan_dev_get_stats64,
V
Vasu Dev 已提交
951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966
#if defined(CONFIG_FCOE) || defined(CONFIG_FCOE_MODULE)
	.ndo_fcoe_ddp_setup	= vlan_dev_fcoe_ddp_setup,
	.ndo_fcoe_ddp_done	= vlan_dev_fcoe_ddp_done,
	.ndo_fcoe_enable	= vlan_dev_fcoe_enable,
	.ndo_fcoe_disable	= vlan_dev_fcoe_disable,
	.ndo_fcoe_get_wwn	= vlan_dev_fcoe_get_wwn,
#endif
};

static const struct net_device_ops vlan_netdev_accel_ops_sq = {
	.ndo_select_queue	= vlan_dev_select_queue,
	.ndo_change_mtu		= vlan_dev_change_mtu,
	.ndo_init		= vlan_dev_init,
	.ndo_uninit		= vlan_dev_uninit,
	.ndo_open		= vlan_dev_open,
	.ndo_stop		= vlan_dev_stop,
967
	.ndo_start_xmit =  vlan_dev_hwaccel_hard_start_xmit,
968 969 970 971 972 973
	.ndo_validate_addr	= eth_validate_addr,
	.ndo_set_mac_address	= vlan_dev_set_mac_address,
	.ndo_set_rx_mode	= vlan_dev_set_rx_mode,
	.ndo_set_multicast_list	= vlan_dev_set_rx_mode,
	.ndo_change_rx_flags	= vlan_dev_change_rx_flags,
	.ndo_do_ioctl		= vlan_dev_ioctl,
F
Frank Blaschka 已提交
974
	.ndo_neigh_setup	= vlan_dev_neigh_setup,
E
Eric Dumazet 已提交
975
	.ndo_get_stats64	= vlan_dev_get_stats64,
976 977 978
#if defined(CONFIG_FCOE) || defined(CONFIG_FCOE_MODULE)
	.ndo_fcoe_ddp_setup	= vlan_dev_fcoe_ddp_setup,
	.ndo_fcoe_ddp_done	= vlan_dev_fcoe_ddp_done,
979 980
	.ndo_fcoe_enable	= vlan_dev_fcoe_enable,
	.ndo_fcoe_disable	= vlan_dev_fcoe_disable,
981
	.ndo_fcoe_get_wwn	= vlan_dev_fcoe_get_wwn,
982
#endif
983 984
};

985 986 987 988 989
void vlan_setup(struct net_device *dev)
{
	ether_setup(dev);

	dev->priv_flags		|= IFF_802_1Q_VLAN;
990
	dev->priv_flags		&= ~IFF_XMIT_DST_RELEASE;
991 992
	dev->tx_queue_len	= 0;

993
	dev->netdev_ops		= &vlan_netdev_ops;
994
	dev->destructor		= free_netdev;
P
Patrick McHardy 已提交
995
	dev->ethtool_ops	= &vlan_ethtool_ops;
996 997 998

	memset(dev->broadcast, 0, ETH_ALEN);
}