]>
Commit | Line | Data |
---|---|---|
0189197f EB |
1 | #include <linux/types.h> |
2 | #include <linux/skbuff.h> | |
3 | #include <linux/socket.h> | |
7720c01f | 4 | #include <linux/sysctl.h> |
0189197f EB |
5 | #include <linux/net.h> |
6 | #include <linux/module.h> | |
7 | #include <linux/if_arp.h> | |
8 | #include <linux/ipv6.h> | |
9 | #include <linux/mpls.h> | |
10 | #include <net/ip.h> | |
11 | #include <net/dst.h> | |
12 | #include <net/sock.h> | |
13 | #include <net/arp.h> | |
14 | #include <net/ip_fib.h> | |
15 | #include <net/netevent.h> | |
16 | #include <net/netns/generic.h> | |
17 | #include "internal.h" | |
18 | ||
a2519929 | 19 | #define LABEL_NOT_SPECIFIED (1<<20) |
0189197f EB |
20 | #define MAX_NEW_LABELS 2 |
21 | ||
22 | /* This maximum ha length copied from the definition of struct neighbour */ | |
23 | #define MAX_VIA_ALEN (ALIGN(MAX_ADDR_LEN, sizeof(unsigned long))) | |
24 | ||
25 | struct mpls_route { /* next hop label forwarding entry */ | |
26 | struct net_device *rt_dev; | |
27 | struct rcu_head rt_rcu; | |
28 | u32 rt_label[MAX_NEW_LABELS]; | |
29 | u8 rt_protocol; /* routing protocol that set this entry */ | |
30 | u8 rt_labels:2, | |
31 | rt_via_alen:6; | |
32 | unsigned short rt_via_family; | |
33 | u8 rt_via[0]; | |
34 | }; | |
35 | ||
7720c01f EB |
36 | static int zero = 0; |
37 | static int label_limit = (1 << 20) - 1; | |
38 | ||
0189197f EB |
39 | static struct mpls_route *mpls_route_input_rcu(struct net *net, unsigned index) |
40 | { | |
41 | struct mpls_route *rt = NULL; | |
42 | ||
43 | if (index < net->mpls.platform_labels) { | |
44 | struct mpls_route __rcu **platform_label = | |
45 | rcu_dereference(net->mpls.platform_label); | |
46 | rt = rcu_dereference(platform_label[index]); | |
47 | } | |
48 | return rt; | |
49 | } | |
50 | ||
51 | static bool mpls_output_possible(const struct net_device *dev) | |
52 | { | |
53 | return dev && (dev->flags & IFF_UP) && netif_carrier_ok(dev); | |
54 | } | |
55 | ||
56 | static unsigned int mpls_rt_header_size(const struct mpls_route *rt) | |
57 | { | |
58 | /* The size of the layer 2.5 labels to be added for this route */ | |
59 | return rt->rt_labels * sizeof(struct mpls_shim_hdr); | |
60 | } | |
61 | ||
62 | static unsigned int mpls_dev_mtu(const struct net_device *dev) | |
63 | { | |
64 | /* The amount of data the layer 2 frame can hold */ | |
65 | return dev->mtu; | |
66 | } | |
67 | ||
68 | static bool mpls_pkt_too_big(const struct sk_buff *skb, unsigned int mtu) | |
69 | { | |
70 | if (skb->len <= mtu) | |
71 | return false; | |
72 | ||
73 | if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu) | |
74 | return false; | |
75 | ||
76 | return true; | |
77 | } | |
78 | ||
79 | static bool mpls_egress(struct mpls_route *rt, struct sk_buff *skb, | |
80 | struct mpls_entry_decoded dec) | |
81 | { | |
82 | /* RFC4385 and RFC5586 encode other packets in mpls such that | |
83 | * they don't conflict with the ip version number, making | |
84 | * decoding by examining the ip version correct in everything | |
85 | * except for the strangest cases. | |
86 | * | |
87 | * The strange cases if we choose to support them will require | |
88 | * manual configuration. | |
89 | */ | |
90 | struct iphdr *hdr4 = ip_hdr(skb); | |
91 | bool success = true; | |
92 | ||
93 | if (hdr4->version == 4) { | |
94 | skb->protocol = htons(ETH_P_IP); | |
95 | csum_replace2(&hdr4->check, | |
96 | htons(hdr4->ttl << 8), | |
97 | htons(dec.ttl << 8)); | |
98 | hdr4->ttl = dec.ttl; | |
99 | } | |
100 | else if (hdr4->version == 6) { | |
101 | struct ipv6hdr *hdr6 = ipv6_hdr(skb); | |
102 | skb->protocol = htons(ETH_P_IPV6); | |
103 | hdr6->hop_limit = dec.ttl; | |
104 | } | |
105 | else | |
106 | /* version 0 and version 1 are used by pseudo wires */ | |
107 | success = false; | |
108 | return success; | |
109 | } | |
110 | ||
111 | static int mpls_forward(struct sk_buff *skb, struct net_device *dev, | |
112 | struct packet_type *pt, struct net_device *orig_dev) | |
113 | { | |
114 | struct net *net = dev_net(dev); | |
115 | struct mpls_shim_hdr *hdr; | |
116 | struct mpls_route *rt; | |
117 | struct mpls_entry_decoded dec; | |
118 | struct net_device *out_dev; | |
119 | unsigned int hh_len; | |
120 | unsigned int new_header_size; | |
121 | unsigned int mtu; | |
122 | int err; | |
123 | ||
124 | /* Careful this entire function runs inside of an rcu critical section */ | |
125 | ||
126 | if (skb->pkt_type != PACKET_HOST) | |
127 | goto drop; | |
128 | ||
129 | if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) | |
130 | goto drop; | |
131 | ||
132 | if (!pskb_may_pull(skb, sizeof(*hdr))) | |
133 | goto drop; | |
134 | ||
135 | /* Read and decode the label */ | |
136 | hdr = mpls_hdr(skb); | |
137 | dec = mpls_entry_decode(hdr); | |
138 | ||
139 | /* Pop the label */ | |
140 | skb_pull(skb, sizeof(*hdr)); | |
141 | skb_reset_network_header(skb); | |
142 | ||
143 | skb_orphan(skb); | |
144 | ||
145 | rt = mpls_route_input_rcu(net, dec.label); | |
146 | if (!rt) | |
147 | goto drop; | |
148 | ||
149 | /* Find the output device */ | |
150 | out_dev = rt->rt_dev; | |
151 | if (!mpls_output_possible(out_dev)) | |
152 | goto drop; | |
153 | ||
154 | if (skb_warn_if_lro(skb)) | |
155 | goto drop; | |
156 | ||
157 | skb_forward_csum(skb); | |
158 | ||
159 | /* Verify ttl is valid */ | |
160 | if (dec.ttl <= 2) | |
161 | goto drop; | |
162 | dec.ttl -= 1; | |
163 | ||
164 | /* Verify the destination can hold the packet */ | |
165 | new_header_size = mpls_rt_header_size(rt); | |
166 | mtu = mpls_dev_mtu(out_dev); | |
167 | if (mpls_pkt_too_big(skb, mtu - new_header_size)) | |
168 | goto drop; | |
169 | ||
170 | hh_len = LL_RESERVED_SPACE(out_dev); | |
171 | if (!out_dev->header_ops) | |
172 | hh_len = 0; | |
173 | ||
174 | /* Ensure there is enough space for the headers in the skb */ | |
175 | if (skb_cow(skb, hh_len + new_header_size)) | |
176 | goto drop; | |
177 | ||
178 | skb->dev = out_dev; | |
179 | skb->protocol = htons(ETH_P_MPLS_UC); | |
180 | ||
181 | if (unlikely(!new_header_size && dec.bos)) { | |
182 | /* Penultimate hop popping */ | |
183 | if (!mpls_egress(rt, skb, dec)) | |
184 | goto drop; | |
185 | } else { | |
186 | bool bos; | |
187 | int i; | |
188 | skb_push(skb, new_header_size); | |
189 | skb_reset_network_header(skb); | |
190 | /* Push the new labels */ | |
191 | hdr = mpls_hdr(skb); | |
192 | bos = dec.bos; | |
193 | for (i = rt->rt_labels - 1; i >= 0; i--) { | |
194 | hdr[i] = mpls_entry_encode(rt->rt_label[i], dec.ttl, 0, bos); | |
195 | bos = false; | |
196 | } | |
197 | } | |
198 | ||
199 | err = neigh_xmit(rt->rt_via_family, out_dev, rt->rt_via, skb); | |
200 | if (err) | |
201 | net_dbg_ratelimited("%s: packet transmission failed: %d\n", | |
202 | __func__, err); | |
203 | return 0; | |
204 | ||
205 | drop: | |
206 | kfree_skb(skb); | |
207 | return NET_RX_DROP; | |
208 | } | |
209 | ||
210 | static struct packet_type mpls_packet_type __read_mostly = { | |
211 | .type = cpu_to_be16(ETH_P_MPLS_UC), | |
212 | .func = mpls_forward, | |
213 | }; | |
214 | ||
a2519929 EB |
215 | struct mpls_route_config { |
216 | u32 rc_protocol; | |
217 | u32 rc_ifindex; | |
218 | u16 rc_via_family; | |
219 | u16 rc_via_alen; | |
220 | u8 rc_via[MAX_VIA_ALEN]; | |
221 | u32 rc_label; | |
222 | u32 rc_output_labels; | |
223 | u32 rc_output_label[MAX_NEW_LABELS]; | |
224 | u32 rc_nlflags; | |
225 | struct nl_info rc_nlinfo; | |
226 | }; | |
227 | ||
0189197f EB |
228 | static struct mpls_route *mpls_rt_alloc(size_t alen) |
229 | { | |
230 | struct mpls_route *rt; | |
231 | ||
232 | rt = kzalloc(GFP_KERNEL, sizeof(*rt) + alen); | |
233 | if (rt) | |
234 | rt->rt_via_alen = alen; | |
235 | return rt; | |
236 | } | |
237 | ||
238 | static void mpls_rt_free(struct mpls_route *rt) | |
239 | { | |
240 | if (rt) | |
241 | kfree_rcu(rt, rt_rcu); | |
242 | } | |
243 | ||
244 | static void mpls_route_update(struct net *net, unsigned index, | |
245 | struct net_device *dev, struct mpls_route *new, | |
246 | const struct nl_info *info) | |
247 | { | |
248 | struct mpls_route *rt, *old = NULL; | |
249 | ||
250 | ASSERT_RTNL(); | |
251 | ||
252 | rt = net->mpls.platform_label[index]; | |
253 | if (!dev || (rt && (rt->rt_dev == dev))) { | |
254 | rcu_assign_pointer(net->mpls.platform_label[index], new); | |
255 | old = rt; | |
256 | } | |
257 | ||
258 | /* If we removed a route free it now */ | |
259 | mpls_rt_free(old); | |
260 | } | |
261 | ||
a2519929 EB |
262 | static unsigned find_free_label(struct net *net) |
263 | { | |
264 | unsigned index; | |
265 | for (index = 16; index < net->mpls.platform_labels; index++) { | |
266 | if (!net->mpls.platform_label[index]) | |
267 | return index; | |
268 | } | |
269 | return LABEL_NOT_SPECIFIED; | |
270 | } | |
271 | ||
272 | static int mpls_route_add(struct mpls_route_config *cfg) | |
273 | { | |
274 | struct net *net = cfg->rc_nlinfo.nl_net; | |
275 | struct net_device *dev = NULL; | |
276 | struct mpls_route *rt, *old; | |
277 | unsigned index; | |
278 | int i; | |
279 | int err = -EINVAL; | |
280 | ||
281 | index = cfg->rc_label; | |
282 | ||
283 | /* If a label was not specified during insert pick one */ | |
284 | if ((index == LABEL_NOT_SPECIFIED) && | |
285 | (cfg->rc_nlflags & NLM_F_CREATE)) { | |
286 | index = find_free_label(net); | |
287 | } | |
288 | ||
289 | /* The first 16 labels are reserved, and may not be set */ | |
290 | if (index < 16) | |
291 | goto errout; | |
292 | ||
293 | /* The full 20 bit range may not be supported. */ | |
294 | if (index >= net->mpls.platform_labels) | |
295 | goto errout; | |
296 | ||
297 | /* Ensure only a supported number of labels are present */ | |
298 | if (cfg->rc_output_labels > MAX_NEW_LABELS) | |
299 | goto errout; | |
300 | ||
301 | err = -ENODEV; | |
302 | dev = dev_get_by_index(net, cfg->rc_ifindex); | |
303 | if (!dev) | |
304 | goto errout; | |
305 | ||
306 | /* For now just support ethernet devices */ | |
307 | err = -EINVAL; | |
308 | if ((dev->type != ARPHRD_ETHER) && (dev->type != ARPHRD_LOOPBACK)) | |
309 | goto errout; | |
310 | ||
311 | err = -EINVAL; | |
312 | if ((cfg->rc_via_family == AF_PACKET) && | |
313 | (dev->addr_len != cfg->rc_via_alen)) | |
314 | goto errout; | |
315 | ||
316 | /* Append makes no sense with mpls */ | |
317 | err = -EINVAL; | |
318 | if (cfg->rc_nlflags & NLM_F_APPEND) | |
319 | goto errout; | |
320 | ||
321 | err = -EEXIST; | |
322 | old = net->mpls.platform_label[index]; | |
323 | if ((cfg->rc_nlflags & NLM_F_EXCL) && old) | |
324 | goto errout; | |
325 | ||
326 | err = -EEXIST; | |
327 | if (!(cfg->rc_nlflags & NLM_F_REPLACE) && old) | |
328 | goto errout; | |
329 | ||
330 | err = -ENOENT; | |
331 | if (!(cfg->rc_nlflags & NLM_F_CREATE) && !old) | |
332 | goto errout; | |
333 | ||
334 | err = -ENOMEM; | |
335 | rt = mpls_rt_alloc(cfg->rc_via_alen); | |
336 | if (!rt) | |
337 | goto errout; | |
338 | ||
339 | rt->rt_labels = cfg->rc_output_labels; | |
340 | for (i = 0; i < rt->rt_labels; i++) | |
341 | rt->rt_label[i] = cfg->rc_output_label[i]; | |
342 | rt->rt_protocol = cfg->rc_protocol; | |
343 | rt->rt_dev = dev; | |
344 | rt->rt_via_family = cfg->rc_via_family; | |
345 | memcpy(rt->rt_via, cfg->rc_via, cfg->rc_via_alen); | |
346 | ||
347 | mpls_route_update(net, index, NULL, rt, &cfg->rc_nlinfo); | |
348 | ||
349 | dev_put(dev); | |
350 | return 0; | |
351 | ||
352 | errout: | |
353 | if (dev) | |
354 | dev_put(dev); | |
355 | return err; | |
356 | } | |
357 | ||
358 | static int mpls_route_del(struct mpls_route_config *cfg) | |
359 | { | |
360 | struct net *net = cfg->rc_nlinfo.nl_net; | |
361 | unsigned index; | |
362 | int err = -EINVAL; | |
363 | ||
364 | index = cfg->rc_label; | |
365 | ||
366 | /* The first 16 labels are reserved, and may not be removed */ | |
367 | if (index < 16) | |
368 | goto errout; | |
369 | ||
370 | /* The full 20 bit range may not be supported */ | |
371 | if (index >= net->mpls.platform_labels) | |
372 | goto errout; | |
373 | ||
374 | mpls_route_update(net, index, NULL, NULL, &cfg->rc_nlinfo); | |
375 | ||
376 | err = 0; | |
377 | errout: | |
378 | return err; | |
379 | } | |
380 | ||
0189197f EB |
381 | static void mpls_ifdown(struct net_device *dev) |
382 | { | |
383 | struct net *net = dev_net(dev); | |
384 | unsigned index; | |
385 | ||
386 | for (index = 0; index < net->mpls.platform_labels; index++) { | |
387 | struct mpls_route *rt = net->mpls.platform_label[index]; | |
388 | if (!rt) | |
389 | continue; | |
390 | if (rt->rt_dev != dev) | |
391 | continue; | |
392 | rt->rt_dev = NULL; | |
393 | } | |
394 | } | |
395 | ||
396 | static int mpls_dev_notify(struct notifier_block *this, unsigned long event, | |
397 | void *ptr) | |
398 | { | |
399 | struct net_device *dev = netdev_notifier_info_to_dev(ptr); | |
400 | ||
401 | switch(event) { | |
402 | case NETDEV_UNREGISTER: | |
403 | mpls_ifdown(dev); | |
404 | break; | |
405 | } | |
406 | return NOTIFY_OK; | |
407 | } | |
408 | ||
409 | static struct notifier_block mpls_dev_notifier = { | |
410 | .notifier_call = mpls_dev_notify, | |
411 | }; | |
412 | ||
966bae33 EB |
413 | int nla_put_labels(struct sk_buff *skb, int attrtype, |
414 | u8 labels, const u32 label[]) | |
415 | { | |
416 | struct nlattr *nla; | |
417 | struct mpls_shim_hdr *nla_label; | |
418 | bool bos; | |
419 | int i; | |
420 | nla = nla_reserve(skb, attrtype, labels*4); | |
421 | if (!nla) | |
422 | return -EMSGSIZE; | |
423 | ||
424 | nla_label = nla_data(nla); | |
425 | bos = true; | |
426 | for (i = labels - 1; i >= 0; i--) { | |
427 | nla_label[i] = mpls_entry_encode(label[i], 0, 0, bos); | |
428 | bos = false; | |
429 | } | |
430 | ||
431 | return 0; | |
432 | } | |
433 | ||
434 | int nla_get_labels(const struct nlattr *nla, | |
435 | u32 max_labels, u32 *labels, u32 label[]) | |
436 | { | |
437 | unsigned len = nla_len(nla); | |
438 | unsigned nla_labels; | |
439 | struct mpls_shim_hdr *nla_label; | |
440 | bool bos; | |
441 | int i; | |
442 | ||
443 | /* len needs to be an even multiple of 4 (the label size) */ | |
444 | if (len & 3) | |
445 | return -EINVAL; | |
446 | ||
447 | /* Limit the number of new labels allowed */ | |
448 | nla_labels = len/4; | |
449 | if (nla_labels > max_labels) | |
450 | return -EINVAL; | |
451 | ||
452 | nla_label = nla_data(nla); | |
453 | bos = true; | |
454 | for (i = nla_labels - 1; i >= 0; i--, bos = false) { | |
455 | struct mpls_entry_decoded dec; | |
456 | dec = mpls_entry_decode(nla_label + i); | |
457 | ||
458 | /* Ensure the bottom of stack flag is properly set | |
459 | * and ttl and tc are both clear. | |
460 | */ | |
461 | if ((dec.bos != bos) || dec.ttl || dec.tc) | |
462 | return -EINVAL; | |
463 | ||
464 | label[i] = dec.label; | |
465 | } | |
466 | *labels = nla_labels; | |
467 | return 0; | |
468 | } | |
469 | ||
7720c01f EB |
470 | static int resize_platform_label_table(struct net *net, size_t limit) |
471 | { | |
472 | size_t size = sizeof(struct mpls_route *) * limit; | |
473 | size_t old_limit; | |
474 | size_t cp_size; | |
475 | struct mpls_route __rcu **labels = NULL, **old; | |
476 | struct mpls_route *rt0 = NULL, *rt2 = NULL; | |
477 | unsigned index; | |
478 | ||
479 | if (size) { | |
480 | labels = kzalloc(size, GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY); | |
481 | if (!labels) | |
482 | labels = vzalloc(size); | |
483 | ||
484 | if (!labels) | |
485 | goto nolabels; | |
486 | } | |
487 | ||
488 | /* In case the predefined labels need to be populated */ | |
489 | if (limit > LABEL_IPV4_EXPLICIT_NULL) { | |
490 | struct net_device *lo = net->loopback_dev; | |
491 | rt0 = mpls_rt_alloc(lo->addr_len); | |
492 | if (!rt0) | |
493 | goto nort0; | |
494 | rt0->rt_dev = lo; | |
495 | rt0->rt_protocol = RTPROT_KERNEL; | |
496 | rt0->rt_via_family = AF_PACKET; | |
497 | memcpy(rt0->rt_via, lo->dev_addr, lo->addr_len); | |
498 | } | |
499 | if (limit > LABEL_IPV6_EXPLICIT_NULL) { | |
500 | struct net_device *lo = net->loopback_dev; | |
501 | rt2 = mpls_rt_alloc(lo->addr_len); | |
502 | if (!rt2) | |
503 | goto nort2; | |
504 | rt2->rt_dev = lo; | |
505 | rt2->rt_protocol = RTPROT_KERNEL; | |
506 | rt2->rt_via_family = AF_PACKET; | |
507 | memcpy(rt2->rt_via, lo->dev_addr, lo->addr_len); | |
508 | } | |
509 | ||
510 | rtnl_lock(); | |
511 | /* Remember the original table */ | |
512 | old = net->mpls.platform_label; | |
513 | old_limit = net->mpls.platform_labels; | |
514 | ||
515 | /* Free any labels beyond the new table */ | |
516 | for (index = limit; index < old_limit; index++) | |
517 | mpls_route_update(net, index, NULL, NULL, NULL); | |
518 | ||
519 | /* Copy over the old labels */ | |
520 | cp_size = size; | |
521 | if (old_limit < limit) | |
522 | cp_size = old_limit * sizeof(struct mpls_route *); | |
523 | ||
524 | memcpy(labels, old, cp_size); | |
525 | ||
526 | /* If needed set the predefined labels */ | |
527 | if ((old_limit <= LABEL_IPV6_EXPLICIT_NULL) && | |
528 | (limit > LABEL_IPV6_EXPLICIT_NULL)) { | |
529 | labels[LABEL_IPV6_EXPLICIT_NULL] = rt2; | |
530 | rt2 = NULL; | |
531 | } | |
532 | ||
533 | if ((old_limit <= LABEL_IPV4_EXPLICIT_NULL) && | |
534 | (limit > LABEL_IPV4_EXPLICIT_NULL)) { | |
535 | labels[LABEL_IPV4_EXPLICIT_NULL] = rt0; | |
536 | rt0 = NULL; | |
537 | } | |
538 | ||
539 | /* Update the global pointers */ | |
540 | net->mpls.platform_labels = limit; | |
541 | net->mpls.platform_label = labels; | |
542 | ||
543 | rtnl_unlock(); | |
544 | ||
545 | mpls_rt_free(rt2); | |
546 | mpls_rt_free(rt0); | |
547 | ||
548 | if (old) { | |
549 | synchronize_rcu(); | |
550 | kvfree(old); | |
551 | } | |
552 | return 0; | |
553 | ||
554 | nort2: | |
555 | mpls_rt_free(rt0); | |
556 | nort0: | |
557 | kvfree(labels); | |
558 | nolabels: | |
559 | return -ENOMEM; | |
560 | } | |
561 | ||
562 | static int mpls_platform_labels(struct ctl_table *table, int write, | |
563 | void __user *buffer, size_t *lenp, loff_t *ppos) | |
564 | { | |
565 | struct net *net = table->data; | |
566 | int platform_labels = net->mpls.platform_labels; | |
567 | int ret; | |
568 | struct ctl_table tmp = { | |
569 | .procname = table->procname, | |
570 | .data = &platform_labels, | |
571 | .maxlen = sizeof(int), | |
572 | .mode = table->mode, | |
573 | .extra1 = &zero, | |
574 | .extra2 = &label_limit, | |
575 | }; | |
576 | ||
577 | ret = proc_dointvec_minmax(&tmp, write, buffer, lenp, ppos); | |
578 | ||
579 | if (write && ret == 0) | |
580 | ret = resize_platform_label_table(net, platform_labels); | |
581 | ||
582 | return ret; | |
583 | } | |
584 | ||
585 | static struct ctl_table mpls_table[] = { | |
586 | { | |
587 | .procname = "platform_labels", | |
588 | .data = NULL, | |
589 | .maxlen = sizeof(int), | |
590 | .mode = 0644, | |
591 | .proc_handler = mpls_platform_labels, | |
592 | }, | |
593 | { } | |
594 | }; | |
595 | ||
0189197f EB |
596 | static int mpls_net_init(struct net *net) |
597 | { | |
7720c01f EB |
598 | struct ctl_table *table; |
599 | ||
0189197f EB |
600 | net->mpls.platform_labels = 0; |
601 | net->mpls.platform_label = NULL; | |
602 | ||
7720c01f EB |
603 | table = kmemdup(mpls_table, sizeof(mpls_table), GFP_KERNEL); |
604 | if (table == NULL) | |
605 | return -ENOMEM; | |
606 | ||
607 | table[0].data = net; | |
608 | net->mpls.ctl = register_net_sysctl(net, "net/mpls", table); | |
609 | if (net->mpls.ctl == NULL) | |
610 | return -ENOMEM; | |
611 | ||
0189197f EB |
612 | return 0; |
613 | } | |
614 | ||
615 | static void mpls_net_exit(struct net *net) | |
616 | { | |
7720c01f | 617 | struct ctl_table *table; |
0189197f EB |
618 | unsigned int index; |
619 | ||
7720c01f EB |
620 | table = net->mpls.ctl->ctl_table_arg; |
621 | unregister_net_sysctl_table(net->mpls.ctl); | |
622 | kfree(table); | |
623 | ||
0189197f EB |
624 | /* An rcu grace period haselapsed since there was a device in |
625 | * the network namespace (and thus the last in fqlight packet) | |
626 | * left this network namespace. This is because | |
627 | * unregister_netdevice_many and netdev_run_todo has completed | |
628 | * for each network device that was in this network namespace. | |
629 | * | |
630 | * As such no additional rcu synchronization is necessary when | |
631 | * freeing the platform_label table. | |
632 | */ | |
633 | rtnl_lock(); | |
634 | for (index = 0; index < net->mpls.platform_labels; index++) { | |
635 | struct mpls_route *rt = net->mpls.platform_label[index]; | |
636 | rcu_assign_pointer(net->mpls.platform_label[index], NULL); | |
637 | mpls_rt_free(rt); | |
638 | } | |
639 | rtnl_unlock(); | |
640 | ||
641 | kvfree(net->mpls.platform_label); | |
642 | } | |
643 | ||
644 | static struct pernet_operations mpls_net_ops = { | |
645 | .init = mpls_net_init, | |
646 | .exit = mpls_net_exit, | |
647 | }; | |
648 | ||
649 | static int __init mpls_init(void) | |
650 | { | |
651 | int err; | |
652 | ||
653 | BUILD_BUG_ON(sizeof(struct mpls_shim_hdr) != 4); | |
654 | ||
655 | err = register_pernet_subsys(&mpls_net_ops); | |
656 | if (err) | |
657 | goto out; | |
658 | ||
659 | err = register_netdevice_notifier(&mpls_dev_notifier); | |
660 | if (err) | |
661 | goto out_unregister_pernet; | |
662 | ||
663 | dev_add_pack(&mpls_packet_type); | |
664 | ||
665 | err = 0; | |
666 | out: | |
667 | return err; | |
668 | ||
669 | out_unregister_pernet: | |
670 | unregister_pernet_subsys(&mpls_net_ops); | |
671 | goto out; | |
672 | } | |
673 | module_init(mpls_init); | |
674 | ||
675 | static void __exit mpls_exit(void) | |
676 | { | |
677 | dev_remove_pack(&mpls_packet_type); | |
678 | unregister_netdevice_notifier(&mpls_dev_notifier); | |
679 | unregister_pernet_subsys(&mpls_net_ops); | |
680 | } | |
681 | module_exit(mpls_exit); | |
682 | ||
683 | MODULE_DESCRIPTION("MultiProtocol Label Switching"); | |
684 | MODULE_LICENSE("GPL v2"); | |
685 | MODULE_ALIAS_NETPROTO(PF_MPLS); |