af_mpls.c 49.7 KB
Newer Older
E
Eric W. Biederman 已提交
1 2 3
#include <linux/types.h>
#include <linux/skbuff.h>
#include <linux/socket.h>
4
#include <linux/sysctl.h>
E
Eric W. Biederman 已提交
5 6 7 8 9
#include <linux/net.h>
#include <linux/module.h>
#include <linux/if_arp.h>
#include <linux/ipv6.h>
#include <linux/mpls.h>
D
David Ahern 已提交
10
#include <linux/netconf.h>
11
#include <linux/vmalloc.h>
R
Robert Shearman 已提交
12
#include <linux/percpu.h>
E
Eric W. Biederman 已提交
13 14 15 16 17 18 19
#include <net/ip.h>
#include <net/dst.h>
#include <net/sock.h>
#include <net/arp.h>
#include <net/ip_fib.h>
#include <net/netevent.h>
#include <net/netns/generic.h>
20 21 22
#if IS_ENABLED(CONFIG_IPV6)
#include <net/ipv6.h>
#endif
R
Robert Shearman 已提交
23
#include <net/addrconf.h>
R
Roopa Prabhu 已提交
24
#include <net/nexthop.h>
E
Eric W. Biederman 已提交
25 26
#include "internal.h"

27 28 29 30 31
/* Maximum number of labels to look ahead at when selecting a path of
 * a multipath route
 */
#define MAX_MP_SELECT_LABELS 4

32 33
#define MPLS_NEIGH_TABLE_UNSPEC (NEIGH_LINK_TABLE + 1)

34
static int zero = 0;
35
static int one = 1;
36
static int label_limit = (1 << 20) - 1;
37
static int ttl_max = 255;
38

39 40 41 42
static void rtmsg_lfib(int event, u32 label, struct mpls_route *rt,
		       struct nlmsghdr *nlh, struct net *net, u32 portid,
		       unsigned int nlm_flags);

E
Eric W. Biederman 已提交
43 44 45 46 47 48 49 50 51 52 53 54
static struct mpls_route *mpls_route_input_rcu(struct net *net, unsigned index)
{
	struct mpls_route *rt = NULL;

	if (index < net->mpls.platform_labels) {
		struct mpls_route __rcu **platform_label =
			rcu_dereference(net->mpls.platform_label);
		rt = rcu_dereference(platform_label[index]);
	}
	return rt;
}

55
bool mpls_output_possible(const struct net_device *dev)
E
Eric W. Biederman 已提交
56 57 58
{
	return dev && (dev->flags & IFF_UP) && netif_carrier_ok(dev);
}
59
EXPORT_SYMBOL_GPL(mpls_output_possible);
E
Eric W. Biederman 已提交
60

61 62 63 64 65 66 67 68 69 70 71 72 73 74
static u8 *__mpls_nh_via(struct mpls_route *rt, struct mpls_nh *nh)
{
	u8 *nh0_via = PTR_ALIGN((u8 *)&rt->rt_nh[rt->rt_nhn], VIA_ALEN_ALIGN);
	int nh_index = nh - rt->rt_nh;

	return nh0_via + rt->rt_max_alen * nh_index;
}

static const u8 *mpls_nh_via(const struct mpls_route *rt,
			     const struct mpls_nh *nh)
{
	return __mpls_nh_via((struct mpls_route *)rt, (struct mpls_nh *)nh);
}

R
Roopa Prabhu 已提交
75
static unsigned int mpls_nh_header_size(const struct mpls_nh *nh)
E
Eric W. Biederman 已提交
76 77
{
	/* The size of the layer 2.5 labels to be added for this route */
R
Roopa Prabhu 已提交
78
	return nh->nh_labels * sizeof(struct mpls_shim_hdr);
E
Eric W. Biederman 已提交
79 80
}

81
unsigned int mpls_dev_mtu(const struct net_device *dev)
E
Eric W. Biederman 已提交
82 83 84 85
{
	/* The amount of data the layer 2 frame can hold */
	return dev->mtu;
}
86
EXPORT_SYMBOL_GPL(mpls_dev_mtu);
E
Eric W. Biederman 已提交
87

88
bool mpls_pkt_too_big(const struct sk_buff *skb, unsigned int mtu)
E
Eric W. Biederman 已提交
89 90 91 92
{
	if (skb->len <= mtu)
		return false;

93
	if (skb_is_gso(skb) && skb_gso_validate_mtu(skb, mtu))
E
Eric W. Biederman 已提交
94 95 96 97
		return false;

	return true;
}
98
EXPORT_SYMBOL_GPL(mpls_pkt_too_big);
E
Eric W. Biederman 已提交
99

R
Robert Shearman 已提交
100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124
void mpls_stats_inc_outucastpkts(struct net_device *dev,
				 const struct sk_buff *skb)
{
	struct mpls_dev *mdev;

	if (skb->protocol == htons(ETH_P_MPLS_UC)) {
		mdev = mpls_dev_get(dev);
		if (mdev)
			MPLS_INC_STATS_LEN(mdev, skb->len,
					   tx_packets,
					   tx_bytes);
	} else if (skb->protocol == htons(ETH_P_IP)) {
		IP_UPD_PO_STATS(dev_net(dev), IPSTATS_MIB_OUT, skb->len);
#if IS_ENABLED(CONFIG_IPV6)
	} else if (skb->protocol == htons(ETH_P_IPV6)) {
		struct inet6_dev *in6dev = __in6_dev_get(dev);

		if (in6dev)
			IP6_UPD_PO_STATS(dev_net(dev), in6dev,
					 IPSTATS_MIB_OUT, skb->len);
#endif
	}
}
EXPORT_SYMBOL_GPL(mpls_stats_inc_outucastpkts);

125
static u32 mpls_multipath_hash(struct mpls_route *rt, struct sk_buff *skb)
R
Roopa Prabhu 已提交
126
{
127
	struct mpls_entry_decoded dec;
128
	unsigned int mpls_hdr_len = 0;
129 130 131 132 133
	struct mpls_shim_hdr *hdr;
	bool eli_seen = false;
	int label_index;
	u32 hash = 0;

134
	for (label_index = 0; label_index < MAX_MP_SELECT_LABELS;
135
	     label_index++) {
136 137
		mpls_hdr_len += sizeof(*hdr);
		if (!pskb_may_pull(skb, mpls_hdr_len))
138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161
			break;

		/* Read and decode the current label */
		hdr = mpls_hdr(skb) + label_index;
		dec = mpls_entry_decode(hdr);

		/* RFC6790 - reserved labels MUST NOT be used as keys
		 * for the load-balancing function
		 */
		if (likely(dec.label >= MPLS_LABEL_FIRST_UNRESERVED)) {
			hash = jhash_1word(dec.label, hash);

			/* The entropy label follows the entropy label
			 * indicator, so this means that the entropy
			 * label was just added to the hash - no need to
			 * go any deeper either in the label stack or in the
			 * payload
			 */
			if (eli_seen)
				break;
		} else if (dec.label == MPLS_LABEL_ENTROPY) {
			eli_seen = true;
		}

162 163 164 165 166
		if (!dec.bos)
			continue;

		/* found bottom label; does skb have room for a header? */
		if (pskb_may_pull(skb, mpls_hdr_len + sizeof(struct iphdr))) {
167 168
			const struct iphdr *v4hdr;

169
			v4hdr = (const struct iphdr *)(hdr + 1);
170 171 172 173 174
			if (v4hdr->version == 4) {
				hash = jhash_3words(ntohl(v4hdr->saddr),
						    ntohl(v4hdr->daddr),
						    v4hdr->protocol, hash);
			} else if (v4hdr->version == 6 &&
175 176
				   pskb_may_pull(skb, mpls_hdr_len +
						 sizeof(struct ipv6hdr))) {
177 178
				const struct ipv6hdr *v6hdr;

179
				v6hdr = (const struct ipv6hdr *)(hdr + 1);
180 181 182 183 184
				hash = __ipv6_addr_jhash(&v6hdr->saddr, hash);
				hash = __ipv6_addr_jhash(&v6hdr->daddr, hash);
				hash = jhash_1word(v6hdr->nexthdr, hash);
			}
		}
185 186

		break;
187 188
	}

R
Roopa Prabhu 已提交
189 190 191 192
	return hash;
}

static struct mpls_nh *mpls_select_multipath(struct mpls_route *rt,
193
					     struct sk_buff *skb)
R
Roopa Prabhu 已提交
194 195 196 197 198 199 200 201 202 203 204 205 206 207 208
{
	int alive = ACCESS_ONCE(rt->rt_nhn_alive);
	u32 hash = 0;
	int nh_index = 0;
	int n = 0;

	/* No need to look further into packet if there's only
	 * one path
	 */
	if (rt->rt_nhn == 1)
		goto out;

	if (alive <= 0)
		return NULL;

209
	hash = mpls_multipath_hash(rt, skb);
R
Roopa Prabhu 已提交
210 211 212 213 214 215 216 217 218 219 220
	nh_index = hash % alive;
	if (alive == rt->rt_nhn)
		goto out;
	for_nexthops(rt) {
		if (nh->nh_flags & (RTNH_F_DEAD | RTNH_F_LINKDOWN))
			continue;
		if (n == nh_index)
			return nh;
		n++;
	} endfor_nexthops(rt);

221 222
out:
	return &rt->rt_nh[nh_index];
R
Roopa Prabhu 已提交
223 224
}

225 226
static bool mpls_egress(struct net *net, struct mpls_route *rt,
			struct sk_buff *skb, struct mpls_entry_decoded dec)
E
Eric W. Biederman 已提交
227
{
228 229
	enum mpls_payload_type payload_type;
	bool success = false;
E
Eric W. Biederman 已提交
230

231 232 233 234 235 236 237 238 239 240 241 242 243
	/* The IPv4 code below accesses through the IPv4 header
	 * checksum, which is 12 bytes into the packet.
	 * The IPv6 code below accesses through the IPv6 hop limit
	 * which is 8 bytes into the packet.
	 *
	 * For all supported cases there should always be at least 12
	 * bytes of packet data present.  The IPv4 header is 20 bytes
	 * without options and the IPv6 header is always 40 bytes
	 * long.
	 */
	if (!pskb_may_pull(skb, 12))
		return false;

244 245 246 247 248 249 250
	payload_type = rt->rt_payload_type;
	if (payload_type == MPT_UNSPEC)
		payload_type = ip_hdr(skb)->version;

	switch (payload_type) {
	case MPT_IPV4: {
		struct iphdr *hdr4 = ip_hdr(skb);
251
		u8 new_ttl;
E
Eric W. Biederman 已提交
252
		skb->protocol = htons(ETH_P_IP);
253 254 255 256 257 258 259 260 261 262 263 264

		/* If propagating TTL, take the decremented TTL from
		 * the incoming MPLS header, otherwise decrement the
		 * TTL, but only if not 0 to avoid underflow.
		 */
		if (rt->rt_ttl_propagate == MPLS_TTL_PROP_ENABLED ||
		    (rt->rt_ttl_propagate == MPLS_TTL_PROP_DEFAULT &&
		     net->mpls.ip_ttl_propagate))
			new_ttl = dec.ttl;
		else
			new_ttl = hdr4->ttl ? hdr4->ttl - 1 : 0;

E
Eric W. Biederman 已提交
265 266
		csum_replace2(&hdr4->check,
			      htons(hdr4->ttl << 8),
267 268
			      htons(new_ttl << 8));
		hdr4->ttl = new_ttl;
269 270
		success = true;
		break;
E
Eric W. Biederman 已提交
271
	}
272
	case MPT_IPV6: {
E
Eric W. Biederman 已提交
273 274
		struct ipv6hdr *hdr6 = ipv6_hdr(skb);
		skb->protocol = htons(ETH_P_IPV6);
275 276 277 278 279 280 281 282 283 284 285

		/* If propagating TTL, take the decremented TTL from
		 * the incoming MPLS header, otherwise decrement the
		 * hop limit, but only if not 0 to avoid underflow.
		 */
		if (rt->rt_ttl_propagate == MPLS_TTL_PROP_ENABLED ||
		    (rt->rt_ttl_propagate == MPLS_TTL_PROP_DEFAULT &&
		     net->mpls.ip_ttl_propagate))
			hdr6->hop_limit = dec.ttl;
		else if (hdr6->hop_limit)
			hdr6->hop_limit = hdr6->hop_limit - 1;
286 287
		success = true;
		break;
E
Eric W. Biederman 已提交
288
	}
289
	case MPT_UNSPEC:
290
		/* Should have decided which protocol it is by now */
291 292 293
		break;
	}

E
Eric W. Biederman 已提交
294 295 296 297 298 299 300 301 302
	return success;
}

static int mpls_forward(struct sk_buff *skb, struct net_device *dev,
			struct packet_type *pt, struct net_device *orig_dev)
{
	struct net *net = dev_net(dev);
	struct mpls_shim_hdr *hdr;
	struct mpls_route *rt;
R
Roopa Prabhu 已提交
303
	struct mpls_nh *nh;
E
Eric W. Biederman 已提交
304 305
	struct mpls_entry_decoded dec;
	struct net_device *out_dev;
R
Robert Shearman 已提交
306
	struct mpls_dev *out_mdev;
R
Robert Shearman 已提交
307
	struct mpls_dev *mdev;
E
Eric W. Biederman 已提交
308 309 310 311 312 313 314
	unsigned int hh_len;
	unsigned int new_header_size;
	unsigned int mtu;
	int err;

	/* Careful this entire function runs inside of an rcu critical section */

R
Robert Shearman 已提交
315
	mdev = mpls_dev_get(dev);
R
Robert Shearman 已提交
316
	if (!mdev)
R
Robert Shearman 已提交
317 318
		goto drop;

R
Robert Shearman 已提交
319 320 321 322 323
	MPLS_INC_STATS_LEN(mdev, skb->len, rx_packets,
			   rx_bytes);

	if (!mdev->input_enabled) {
		MPLS_INC_STATS(mdev, rx_dropped);
E
Eric W. Biederman 已提交
324
		goto drop;
R
Robert Shearman 已提交
325 326 327 328
	}

	if (skb->pkt_type != PACKET_HOST)
		goto err;
E
Eric W. Biederman 已提交
329 330

	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL)
R
Robert Shearman 已提交
331
		goto err;
E
Eric W. Biederman 已提交
332 333

	if (!pskb_may_pull(skb, sizeof(*hdr)))
R
Robert Shearman 已提交
334
		goto err;
E
Eric W. Biederman 已提交
335 336 337 338 339 340

	/* Read and decode the label */
	hdr = mpls_hdr(skb);
	dec = mpls_entry_decode(hdr);

	rt = mpls_route_input_rcu(net, dec.label);
R
Robert Shearman 已提交
341 342
	if (!rt) {
		MPLS_INC_STATS(mdev, rx_noroute);
E
Eric W. Biederman 已提交
343
		goto drop;
R
Robert Shearman 已提交
344
	}
E
Eric W. Biederman 已提交
345

346
	nh = mpls_select_multipath(rt, skb);
R
Roopa Prabhu 已提交
347
	if (!nh)
R
Robert Shearman 已提交
348
		goto err;
E
Eric W. Biederman 已提交
349

350 351 352 353 354 355
	/* Pop the label */
	skb_pull(skb, sizeof(*hdr));
	skb_reset_network_header(skb);

	skb_orphan(skb);

E
Eric W. Biederman 已提交
356
	if (skb_warn_if_lro(skb))
R
Robert Shearman 已提交
357
		goto err;
E
Eric W. Biederman 已提交
358 359 360 361

	skb_forward_csum(skb);

	/* Verify ttl is valid */
362
	if (dec.ttl <= 1)
R
Robert Shearman 已提交
363
		goto err;
E
Eric W. Biederman 已提交
364 365
	dec.ttl -= 1;

R
Robert Shearman 已提交
366 367 368 369 370
	/* Find the output device */
	out_dev = rcu_dereference(nh->nh_dev);
	if (!mpls_output_possible(out_dev))
		goto tx_err;

E
Eric W. Biederman 已提交
371
	/* Verify the destination can hold the packet */
R
Roopa Prabhu 已提交
372
	new_header_size = mpls_nh_header_size(nh);
E
Eric W. Biederman 已提交
373 374
	mtu = mpls_dev_mtu(out_dev);
	if (mpls_pkt_too_big(skb, mtu - new_header_size))
R
Robert Shearman 已提交
375
		goto tx_err;
E
Eric W. Biederman 已提交
376 377 378 379 380 381 382

	hh_len = LL_RESERVED_SPACE(out_dev);
	if (!out_dev->header_ops)
		hh_len = 0;

	/* Ensure there is enough space for the headers in the skb */
	if (skb_cow(skb, hh_len + new_header_size))
R
Robert Shearman 已提交
383
		goto tx_err;
E
Eric W. Biederman 已提交
384 385 386 387 388 389

	skb->dev = out_dev;
	skb->protocol = htons(ETH_P_MPLS_UC);

	if (unlikely(!new_header_size && dec.bos)) {
		/* Penultimate hop popping */
390
		if (!mpls_egress(dev_net(out_dev), rt, skb, dec))
R
Robert Shearman 已提交
391
			goto err;
E
Eric W. Biederman 已提交
392 393 394 395 396 397 398 399
	} else {
		bool bos;
		int i;
		skb_push(skb, new_header_size);
		skb_reset_network_header(skb);
		/* Push the new labels */
		hdr = mpls_hdr(skb);
		bos = dec.bos;
R
Roopa Prabhu 已提交
400 401 402
		for (i = nh->nh_labels - 1; i >= 0; i--) {
			hdr[i] = mpls_entry_encode(nh->nh_label[i],
						   dec.ttl, 0, bos);
E
Eric W. Biederman 已提交
403 404 405 406
			bos = false;
		}
	}

R
Robert Shearman 已提交
407 408
	mpls_stats_inc_outucastpkts(out_dev, skb);

409 410 411 412 413 414 415
	/* If via wasn't specified then send out using device address */
	if (nh->nh_via_table == MPLS_NEIGH_TABLE_UNSPEC)
		err = neigh_xmit(NEIGH_LINK_TABLE, out_dev,
				 out_dev->dev_addr, skb);
	else
		err = neigh_xmit(nh->nh_via_table, out_dev,
				 mpls_nh_via(rt, nh), skb);
E
Eric W. Biederman 已提交
416 417 418 419 420
	if (err)
		net_dbg_ratelimited("%s: packet transmission failed: %d\n",
				    __func__, err);
	return 0;

R
Robert Shearman 已提交
421 422 423 424 425 426 427
tx_err:
	out_mdev = out_dev ? mpls_dev_get(out_dev) : NULL;
	if (out_mdev)
		MPLS_INC_STATS(out_mdev, tx_errors);
	goto drop;
err:
	MPLS_INC_STATS(mdev, rx_errors);
E
Eric W. Biederman 已提交
428 429 430 431 432 433 434 435 436 437
drop:
	kfree_skb(skb);
	return NET_RX_DROP;
}

static struct packet_type mpls_packet_type __read_mostly = {
	.type = cpu_to_be16(ETH_P_MPLS_UC),
	.func = mpls_forward,
};

438
static const struct nla_policy rtm_mpls_policy[RTA_MAX+1] = {
439 440
	[RTA_DST]		= { .type = NLA_U32 },
	[RTA_OIF]		= { .type = NLA_U32 },
441
	[RTA_TTL_PROPAGATE]	= { .type = NLA_U8 },
442 443
};

444
struct mpls_route_config {
445 446
	u32			rc_protocol;
	u32			rc_ifindex;
R
Roopa Prabhu 已提交
447 448
	u8			rc_via_table;
	u8			rc_via_alen;
449 450
	u8			rc_via[MAX_VIA_ALEN];
	u32			rc_label;
451
	u8			rc_ttl_propagate;
R
Roopa Prabhu 已提交
452
	u8			rc_output_labels;
453 454 455 456
	u32			rc_output_label[MAX_NEW_LABELS];
	u32			rc_nlflags;
	enum mpls_payload_type	rc_payload_type;
	struct nl_info		rc_nlinfo;
R
Roopa Prabhu 已提交
457 458
	struct rtnexthop	*rc_mp;
	int			rc_mp_len;
459 460
};

461
static struct mpls_route *mpls_rt_alloc(int num_nh, u8 max_alen)
E
Eric W. Biederman 已提交
462
{
463
	u8 max_alen_aligned = ALIGN(max_alen, VIA_ALEN_ALIGN);
E
Eric W. Biederman 已提交
464 465
	struct mpls_route *rt;

466 467 468
	rt = kzalloc(ALIGN(sizeof(*rt) + num_nh * sizeof(*rt->rt_nh),
			   VIA_ALEN_ALIGN) +
		     num_nh * max_alen_aligned,
R
Roopa Prabhu 已提交
469
		     GFP_KERNEL);
470
	if (rt) {
R
Roopa Prabhu 已提交
471
		rt->rt_nhn = num_nh;
R
Roopa Prabhu 已提交
472
		rt->rt_nhn_alive = num_nh;
473 474
		rt->rt_max_alen = max_alen_aligned;
	}
R
Roopa Prabhu 已提交
475

E
Eric W. Biederman 已提交
476 477 478 479 480 481 482 483 484
	return rt;
}

static void mpls_rt_free(struct mpls_route *rt)
{
	if (rt)
		kfree_rcu(rt, rt_rcu);
}

485 486 487 488 489 490 491 492 493 494
static void mpls_notify_route(struct net *net, unsigned index,
			      struct mpls_route *old, struct mpls_route *new,
			      const struct nl_info *info)
{
	struct nlmsghdr *nlh = info ? info->nlh : NULL;
	unsigned portid = info ? info->portid : 0;
	int event = new ? RTM_NEWROUTE : RTM_DELROUTE;
	struct mpls_route *rt = new ? new : old;
	unsigned nlm_flags = (old && new) ? NLM_F_REPLACE : 0;
	/* Ignore reserved labels for now */
495
	if (rt && (index >= MPLS_LABEL_FIRST_UNRESERVED))
496 497 498
		rtmsg_lfib(event, index, rt, nlh, net, portid, nlm_flags);
}

E
Eric W. Biederman 已提交
499
static void mpls_route_update(struct net *net, unsigned index,
R
Roopa Prabhu 已提交
500
			      struct mpls_route *new,
E
Eric W. Biederman 已提交
501 502
			      const struct nl_info *info)
{
503
	struct mpls_route __rcu **platform_label;
R
Roopa Prabhu 已提交
504
	struct mpls_route *rt;
E
Eric W. Biederman 已提交
505 506 507

	ASSERT_RTNL();

508 509
	platform_label = rtnl_dereference(net->mpls.platform_label);
	rt = rtnl_dereference(platform_label[index]);
R
Roopa Prabhu 已提交
510
	rcu_assign_pointer(platform_label[index], new);
E
Eric W. Biederman 已提交
511

R
Roopa Prabhu 已提交
512
	mpls_notify_route(net, index, rt, new, info);
513

E
Eric W. Biederman 已提交
514
	/* If we removed a route free it now */
R
Roopa Prabhu 已提交
515
	mpls_rt_free(rt);
E
Eric W. Biederman 已提交
516 517
}

518 519
static unsigned find_free_label(struct net *net)
{
520 521
	struct mpls_route __rcu **platform_label;
	size_t platform_labels;
522
	unsigned index;
523 524 525

	platform_label = rtnl_dereference(net->mpls.platform_label);
	platform_labels = net->mpls.platform_labels;
526 527
	for (index = MPLS_LABEL_FIRST_UNRESERVED; index < platform_labels;
	     index++) {
528
		if (!rtnl_dereference(platform_label[index]))
529 530 531 532 533
			return index;
	}
	return LABEL_NOT_SPECIFIED;
}

534
#if IS_ENABLED(CONFIG_INET)
535 536
static struct net_device *inet_fib_lookup_dev(struct net *net,
					      const void *addr)
R
Roopa Prabhu 已提交
537
{
538
	struct net_device *dev;
R
Roopa Prabhu 已提交
539 540 541 542 543 544
	struct rtable *rt;
	struct in_addr daddr;

	memcpy(&daddr, addr, sizeof(struct in_addr));
	rt = ip_route_output(net, daddr.s_addr, 0, 0, 0);
	if (IS_ERR(rt))
545
		return ERR_CAST(rt);
R
Roopa Prabhu 已提交
546 547 548 549 550 551 552

	dev = rt->dst.dev;
	dev_hold(dev);

	ip_rt_put(rt);

	return dev;
553 554
}
#else
555 556
static struct net_device *inet_fib_lookup_dev(struct net *net,
					      const void *addr)
557 558
{
	return ERR_PTR(-EAFNOSUPPORT);
R
Roopa Prabhu 已提交
559
}
560
#endif
R
Roopa Prabhu 已提交
561

562
#if IS_ENABLED(CONFIG_IPV6)
563 564
static struct net_device *inet6_fib_lookup_dev(struct net *net,
					       const void *addr)
R
Roopa Prabhu 已提交
565
{
566
	struct net_device *dev;
R
Roopa Prabhu 已提交
567 568
	struct dst_entry *dst;
	struct flowi6 fl6;
569 570 571 572
	int err;

	if (!ipv6_stub)
		return ERR_PTR(-EAFNOSUPPORT);
R
Roopa Prabhu 已提交
573 574 575

	memset(&fl6, 0, sizeof(fl6));
	memcpy(&fl6.daddr, addr, sizeof(struct in6_addr));
576 577
	err = ipv6_stub->ipv6_dst_lookup(net, NULL, &dst, &fl6);
	if (err)
578
		return ERR_PTR(err);
R
Roopa Prabhu 已提交
579 580 581 582 583 584 585

	dev = dst->dev;
	dev_hold(dev);
	dst_release(dst);

	return dev;
}
586
#else
587 588
static struct net_device *inet6_fib_lookup_dev(struct net *net,
					       const void *addr)
589 590 591 592
{
	return ERR_PTR(-EAFNOSUPPORT);
}
#endif
R
Roopa Prabhu 已提交
593 594

static struct net_device *find_outdev(struct net *net,
595
				      struct mpls_route *rt,
R
Roopa Prabhu 已提交
596
				      struct mpls_nh *nh, int oif)
R
Roopa Prabhu 已提交
597 598 599
{
	struct net_device *dev = NULL;

R
Roopa Prabhu 已提交
600 601
	if (!oif) {
		switch (nh->nh_via_table) {
R
Roopa Prabhu 已提交
602
		case NEIGH_ARP_TABLE:
603
			dev = inet_fib_lookup_dev(net, mpls_nh_via(rt, nh));
R
Roopa Prabhu 已提交
604 605
			break;
		case NEIGH_ND_TABLE:
606
			dev = inet6_fib_lookup_dev(net, mpls_nh_via(rt, nh));
R
Roopa Prabhu 已提交
607 608 609 610 611
			break;
		case NEIGH_LINK_TABLE:
			break;
		}
	} else {
R
Roopa Prabhu 已提交
612
		dev = dev_get_by_index(net, oif);
R
Roopa Prabhu 已提交
613 614
	}

615 616 617
	if (!dev)
		return ERR_PTR(-ENODEV);

618 619 620
	if (IS_ERR(dev))
		return dev;

R
Roopa Prabhu 已提交
621 622 623
	/* The caller is holding rtnl anyways, so release the dev reference */
	dev_put(dev);

R
Roopa Prabhu 已提交
624 625 626
	return dev;
}

627 628
static int mpls_nh_assign_dev(struct net *net, struct mpls_route *rt,
			      struct mpls_nh *nh, int oif)
R
Roopa Prabhu 已提交
629 630 631 632
{
	struct net_device *dev = NULL;
	int err = -ENODEV;

633
	dev = find_outdev(net, rt, nh, oif);
R
Roopa Prabhu 已提交
634 635 636 637 638 639 640 641 642 643 644
	if (IS_ERR(dev)) {
		err = PTR_ERR(dev);
		dev = NULL;
		goto errout;
	}

	/* Ensure this is a supported device */
	err = -EINVAL;
	if (!mpls_dev_get(dev))
		goto errout;

645 646 647 648
	if ((nh->nh_via_table == NEIGH_LINK_TABLE) &&
	    (dev->addr_len != nh->nh_via_alen))
		goto errout;

R
Roopa Prabhu 已提交
649 650
	RCU_INIT_POINTER(nh->nh_dev, dev);

R
Roopa Prabhu 已提交
651 652 653 654 655 656 657 658 659 660
	if (!(dev->flags & IFF_UP)) {
		nh->nh_flags |= RTNH_F_DEAD;
	} else {
		unsigned int flags;

		flags = dev_get_flags(dev);
		if (!(flags & (IFF_RUNNING | IFF_LOWER_UP)))
			nh->nh_flags |= RTNH_F_LINKDOWN;
	}

R
Roopa Prabhu 已提交
661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687
	return 0;

errout:
	return err;
}

static int mpls_nh_build_from_cfg(struct mpls_route_config *cfg,
				  struct mpls_route *rt)
{
	struct net *net = cfg->rc_nlinfo.nl_net;
	struct mpls_nh *nh = rt->rt_nh;
	int err;
	int i;

	if (!nh)
		return -ENOMEM;

	err = -EINVAL;
	/* Ensure only a supported number of labels are present */
	if (cfg->rc_output_labels > MAX_NEW_LABELS)
		goto errout;

	nh->nh_labels = cfg->rc_output_labels;
	for (i = 0; i < nh->nh_labels; i++)
		nh->nh_label[i] = cfg->rc_output_label[i];

	nh->nh_via_table = cfg->rc_via_table;
688
	memcpy(__mpls_nh_via(rt, nh), cfg->rc_via, cfg->rc_via_alen);
R
Roopa Prabhu 已提交
689 690
	nh->nh_via_alen = cfg->rc_via_alen;

691
	err = mpls_nh_assign_dev(net, rt, nh, cfg->rc_ifindex);
R
Roopa Prabhu 已提交
692 693 694
	if (err)
		goto errout;

R
Roopa Prabhu 已提交
695 696 697
	if (nh->nh_flags & (RTNH_F_DEAD | RTNH_F_LINKDOWN))
		rt->rt_nhn_alive--;

R
Roopa Prabhu 已提交
698 699 700 701 702 703
	return 0;

errout:
	return err;
}

704
static int mpls_nh_build(struct net *net, struct mpls_route *rt,
R
Roopa Prabhu 已提交
705 706
			 struct mpls_nh *nh, int oif, struct nlattr *via,
			 struct nlattr *newdst)
R
Roopa Prabhu 已提交
707 708 709 710 711 712 713 714 715 716 717 718 719
{
	int err = -ENOMEM;

	if (!nh)
		goto errout;

	if (newdst) {
		err = nla_get_labels(newdst, MAX_NEW_LABELS,
				     &nh->nh_labels, nh->nh_label);
		if (err)
			goto errout;
	}

720 721 722 723 724 725 726 727
	if (via) {
		err = nla_get_via(via, &nh->nh_via_alen, &nh->nh_via_table,
				  __mpls_nh_via(rt, nh));
		if (err)
			goto errout;
	} else {
		nh->nh_via_table = MPLS_NEIGH_TABLE_UNSPEC;
	}
R
Roopa Prabhu 已提交
728

729
	err = mpls_nh_assign_dev(net, rt, nh, oif);
R
Roopa Prabhu 已提交
730 731 732 733 734 735 736 737 738
	if (err)
		goto errout;

	return 0;

errout:
	return err;
}

739 740
static int mpls_count_nexthops(struct rtnexthop *rtnh, int len,
			       u8 cfg_via_alen, u8 *max_via_alen)
R
Roopa Prabhu 已提交
741 742 743 744
{
	int nhs = 0;
	int remaining = len;

745 746 747 748 749 750 751
	if (!rtnh) {
		*max_via_alen = cfg_via_alen;
		return 1;
	}

	*max_via_alen = 0;

R
Roopa Prabhu 已提交
752
	while (rtnh_ok(rtnh, remaining)) {
753 754 755 756 757 758 759 760 761 762 763 764 765 766 767
		struct nlattr *nla, *attrs = rtnh_attrs(rtnh);
		int attrlen;

		attrlen = rtnh_attrlen(rtnh);
		nla = nla_find(attrs, attrlen, RTA_VIA);
		if (nla && nla_len(nla) >=
		    offsetof(struct rtvia, rtvia_addr)) {
			int via_alen = nla_len(nla) -
				offsetof(struct rtvia, rtvia_addr);

			if (via_alen <= MAX_VIA_ALEN)
				*max_via_alen = max_t(u16, *max_via_alen,
						      via_alen);
		}

R
Roopa Prabhu 已提交
768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794
		nhs++;
		rtnh = rtnh_next(rtnh, &remaining);
	}

	/* leftover implies invalid nexthop configuration, discard it */
	return remaining > 0 ? 0 : nhs;
}

static int mpls_nh_build_multi(struct mpls_route_config *cfg,
			       struct mpls_route *rt)
{
	struct rtnexthop *rtnh = cfg->rc_mp;
	struct nlattr *nla_via, *nla_newdst;
	int remaining = cfg->rc_mp_len;
	int nhs = 0;
	int err = 0;

	change_nexthops(rt) {
		int attrlen;

		nla_via = NULL;
		nla_newdst = NULL;

		err = -EINVAL;
		if (!rtnh_ok(rtnh, remaining))
			goto errout;

795 796 797 798 799 800
		/* neither weighted multipath nor any flags
		 * are supported
		 */
		if (rtnh->rtnh_hops || rtnh->rtnh_flags)
			goto errout;

R
Roopa Prabhu 已提交
801 802 803 804 805 806 807 808
		attrlen = rtnh_attrlen(rtnh);
		if (attrlen > 0) {
			struct nlattr *attrs = rtnh_attrs(rtnh);

			nla_via = nla_find(attrs, attrlen, RTA_VIA);
			nla_newdst = nla_find(attrs, attrlen, RTA_NEWDST);
		}

809
		err = mpls_nh_build(cfg->rc_nlinfo.nl_net, rt, nh,
R
Roopa Prabhu 已提交
810
				    rtnh->rtnh_ifindex, nla_via, nla_newdst);
R
Roopa Prabhu 已提交
811 812 813
		if (err)
			goto errout;

R
Roopa Prabhu 已提交
814 815 816
		if (nh->nh_flags & (RTNH_F_DEAD | RTNH_F_LINKDOWN))
			rt->rt_nhn_alive--;

R
Roopa Prabhu 已提交
817 818 819 820 821 822 823 824 825 826 827 828
		rtnh = rtnh_next(rtnh, &remaining);
		nhs++;
	} endfor_nexthops(rt);

	rt->rt_nhn = nhs;

	return 0;

errout:
	return err;
}

829 830
static int mpls_route_add(struct mpls_route_config *cfg)
{
831
	struct mpls_route __rcu **platform_label;
832 833 834
	struct net *net = cfg->rc_nlinfo.nl_net;
	struct mpls_route *rt, *old;
	int err = -EINVAL;
835
	u8 max_via_alen;
R
Roopa Prabhu 已提交
836
	unsigned index;
837
	int nhs;
838 839 840 841 842 843 844 845 846

	index = cfg->rc_label;

	/* If a label was not specified during insert pick one */
	if ((index == LABEL_NOT_SPECIFIED) &&
	    (cfg->rc_nlflags & NLM_F_CREATE)) {
		index = find_free_label(net);
	}

847 848
	/* Reserved labels may not be set */
	if (index < MPLS_LABEL_FIRST_UNRESERVED)
849 850 851 852 853 854 855
		goto errout;

	/* The full 20 bit range may not be supported. */
	if (index >= net->mpls.platform_labels)
		goto errout;

	/* Append makes no sense with mpls */
856
	err = -EOPNOTSUPP;
857 858 859 860
	if (cfg->rc_nlflags & NLM_F_APPEND)
		goto errout;

	err = -EEXIST;
861 862
	platform_label = rtnl_dereference(net->mpls.platform_label);
	old = rtnl_dereference(platform_label[index]);
863 864 865 866 867 868 869 870 871 872 873
	if ((cfg->rc_nlflags & NLM_F_EXCL) && old)
		goto errout;

	err = -EEXIST;
	if (!(cfg->rc_nlflags & NLM_F_REPLACE) && old)
		goto errout;

	err = -ENOENT;
	if (!(cfg->rc_nlflags & NLM_F_CREATE) && !old)
		goto errout;

874 875 876 877 878
	err = -EINVAL;
	nhs = mpls_count_nexthops(cfg->rc_mp, cfg->rc_mp_len,
				  cfg->rc_via_alen, &max_via_alen);
	if (nhs == 0)
		goto errout;
R
Roopa Prabhu 已提交
879

880
	err = -ENOMEM;
881
	rt = mpls_rt_alloc(nhs, max_via_alen);
882 883 884 885
	if (!rt)
		goto errout;

	rt->rt_protocol = cfg->rc_protocol;
886
	rt->rt_payload_type = cfg->rc_payload_type;
887
	rt->rt_ttl_propagate = cfg->rc_ttl_propagate;
888

R
Roopa Prabhu 已提交
889 890 891 892 893 894 895 896
	if (cfg->rc_mp)
		err = mpls_nh_build_multi(cfg, rt);
	else
		err = mpls_nh_build_from_cfg(cfg, rt);
	if (err)
		goto freert;

	mpls_route_update(net, index, rt, &cfg->rc_nlinfo);
897 898 899

	return 0;

R
Roopa Prabhu 已提交
900 901
freert:
	mpls_rt_free(rt);
902 903 904 905 906 907 908 909 910 911 912 913
errout:
	return err;
}

static int mpls_route_del(struct mpls_route_config *cfg)
{
	struct net *net = cfg->rc_nlinfo.nl_net;
	unsigned index;
	int err = -EINVAL;

	index = cfg->rc_label;

914 915
	/* Reserved labels may not be removed */
	if (index < MPLS_LABEL_FIRST_UNRESERVED)
916 917 918 919 920 921
		goto errout;

	/* The full 20 bit range may not be supported */
	if (index >= net->mpls.platform_labels)
		goto errout;

R
Roopa Prabhu 已提交
922
	mpls_route_update(net, index, NULL, &cfg->rc_nlinfo);
923 924 925 926 927 928

	err = 0;
errout:
	return err;
}

R
Robert Shearman 已提交
929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992
static void mpls_get_stats(struct mpls_dev *mdev,
			   struct mpls_link_stats *stats)
{
	struct mpls_pcpu_stats *p;
	int i;

	memset(stats, 0, sizeof(*stats));

	for_each_possible_cpu(i) {
		struct mpls_link_stats local;
		unsigned int start;

		p = per_cpu_ptr(mdev->stats, i);
		do {
			start = u64_stats_fetch_begin(&p->syncp);
			local = p->stats;
		} while (u64_stats_fetch_retry(&p->syncp, start));

		stats->rx_packets	+= local.rx_packets;
		stats->rx_bytes		+= local.rx_bytes;
		stats->tx_packets	+= local.tx_packets;
		stats->tx_bytes		+= local.tx_bytes;
		stats->rx_errors	+= local.rx_errors;
		stats->tx_errors	+= local.tx_errors;
		stats->rx_dropped	+= local.rx_dropped;
		stats->tx_dropped	+= local.tx_dropped;
		stats->rx_noroute	+= local.rx_noroute;
	}
}

static int mpls_fill_stats_af(struct sk_buff *skb,
			      const struct net_device *dev)
{
	struct mpls_link_stats *stats;
	struct mpls_dev *mdev;
	struct nlattr *nla;

	mdev = mpls_dev_get(dev);
	if (!mdev)
		return -ENODATA;

	nla = nla_reserve_64bit(skb, MPLS_STATS_LINK,
				sizeof(struct mpls_link_stats),
				MPLS_STATS_UNSPEC);
	if (!nla)
		return -EMSGSIZE;

	stats = nla_data(nla);
	mpls_get_stats(mdev, stats);

	return 0;
}

static size_t mpls_get_stats_af_size(const struct net_device *dev)
{
	struct mpls_dev *mdev;

	mdev = mpls_dev_get(dev);
	if (!mdev)
		return 0;

	return nla_total_size_64bit(sizeof(struct mpls_link_stats));
}

D
David Ahern 已提交
993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168
static int mpls_netconf_fill_devconf(struct sk_buff *skb, struct mpls_dev *mdev,
				     u32 portid, u32 seq, int event,
				     unsigned int flags, int type)
{
	struct nlmsghdr  *nlh;
	struct netconfmsg *ncm;
	bool all = false;

	nlh = nlmsg_put(skb, portid, seq, event, sizeof(struct netconfmsg),
			flags);
	if (!nlh)
		return -EMSGSIZE;

	if (type == NETCONFA_ALL)
		all = true;

	ncm = nlmsg_data(nlh);
	ncm->ncm_family = AF_MPLS;

	if (nla_put_s32(skb, NETCONFA_IFINDEX, mdev->dev->ifindex) < 0)
		goto nla_put_failure;

	if ((all || type == NETCONFA_INPUT) &&
	    nla_put_s32(skb, NETCONFA_INPUT,
			mdev->input_enabled) < 0)
		goto nla_put_failure;

	nlmsg_end(skb, nlh);
	return 0;

nla_put_failure:
	nlmsg_cancel(skb, nlh);
	return -EMSGSIZE;
}

static int mpls_netconf_msgsize_devconf(int type)
{
	int size = NLMSG_ALIGN(sizeof(struct netconfmsg))
			+ nla_total_size(4); /* NETCONFA_IFINDEX */
	bool all = false;

	if (type == NETCONFA_ALL)
		all = true;

	if (all || type == NETCONFA_INPUT)
		size += nla_total_size(4);

	return size;
}

static void mpls_netconf_notify_devconf(struct net *net, int type,
					struct mpls_dev *mdev)
{
	struct sk_buff *skb;
	int err = -ENOBUFS;

	skb = nlmsg_new(mpls_netconf_msgsize_devconf(type), GFP_KERNEL);
	if (!skb)
		goto errout;

	err = mpls_netconf_fill_devconf(skb, mdev, 0, 0, RTM_NEWNETCONF,
					0, type);
	if (err < 0) {
		/* -EMSGSIZE implies BUG in mpls_netconf_msgsize_devconf() */
		WARN_ON(err == -EMSGSIZE);
		kfree_skb(skb);
		goto errout;
	}

	rtnl_notify(skb, net, 0, RTNLGRP_MPLS_NETCONF, NULL, GFP_KERNEL);
	return;
errout:
	if (err < 0)
		rtnl_set_sk_err(net, RTNLGRP_MPLS_NETCONF, err);
}

static const struct nla_policy devconf_mpls_policy[NETCONFA_MAX + 1] = {
	[NETCONFA_IFINDEX]	= { .len = sizeof(int) },
};

static int mpls_netconf_get_devconf(struct sk_buff *in_skb,
				    struct nlmsghdr *nlh)
{
	struct net *net = sock_net(in_skb->sk);
	struct nlattr *tb[NETCONFA_MAX + 1];
	struct netconfmsg *ncm;
	struct net_device *dev;
	struct mpls_dev *mdev;
	struct sk_buff *skb;
	int ifindex;
	int err;

	err = nlmsg_parse(nlh, sizeof(*ncm), tb, NETCONFA_MAX,
			  devconf_mpls_policy);
	if (err < 0)
		goto errout;

	err = -EINVAL;
	if (!tb[NETCONFA_IFINDEX])
		goto errout;

	ifindex = nla_get_s32(tb[NETCONFA_IFINDEX]);
	dev = __dev_get_by_index(net, ifindex);
	if (!dev)
		goto errout;

	mdev = mpls_dev_get(dev);
	if (!mdev)
		goto errout;

	err = -ENOBUFS;
	skb = nlmsg_new(mpls_netconf_msgsize_devconf(NETCONFA_ALL), GFP_KERNEL);
	if (!skb)
		goto errout;

	err = mpls_netconf_fill_devconf(skb, mdev,
					NETLINK_CB(in_skb).portid,
					nlh->nlmsg_seq, RTM_NEWNETCONF, 0,
					NETCONFA_ALL);
	if (err < 0) {
		/* -EMSGSIZE implies BUG in mpls_netconf_msgsize_devconf() */
		WARN_ON(err == -EMSGSIZE);
		kfree_skb(skb);
		goto errout;
	}
	err = rtnl_unicast(skb, net, NETLINK_CB(in_skb).portid);
errout:
	return err;
}

static int mpls_netconf_dump_devconf(struct sk_buff *skb,
				     struct netlink_callback *cb)
{
	struct net *net = sock_net(skb->sk);
	struct hlist_head *head;
	struct net_device *dev;
	struct mpls_dev *mdev;
	int idx, s_idx;
	int h, s_h;

	s_h = cb->args[0];
	s_idx = idx = cb->args[1];

	for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) {
		idx = 0;
		head = &net->dev_index_head[h];
		rcu_read_lock();
		cb->seq = net->dev_base_seq;
		hlist_for_each_entry_rcu(dev, head, index_hlist) {
			if (idx < s_idx)
				goto cont;
			mdev = mpls_dev_get(dev);
			if (!mdev)
				goto cont;
			if (mpls_netconf_fill_devconf(skb, mdev,
						      NETLINK_CB(cb->skb).portid,
						      cb->nlh->nlmsg_seq,
						      RTM_NEWNETCONF,
						      NLM_F_MULTI,
						      NETCONFA_ALL) < 0) {
				rcu_read_unlock();
				goto done;
			}
			nl_dump_check_consistent(cb, nlmsg_hdr(skb));
cont:
			idx++;
		}
		rcu_read_unlock();
	}
done:
	cb->args[0] = h;
	cb->args[1] = idx;

	return skb->len;
}

1169 1170 1171
#define MPLS_PERDEV_SYSCTL_OFFSET(field)	\
	(&((struct mpls_dev *)0)->field)

D
David Ahern 已提交
1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195
static int mpls_conf_proc(struct ctl_table *ctl, int write,
			  void __user *buffer,
			  size_t *lenp, loff_t *ppos)
{
	int oval = *(int *)ctl->data;
	int ret = proc_dointvec(ctl, write, buffer, lenp, ppos);

	if (write) {
		struct mpls_dev *mdev = ctl->extra1;
		int i = (int *)ctl->data - (int *)mdev;
		struct net *net = ctl->extra2;
		int val = *(int *)ctl->data;

		if (i == offsetof(struct mpls_dev, input_enabled) &&
		    val != oval) {
			mpls_netconf_notify_devconf(net,
						    NETCONFA_INPUT,
						    mdev);
		}
	}

	return ret;
}

1196 1197 1198 1199 1200
static const struct ctl_table mpls_dev_table[] = {
	{
		.procname	= "input",
		.maxlen		= sizeof(int),
		.mode		= 0644,
D
David Ahern 已提交
1201
		.proc_handler	= mpls_conf_proc,
1202 1203 1204 1205 1206 1207 1208 1209 1210
		.data		= MPLS_PERDEV_SYSCTL_OFFSET(input_enabled),
	},
	{ }
};

static int mpls_dev_sysctl_register(struct net_device *dev,
				    struct mpls_dev *mdev)
{
	char path[sizeof("net/mpls/conf/") + IFNAMSIZ];
D
David Ahern 已提交
1211
	struct net *net = dev_net(dev);
1212 1213 1214 1215 1216 1217 1218 1219 1220 1221
	struct ctl_table *table;
	int i;

	table = kmemdup(&mpls_dev_table, sizeof(mpls_dev_table), GFP_KERNEL);
	if (!table)
		goto out;

	/* Table data contains only offsets relative to the base of
	 * the mdev at this point, so make them absolute.
	 */
D
David Ahern 已提交
1222
	for (i = 0; i < ARRAY_SIZE(mpls_dev_table); i++) {
1223
		table[i].data = (char *)mdev + (uintptr_t)table[i].data;
D
David Ahern 已提交
1224 1225 1226
		table[i].extra1 = mdev;
		table[i].extra2 = net;
	}
1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250

	snprintf(path, sizeof(path), "net/mpls/conf/%s", dev->name);

	mdev->sysctl = register_net_sysctl(dev_net(dev), path, table);
	if (!mdev->sysctl)
		goto free;

	return 0;

free:
	kfree(table);
out:
	return -ENOBUFS;
}

static void mpls_dev_sysctl_unregister(struct mpls_dev *mdev)
{
	struct ctl_table *table;

	table = mdev->sysctl->ctl_table_arg;
	unregister_net_sysctl_table(mdev->sysctl);
	kfree(table);
}

R
Robert Shearman 已提交
1251 1252 1253 1254
static struct mpls_dev *mpls_add_dev(struct net_device *dev)
{
	struct mpls_dev *mdev;
	int err = -ENOMEM;
R
Robert Shearman 已提交
1255
	int i;
R
Robert Shearman 已提交
1256 1257 1258 1259 1260 1261 1262

	ASSERT_RTNL();

	mdev = kzalloc(sizeof(*mdev), GFP_KERNEL);
	if (!mdev)
		return ERR_PTR(err);

R
Robert Shearman 已提交
1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273
	mdev->stats = alloc_percpu(struct mpls_pcpu_stats);
	if (!mdev->stats)
		goto free;

	for_each_possible_cpu(i) {
		struct mpls_pcpu_stats *mpls_stats;

		mpls_stats = per_cpu_ptr(mdev->stats, i);
		u64_stats_init(&mpls_stats->syncp);
	}

1274 1275 1276 1277
	err = mpls_dev_sysctl_register(dev, mdev);
	if (err)
		goto free;

D
David Ahern 已提交
1278
	mdev->dev = dev;
R
Robert Shearman 已提交
1279 1280 1281
	rcu_assign_pointer(dev->mpls_ptr, mdev);

	return mdev;
1282 1283

free:
R
Robert Shearman 已提交
1284
	free_percpu(mdev->stats);
1285 1286
	kfree(mdev);
	return ERR_PTR(err);
R
Robert Shearman 已提交
1287 1288
}

R
Robert Shearman 已提交
1289 1290 1291 1292 1293 1294 1295 1296
static void mpls_dev_destroy_rcu(struct rcu_head *head)
{
	struct mpls_dev *mdev = container_of(head, struct mpls_dev, rcu);

	free_percpu(mdev->stats);
	kfree(mdev);
}

R
Roopa Prabhu 已提交
1297
static void mpls_ifdown(struct net_device *dev, int event)
E
Eric W. Biederman 已提交
1298
{
1299
	struct mpls_route __rcu **platform_label;
E
Eric W. Biederman 已提交
1300
	struct net *net = dev_net(dev);
1301
	unsigned int nh_flags = RTNH_F_DEAD | RTNH_F_LINKDOWN;
1302
	unsigned int alive, deleted;
E
Eric W. Biederman 已提交
1303 1304
	unsigned index;

1305
	platform_label = rtnl_dereference(net->mpls.platform_label);
E
Eric W. Biederman 已提交
1306
	for (index = 0; index < net->mpls.platform_labels; index++) {
1307
		struct mpls_route *rt = rtnl_dereference(platform_label[index]);
R
Roopa Prabhu 已提交
1308

E
Eric W. Biederman 已提交
1309 1310
		if (!rt)
			continue;
R
Roopa Prabhu 已提交
1311

1312
		alive = 0;
1313
		deleted = 0;
R
Roopa Prabhu 已提交
1314
		change_nexthops(rt) {
R
Roopa Prabhu 已提交
1315
			if (rtnl_dereference(nh->nh_dev) != dev)
1316 1317
				goto next;

R
Roopa Prabhu 已提交
1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328
			switch (event) {
			case NETDEV_DOWN:
			case NETDEV_UNREGISTER:
				nh->nh_flags |= RTNH_F_DEAD;
				/* fall through */
			case NETDEV_CHANGE:
				nh->nh_flags |= RTNH_F_LINKDOWN;
				break;
			}
			if (event == NETDEV_UNREGISTER)
				RCU_INIT_POINTER(nh->nh_dev, NULL);
1329 1330 1331
next:
			if (!(nh->nh_flags & nh_flags))
				alive++;
1332 1333
			if (!rtnl_dereference(nh->nh_dev))
				deleted++;
R
Roopa Prabhu 已提交
1334
		} endfor_nexthops(rt);
1335 1336

		WRITE_ONCE(rt->rt_nhn_alive, alive);
1337 1338 1339 1340

		/* if there are no more nexthops, delete the route */
		if (event == NETDEV_UNREGISTER && deleted == rt->rt_nhn)
			mpls_route_update(net, index, NULL, NULL);
E
Eric W. Biederman 已提交
1341
	}
R
Roopa Prabhu 已提交
1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356
}

static void mpls_ifup(struct net_device *dev, unsigned int nh_flags)
{
	struct mpls_route __rcu **platform_label;
	struct net *net = dev_net(dev);
	unsigned index;
	int alive;

	platform_label = rtnl_dereference(net->mpls.platform_label);
	for (index = 0; index < net->mpls.platform_labels; index++) {
		struct mpls_route *rt = rtnl_dereference(platform_label[index]);

		if (!rt)
			continue;
1357

R
Roopa Prabhu 已提交
1358 1359 1360 1361
		alive = 0;
		change_nexthops(rt) {
			struct net_device *nh_dev =
				rtnl_dereference(nh->nh_dev);
R
Robert Shearman 已提交
1362

R
Roopa Prabhu 已提交
1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374
			if (!(nh->nh_flags & nh_flags)) {
				alive++;
				continue;
			}
			if (nh_dev != dev)
				continue;
			alive++;
			nh->nh_flags &= ~nh_flags;
		} endfor_nexthops(rt);

		ACCESS_ONCE(rt->rt_nhn_alive) = alive;
	}
E
Eric W. Biederman 已提交
1375 1376 1377 1378 1379 1380
}

static int mpls_dev_notify(struct notifier_block *this, unsigned long event,
			   void *ptr)
{
	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
R
Robert Shearman 已提交
1381
	struct mpls_dev *mdev;
R
Roopa Prabhu 已提交
1382
	unsigned int flags;
E
Eric W. Biederman 已提交
1383

R
Roopa Prabhu 已提交
1384
	if (event == NETDEV_REGISTER) {
1385
		/* For now just support Ethernet, IPGRE, SIT and IPIP devices */
1386 1387
		if (dev->type == ARPHRD_ETHER ||
		    dev->type == ARPHRD_LOOPBACK ||
1388 1389 1390
		    dev->type == ARPHRD_IPGRE ||
		    dev->type == ARPHRD_SIT ||
		    dev->type == ARPHRD_TUNNEL) {
R
Robert Shearman 已提交
1391 1392 1393 1394
			mdev = mpls_add_dev(dev);
			if (IS_ERR(mdev))
				return notifier_from_errno(PTR_ERR(mdev));
		}
R
Roopa Prabhu 已提交
1395 1396
		return NOTIFY_OK;
	}
R
Robert Shearman 已提交
1397

R
Roopa Prabhu 已提交
1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419
	mdev = mpls_dev_get(dev);
	if (!mdev)
		return NOTIFY_OK;

	switch (event) {
	case NETDEV_DOWN:
		mpls_ifdown(dev, event);
		break;
	case NETDEV_UP:
		flags = dev_get_flags(dev);
		if (flags & (IFF_RUNNING | IFF_LOWER_UP))
			mpls_ifup(dev, RTNH_F_DEAD | RTNH_F_LINKDOWN);
		else
			mpls_ifup(dev, RTNH_F_DEAD);
		break;
	case NETDEV_CHANGE:
		flags = dev_get_flags(dev);
		if (flags & (IFF_RUNNING | IFF_LOWER_UP))
			mpls_ifup(dev, RTNH_F_DEAD | RTNH_F_LINKDOWN);
		else
			mpls_ifdown(dev, event);
		break;
E
Eric W. Biederman 已提交
1420
	case NETDEV_UNREGISTER:
R
Roopa Prabhu 已提交
1421 1422 1423 1424 1425
		mpls_ifdown(dev, event);
		mdev = mpls_dev_get(dev);
		if (mdev) {
			mpls_dev_sysctl_unregister(mdev);
			RCU_INIT_POINTER(dev->mpls_ptr, NULL);
R
Robert Shearman 已提交
1426
			call_rcu(&mdev->rcu, mpls_dev_destroy_rcu);
R
Roopa Prabhu 已提交
1427
		}
E
Eric W. Biederman 已提交
1428
		break;
1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439
	case NETDEV_CHANGENAME:
		mdev = mpls_dev_get(dev);
		if (mdev) {
			int err;

			mpls_dev_sysctl_unregister(mdev);
			err = mpls_dev_sysctl_register(dev, mdev);
			if (err)
				return notifier_from_errno(err);
		}
		break;
E
Eric W. Biederman 已提交
1440 1441 1442 1443 1444 1445 1446 1447
	}
	return NOTIFY_OK;
}

static struct notifier_block mpls_dev_notifier = {
	.notifier_call = mpls_dev_notify,
};

1448
static int nla_put_via(struct sk_buff *skb,
1449
		       u8 table, const void *addr, int alen)
1450
{
1451 1452 1453
	static const int table_to_family[NEIGH_NR_TABLES + 1] = {
		AF_INET, AF_INET6, AF_DECnet, AF_PACKET,
	};
1454 1455
	struct nlattr *nla;
	struct rtvia *via;
1456
	int family = AF_UNSPEC;
1457 1458 1459 1460 1461

	nla = nla_reserve(skb, RTA_VIA, alen + 2);
	if (!nla)
		return -EMSGSIZE;

1462 1463 1464
	if (table <= NEIGH_NR_TABLES)
		family = table_to_family[table];

1465 1466 1467 1468 1469 1470
	via = nla_data(nla);
	via->rtvia_family = family;
	memcpy(via->rtvia_addr, addr, alen);
	return 0;
}

1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490
int nla_put_labels(struct sk_buff *skb, int attrtype,
		   u8 labels, const u32 label[])
{
	struct nlattr *nla;
	struct mpls_shim_hdr *nla_label;
	bool bos;
	int i;
	nla = nla_reserve(skb, attrtype, labels*4);
	if (!nla)
		return -EMSGSIZE;

	nla_label = nla_data(nla);
	bos = true;
	for (i = labels - 1; i >= 0; i--) {
		nla_label[i] = mpls_entry_encode(label[i], 0, 0, bos);
		bos = false;
	}

	return 0;
}
1491
EXPORT_SYMBOL_GPL(nla_put_labels);
1492 1493

int nla_get_labels(const struct nlattr *nla,
R
Roopa Prabhu 已提交
1494
		   u32 max_labels, u8 *labels, u32 label[])
1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522
{
	unsigned len = nla_len(nla);
	unsigned nla_labels;
	struct mpls_shim_hdr *nla_label;
	bool bos;
	int i;

	/* len needs to be an even multiple of 4 (the label size) */
	if (len & 3)
		return -EINVAL;

	/* Limit the number of new labels allowed */
	nla_labels = len/4;
	if (nla_labels > max_labels)
		return -EINVAL;

	nla_label = nla_data(nla);
	bos = true;
	for (i = nla_labels - 1; i >= 0; i--, bos = false) {
		struct mpls_entry_decoded dec;
		dec = mpls_entry_decode(nla_label + i);

		/* Ensure the bottom of stack flag is properly set
		 * and ttl and tc are both clear.
		 */
		if ((dec.bos != bos) || dec.ttl || dec.tc)
			return -EINVAL;

1523
		switch (dec.label) {
1524
		case MPLS_LABEL_IMPLNULL:
1525 1526 1527 1528 1529 1530 1531
			/* RFC3032: This is a label that an LSR may
			 * assign and distribute, but which never
			 * actually appears in the encapsulation.
			 */
			return -EINVAL;
		}

1532 1533 1534 1535 1536
		label[i] = dec.label;
	}
	*labels = nla_labels;
	return 0;
}
1537
EXPORT_SYMBOL_GPL(nla_get_labels);
1538

R
Roopa Prabhu 已提交
1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580
int nla_get_via(const struct nlattr *nla, u8 *via_alen,
		u8 *via_table, u8 via_addr[])
{
	struct rtvia *via = nla_data(nla);
	int err = -EINVAL;
	int alen;

	if (nla_len(nla) < offsetof(struct rtvia, rtvia_addr))
		goto errout;
	alen = nla_len(nla) -
			offsetof(struct rtvia, rtvia_addr);
	if (alen > MAX_VIA_ALEN)
		goto errout;

	/* Validate the address family */
	switch (via->rtvia_family) {
	case AF_PACKET:
		*via_table = NEIGH_LINK_TABLE;
		break;
	case AF_INET:
		*via_table = NEIGH_ARP_TABLE;
		if (alen != 4)
			goto errout;
		break;
	case AF_INET6:
		*via_table = NEIGH_ND_TABLE;
		if (alen != 16)
			goto errout;
		break;
	default:
		/* Unsupported address family */
		goto errout;
	}

	memcpy(via_addr, via->rtvia_addr, alen);
	*via_alen = alen;
	err = 0;

errout:
	return err;
}

1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621
static int rtm_to_route_config(struct sk_buff *skb,  struct nlmsghdr *nlh,
			       struct mpls_route_config *cfg)
{
	struct rtmsg *rtm;
	struct nlattr *tb[RTA_MAX+1];
	int index;
	int err;

	err = nlmsg_parse(nlh, sizeof(*rtm), tb, RTA_MAX, rtm_mpls_policy);
	if (err < 0)
		goto errout;

	err = -EINVAL;
	rtm = nlmsg_data(nlh);
	memset(cfg, 0, sizeof(*cfg));

	if (rtm->rtm_family != AF_MPLS)
		goto errout;
	if (rtm->rtm_dst_len != 20)
		goto errout;
	if (rtm->rtm_src_len != 0)
		goto errout;
	if (rtm->rtm_tos != 0)
		goto errout;
	if (rtm->rtm_table != RT_TABLE_MAIN)
		goto errout;
	/* Any value is acceptable for rtm_protocol */

	/* As mpls uses destination specific addresses
	 * (or source specific address in the case of multicast)
	 * all addresses have universal scope.
	 */
	if (rtm->rtm_scope != RT_SCOPE_UNIVERSE)
		goto errout;
	if (rtm->rtm_type != RTN_UNICAST)
		goto errout;
	if (rtm->rtm_flags != 0)
		goto errout;

	cfg->rc_label		= LABEL_NOT_SPECIFIED;
	cfg->rc_protocol	= rtm->rtm_protocol;
1622
	cfg->rc_via_table	= MPLS_NEIGH_TABLE_UNSPEC;
1623
	cfg->rc_ttl_propagate	= MPLS_TTL_PROP_DEFAULT;
1624 1625 1626 1627 1628 1629 1630 1631 1632 1633
	cfg->rc_nlflags		= nlh->nlmsg_flags;
	cfg->rc_nlinfo.portid	= NETLINK_CB(skb).portid;
	cfg->rc_nlinfo.nlh	= nlh;
	cfg->rc_nlinfo.nl_net	= sock_net(skb->sk);

	for (index = 0; index <= RTA_MAX; index++) {
		struct nlattr *nla = tb[index];
		if (!nla)
			continue;

1634
		switch (index) {
1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645
		case RTA_OIF:
			cfg->rc_ifindex = nla_get_u32(nla);
			break;
		case RTA_NEWDST:
			if (nla_get_labels(nla, MAX_NEW_LABELS,
					   &cfg->rc_output_labels,
					   cfg->rc_output_label))
				goto errout;
			break;
		case RTA_DST:
		{
R
Roopa Prabhu 已提交
1646
			u8 label_count;
1647 1648 1649 1650
			if (nla_get_labels(nla, 1, &label_count,
					   &cfg->rc_label))
				goto errout;

1651 1652
			/* Reserved labels may not be set */
			if (cfg->rc_label < MPLS_LABEL_FIRST_UNRESERVED)
1653 1654 1655 1656 1657 1658
				goto errout;

			break;
		}
		case RTA_VIA:
		{
R
Roopa Prabhu 已提交
1659 1660
			if (nla_get_via(nla, &cfg->rc_via_alen,
					&cfg->rc_via_table, cfg->rc_via))
1661
				goto errout;
R
Roopa Prabhu 已提交
1662 1663 1664 1665 1666 1667
			break;
		}
		case RTA_MULTIPATH:
		{
			cfg->rc_mp = nla_data(nla);
			cfg->rc_mp_len = nla_len(nla);
1668 1669
			break;
		}
1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680
		case RTA_TTL_PROPAGATE:
		{
			u8 ttl_propagate = nla_get_u8(nla);

			if (ttl_propagate > 1)
				goto errout;
			cfg->rc_ttl_propagate = ttl_propagate ?
				MPLS_TTL_PROP_ENABLED :
				MPLS_TTL_PROP_DISABLED;
			break;
		}
1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719
		default:
			/* Unsupported attribute */
			goto errout;
		}
	}

	err = 0;
errout:
	return err;
}

static int mpls_rtm_delroute(struct sk_buff *skb, struct nlmsghdr *nlh)
{
	struct mpls_route_config cfg;
	int err;

	err = rtm_to_route_config(skb, nlh, &cfg);
	if (err < 0)
		return err;

	return mpls_route_del(&cfg);
}


static int mpls_rtm_newroute(struct sk_buff *skb, struct nlmsghdr *nlh)
{
	struct mpls_route_config cfg;
	int err;

	err = rtm_to_route_config(skb, nlh, &cfg);
	if (err < 0)
		return err;

	return mpls_route_add(&cfg);
}

static int mpls_dump_route(struct sk_buff *skb, u32 portid, u32 seq, int event,
			   u32 label, struct mpls_route *rt, int flags)
{
1720
	struct net_device *dev;
1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740
	struct nlmsghdr *nlh;
	struct rtmsg *rtm;

	nlh = nlmsg_put(skb, portid, seq, event, sizeof(*rtm), flags);
	if (nlh == NULL)
		return -EMSGSIZE;

	rtm = nlmsg_data(nlh);
	rtm->rtm_family = AF_MPLS;
	rtm->rtm_dst_len = 20;
	rtm->rtm_src_len = 0;
	rtm->rtm_tos = 0;
	rtm->rtm_table = RT_TABLE_MAIN;
	rtm->rtm_protocol = rt->rt_protocol;
	rtm->rtm_scope = RT_SCOPE_UNIVERSE;
	rtm->rtm_type = RTN_UNICAST;
	rtm->rtm_flags = 0;

	if (nla_put_labels(skb, RTA_DST, 1, &label))
		goto nla_put_failure;
1741 1742 1743 1744 1745 1746 1747 1748 1749

	if (rt->rt_ttl_propagate != MPLS_TTL_PROP_DEFAULT) {
		bool ttl_propagate =
			rt->rt_ttl_propagate == MPLS_TTL_PROP_ENABLED;

		if (nla_put_u8(skb, RTA_TTL_PROPAGATE,
			       ttl_propagate))
			goto nla_put_failure;
	}
R
Roopa Prabhu 已提交
1750
	if (rt->rt_nhn == 1) {
1751
		const struct mpls_nh *nh = rt->rt_nh;
R
Roopa Prabhu 已提交
1752 1753 1754 1755 1756

		if (nh->nh_labels &&
		    nla_put_labels(skb, RTA_NEWDST, nh->nh_labels,
				   nh->nh_label))
			goto nla_put_failure;
1757
		if (nh->nh_via_table != MPLS_NEIGH_TABLE_UNSPEC &&
1758
		    nla_put_via(skb, nh->nh_via_table, mpls_nh_via(rt, nh),
R
Roopa Prabhu 已提交
1759 1760 1761 1762 1763
				nh->nh_via_alen))
			goto nla_put_failure;
		dev = rtnl_dereference(nh->nh_dev);
		if (dev && nla_put_u32(skb, RTA_OIF, dev->ifindex))
			goto nla_put_failure;
R
Roopa Prabhu 已提交
1764 1765 1766 1767
		if (nh->nh_flags & RTNH_F_LINKDOWN)
			rtm->rtm_flags |= RTNH_F_LINKDOWN;
		if (nh->nh_flags & RTNH_F_DEAD)
			rtm->rtm_flags |= RTNH_F_DEAD;
R
Roopa Prabhu 已提交
1768 1769 1770
	} else {
		struct rtnexthop *rtnh;
		struct nlattr *mp;
R
Roopa Prabhu 已提交
1771 1772
		int dead = 0;
		int linkdown = 0;
R
Roopa Prabhu 已提交
1773 1774 1775 1776 1777 1778

		mp = nla_nest_start(skb, RTA_MULTIPATH);
		if (!mp)
			goto nla_put_failure;

		for_nexthops(rt) {
1779 1780 1781 1782
			dev = rtnl_dereference(nh->nh_dev);
			if (!dev)
				continue;

R
Roopa Prabhu 已提交
1783 1784 1785 1786
			rtnh = nla_reserve_nohdr(skb, sizeof(*rtnh));
			if (!rtnh)
				goto nla_put_failure;

1787
			rtnh->rtnh_ifindex = dev->ifindex;
R
Roopa Prabhu 已提交
1788 1789 1790 1791 1792 1793 1794 1795 1796
			if (nh->nh_flags & RTNH_F_LINKDOWN) {
				rtnh->rtnh_flags |= RTNH_F_LINKDOWN;
				linkdown++;
			}
			if (nh->nh_flags & RTNH_F_DEAD) {
				rtnh->rtnh_flags |= RTNH_F_DEAD;
				dead++;
			}

R
Roopa Prabhu 已提交
1797 1798 1799 1800
			if (nh->nh_labels && nla_put_labels(skb, RTA_NEWDST,
							    nh->nh_labels,
							    nh->nh_label))
				goto nla_put_failure;
1801 1802
			if (nh->nh_via_table != MPLS_NEIGH_TABLE_UNSPEC &&
			    nla_put_via(skb, nh->nh_via_table,
1803
					mpls_nh_via(rt, nh),
R
Roopa Prabhu 已提交
1804 1805 1806 1807 1808 1809 1810
					nh->nh_via_alen))
				goto nla_put_failure;

			/* length of rtnetlink header + attributes */
			rtnh->rtnh_len = nlmsg_get_pos(skb) - (void *)rtnh;
		} endfor_nexthops(rt);

R
Roopa Prabhu 已提交
1811 1812 1813 1814 1815
		if (linkdown == rt->rt_nhn)
			rtm->rtm_flags |= RTNH_F_LINKDOWN;
		if (dead == rt->rt_nhn)
			rtm->rtm_flags |= RTNH_F_DEAD;

R
Roopa Prabhu 已提交
1816 1817
		nla_nest_end(skb, mp);
	}
1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829

	nlmsg_end(skb, nlh);
	return 0;

nla_put_failure:
	nlmsg_cancel(skb, nlh);
	return -EMSGSIZE;
}

static int mpls_dump_routes(struct sk_buff *skb, struct netlink_callback *cb)
{
	struct net *net = sock_net(skb->sk);
1830 1831
	struct mpls_route __rcu **platform_label;
	size_t platform_labels;
1832 1833 1834 1835 1836
	unsigned int index;

	ASSERT_RTNL();

	index = cb->args[0];
1837 1838
	if (index < MPLS_LABEL_FIRST_UNRESERVED)
		index = MPLS_LABEL_FIRST_UNRESERVED;
1839

1840 1841 1842
	platform_label = rtnl_dereference(net->mpls.platform_label);
	platform_labels = net->mpls.platform_labels;
	for (; index < platform_labels; index++) {
1843
		struct mpls_route *rt;
1844
		rt = rtnl_dereference(platform_label[index]);
1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857
		if (!rt)
			continue;

		if (mpls_dump_route(skb, NETLINK_CB(cb->skb).portid,
				    cb->nlh->nlmsg_seq, RTM_NEWROUTE,
				    index, rt, NLM_F_MULTI) < 0)
			break;
	}
	cb->args[0] = index;

	return skb->len;
}

1858 1859 1860 1861
static inline size_t lfib_nlmsg_size(struct mpls_route *rt)
{
	size_t payload =
		NLMSG_ALIGN(sizeof(struct rtmsg))
1862 1863
		+ nla_total_size(4)			/* RTA_DST */
		+ nla_total_size(1);			/* RTA_TTL_PROPAGATE */
R
Roopa Prabhu 已提交
1864 1865 1866 1867 1868 1869

	if (rt->rt_nhn == 1) {
		struct mpls_nh *nh = rt->rt_nh;

		if (nh->nh_dev)
			payload += nla_total_size(4); /* RTA_OIF */
1870
		if (nh->nh_via_table != MPLS_NEIGH_TABLE_UNSPEC) /* RTA_VIA */
1871
			payload += nla_total_size(2 + nh->nh_via_alen);
R
Roopa Prabhu 已提交
1872 1873 1874 1875 1876 1877 1878 1879
		if (nh->nh_labels) /* RTA_NEWDST */
			payload += nla_total_size(nh->nh_labels * 4);
	} else {
		/* each nexthop is packed in an attribute */
		size_t nhsize = 0;

		for_nexthops(rt) {
			nhsize += nla_total_size(sizeof(struct rtnexthop));
1880 1881 1882
			/* RTA_VIA */
			if (nh->nh_via_table != MPLS_NEIGH_TABLE_UNSPEC)
				nhsize += nla_total_size(2 + nh->nh_via_alen);
R
Roopa Prabhu 已提交
1883 1884 1885 1886 1887 1888 1889
			if (nh->nh_labels)
				nhsize += nla_total_size(nh->nh_labels * 4);
		} endfor_nexthops(rt);
		/* nested attribute */
		payload += nla_total_size(nhsize);
	}

1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919
	return payload;
}

static void rtmsg_lfib(int event, u32 label, struct mpls_route *rt,
		       struct nlmsghdr *nlh, struct net *net, u32 portid,
		       unsigned int nlm_flags)
{
	struct sk_buff *skb;
	u32 seq = nlh ? nlh->nlmsg_seq : 0;
	int err = -ENOBUFS;

	skb = nlmsg_new(lfib_nlmsg_size(rt), GFP_KERNEL);
	if (skb == NULL)
		goto errout;

	err = mpls_dump_route(skb, portid, seq, event, label, rt, nlm_flags);
	if (err < 0) {
		/* -EMSGSIZE implies BUG in lfib_nlmsg_size */
		WARN_ON(err == -EMSGSIZE);
		kfree_skb(skb);
		goto errout;
	}
	rtnl_notify(skb, net, portid, RTNLGRP_MPLS_ROUTE, nlh, GFP_KERNEL);

	return;
errout:
	if (err < 0)
		rtnl_set_sk_err(net, RTNLGRP_MPLS_ROUTE, err);
}

1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938
static int resize_platform_label_table(struct net *net, size_t limit)
{
	size_t size = sizeof(struct mpls_route *) * limit;
	size_t old_limit;
	size_t cp_size;
	struct mpls_route __rcu **labels = NULL, **old;
	struct mpls_route *rt0 = NULL, *rt2 = NULL;
	unsigned index;

	if (size) {
		labels = kzalloc(size, GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY);
		if (!labels)
			labels = vzalloc(size);

		if (!labels)
			goto nolabels;
	}

	/* In case the predefined labels need to be populated */
1939
	if (limit > MPLS_LABEL_IPV4NULL) {
1940
		struct net_device *lo = net->loopback_dev;
1941
		rt0 = mpls_rt_alloc(1, lo->addr_len);
1942 1943
		if (!rt0)
			goto nort0;
R
Roopa Prabhu 已提交
1944
		RCU_INIT_POINTER(rt0->rt_nh->nh_dev, lo);
1945
		rt0->rt_protocol = RTPROT_KERNEL;
1946
		rt0->rt_payload_type = MPT_IPV4;
1947
		rt0->rt_ttl_propagate = MPLS_TTL_PROP_DEFAULT;
R
Roopa Prabhu 已提交
1948
		rt0->rt_nh->nh_via_table = NEIGH_LINK_TABLE;
1949
		rt0->rt_nh->nh_via_alen = lo->addr_len;
1950 1951
		memcpy(__mpls_nh_via(rt0, rt0->rt_nh), lo->dev_addr,
		       lo->addr_len);
1952
	}
1953
	if (limit > MPLS_LABEL_IPV6NULL) {
1954
		struct net_device *lo = net->loopback_dev;
1955
		rt2 = mpls_rt_alloc(1, lo->addr_len);
1956 1957
		if (!rt2)
			goto nort2;
R
Roopa Prabhu 已提交
1958
		RCU_INIT_POINTER(rt2->rt_nh->nh_dev, lo);
1959
		rt2->rt_protocol = RTPROT_KERNEL;
1960
		rt2->rt_payload_type = MPT_IPV6;
1961
		rt2->rt_ttl_propagate = MPLS_TTL_PROP_DEFAULT;
R
Roopa Prabhu 已提交
1962
		rt2->rt_nh->nh_via_table = NEIGH_LINK_TABLE;
1963
		rt2->rt_nh->nh_via_alen = lo->addr_len;
1964 1965
		memcpy(__mpls_nh_via(rt2, rt2->rt_nh), lo->dev_addr,
		       lo->addr_len);
1966 1967 1968 1969
	}

	rtnl_lock();
	/* Remember the original table */
1970
	old = rtnl_dereference(net->mpls.platform_label);
1971 1972 1973 1974
	old_limit = net->mpls.platform_labels;

	/* Free any labels beyond the new table */
	for (index = limit; index < old_limit; index++)
R
Roopa Prabhu 已提交
1975
		mpls_route_update(net, index, NULL, NULL);
1976 1977 1978 1979 1980 1981 1982 1983 1984

	/* Copy over the old labels */
	cp_size = size;
	if (old_limit < limit)
		cp_size = old_limit * sizeof(struct mpls_route *);

	memcpy(labels, old, cp_size);

	/* If needed set the predefined labels */
1985 1986 1987
	if ((old_limit <= MPLS_LABEL_IPV6NULL) &&
	    (limit > MPLS_LABEL_IPV6NULL)) {
		RCU_INIT_POINTER(labels[MPLS_LABEL_IPV6NULL], rt2);
1988 1989 1990
		rt2 = NULL;
	}

1991 1992 1993
	if ((old_limit <= MPLS_LABEL_IPV4NULL) &&
	    (limit > MPLS_LABEL_IPV4NULL)) {
		RCU_INIT_POINTER(labels[MPLS_LABEL_IPV4NULL], rt0);
1994 1995 1996 1997 1998
		rt0 = NULL;
	}

	/* Update the global pointers */
	net->mpls.platform_labels = limit;
1999
	rcu_assign_pointer(net->mpls.platform_label, labels);
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042

	rtnl_unlock();

	mpls_rt_free(rt2);
	mpls_rt_free(rt0);

	if (old) {
		synchronize_rcu();
		kvfree(old);
	}
	return 0;

nort2:
	mpls_rt_free(rt0);
nort0:
	kvfree(labels);
nolabels:
	return -ENOMEM;
}

static int mpls_platform_labels(struct ctl_table *table, int write,
				void __user *buffer, size_t *lenp, loff_t *ppos)
{
	struct net *net = table->data;
	int platform_labels = net->mpls.platform_labels;
	int ret;
	struct ctl_table tmp = {
		.procname	= table->procname,
		.data		= &platform_labels,
		.maxlen		= sizeof(int),
		.mode		= table->mode,
		.extra1		= &zero,
		.extra2		= &label_limit,
	};

	ret = proc_dointvec_minmax(&tmp, write, buffer, lenp, ppos);

	if (write && ret == 0)
		ret = resize_platform_label_table(net, platform_labels);

	return ret;
}

2043 2044 2045
#define MPLS_NS_SYSCTL_OFFSET(field)		\
	(&((struct net *)0)->field)

2046
static const struct ctl_table mpls_table[] = {
2047 2048 2049 2050 2051 2052 2053
	{
		.procname	= "platform_labels",
		.data		= NULL,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= mpls_platform_labels,
	},
2054 2055 2056 2057 2058 2059 2060 2061 2062
	{
		.procname	= "ip_ttl_propagate",
		.data		= MPLS_NS_SYSCTL_OFFSET(mpls.ip_ttl_propagate),
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &zero,
		.extra2		= &one,
	},
2063 2064 2065 2066 2067 2068 2069 2070 2071
	{
		.procname	= "default_ttl",
		.data		= MPLS_NS_SYSCTL_OFFSET(mpls.default_ttl),
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &one,
		.extra2		= &ttl_max,
	},
2072 2073 2074
	{ }
};

E
Eric W. Biederman 已提交
2075 2076
static int mpls_net_init(struct net *net)
{
2077
	struct ctl_table *table;
2078
	int i;
2079

E
Eric W. Biederman 已提交
2080 2081
	net->mpls.platform_labels = 0;
	net->mpls.platform_label = NULL;
2082
	net->mpls.ip_ttl_propagate = 1;
2083
	net->mpls.default_ttl = 255;
E
Eric W. Biederman 已提交
2084

2085 2086 2087 2088
	table = kmemdup(mpls_table, sizeof(mpls_table), GFP_KERNEL);
	if (table == NULL)
		return -ENOMEM;

2089 2090 2091 2092 2093 2094
	/* Table data contains only offsets relative to the base of
	 * the mdev at this point, so make them absolute.
	 */
	for (i = 0; i < ARRAY_SIZE(mpls_table) - 1; i++)
		table[i].data = (char *)net + (uintptr_t)table[i].data;

2095
	net->mpls.ctl = register_net_sysctl(net, "net/mpls", table);
2096 2097
	if (net->mpls.ctl == NULL) {
		kfree(table);
2098
		return -ENOMEM;
2099
	}
2100

E
Eric W. Biederman 已提交
2101 2102 2103 2104 2105
	return 0;
}

static void mpls_net_exit(struct net *net)
{
2106 2107
	struct mpls_route __rcu **platform_label;
	size_t platform_labels;
2108
	struct ctl_table *table;
E
Eric W. Biederman 已提交
2109 2110
	unsigned int index;

2111 2112 2113 2114
	table = net->mpls.ctl->ctl_table_arg;
	unregister_net_sysctl_table(net->mpls.ctl);
	kfree(table);

2115 2116
	/* An rcu grace period has passed since there was a device in
	 * the network namespace (and thus the last in flight packet)
E
Eric W. Biederman 已提交
2117 2118 2119 2120 2121 2122 2123 2124
	 * left this network namespace.  This is because
	 * unregister_netdevice_many and netdev_run_todo has completed
	 * for each network device that was in this network namespace.
	 *
	 * As such no additional rcu synchronization is necessary when
	 * freeing the platform_label table.
	 */
	rtnl_lock();
2125 2126 2127 2128 2129
	platform_label = rtnl_dereference(net->mpls.platform_label);
	platform_labels = net->mpls.platform_labels;
	for (index = 0; index < platform_labels; index++) {
		struct mpls_route *rt = rtnl_dereference(platform_label[index]);
		RCU_INIT_POINTER(platform_label[index], NULL);
2130
		mpls_notify_route(net, index, rt, NULL, NULL);
E
Eric W. Biederman 已提交
2131 2132 2133 2134
		mpls_rt_free(rt);
	}
	rtnl_unlock();

2135
	kvfree(platform_label);
E
Eric W. Biederman 已提交
2136 2137 2138 2139 2140 2141 2142
}

static struct pernet_operations mpls_net_ops = {
	.init = mpls_net_init,
	.exit = mpls_net_exit,
};

R
Robert Shearman 已提交
2143 2144 2145 2146 2147 2148
static struct rtnl_af_ops mpls_af_ops __read_mostly = {
	.family		   = AF_MPLS,
	.fill_stats_af	   = mpls_fill_stats_af,
	.get_stats_af_size = mpls_get_stats_af_size,
};

E
Eric W. Biederman 已提交
2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164
static int __init mpls_init(void)
{
	int err;

	BUILD_BUG_ON(sizeof(struct mpls_shim_hdr) != 4);

	err = register_pernet_subsys(&mpls_net_ops);
	if (err)
		goto out;

	err = register_netdevice_notifier(&mpls_dev_notifier);
	if (err)
		goto out_unregister_pernet;

	dev_add_pack(&mpls_packet_type);

R
Robert Shearman 已提交
2165 2166
	rtnl_af_register(&mpls_af_ops);

2167 2168 2169
	rtnl_register(PF_MPLS, RTM_NEWROUTE, mpls_rtm_newroute, NULL, NULL);
	rtnl_register(PF_MPLS, RTM_DELROUTE, mpls_rtm_delroute, NULL, NULL);
	rtnl_register(PF_MPLS, RTM_GETROUTE, NULL, mpls_dump_routes, NULL);
D
David Ahern 已提交
2170 2171
	rtnl_register(PF_MPLS, RTM_GETNETCONF, mpls_netconf_get_devconf,
		      mpls_netconf_dump_devconf, NULL);
E
Eric W. Biederman 已提交
2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183
	err = 0;
out:
	return err;

out_unregister_pernet:
	unregister_pernet_subsys(&mpls_net_ops);
	goto out;
}
module_init(mpls_init);

static void __exit mpls_exit(void)
{
2184
	rtnl_unregister_all(PF_MPLS);
R
Robert Shearman 已提交
2185
	rtnl_af_unregister(&mpls_af_ops);
E
Eric W. Biederman 已提交
2186 2187 2188 2189 2190 2191 2192 2193 2194
	dev_remove_pack(&mpls_packet_type);
	unregister_netdevice_notifier(&mpls_dev_notifier);
	unregister_pernet_subsys(&mpls_net_ops);
}
module_exit(mpls_exit);

MODULE_DESCRIPTION("MultiProtocol Label Switching");
MODULE_LICENSE("GPL v2");
MODULE_ALIAS_NETPROTO(PF_MPLS);