Loading...
1// SPDX-License-Identifier: GPL-2.0-only
2/*
3 * net/core/dst.c Protocol independent destination cache.
4 *
5 * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
6 *
7 */
8
9#include <linux/bitops.h>
10#include <linux/errno.h>
11#include <linux/init.h>
12#include <linux/kernel.h>
13#include <linux/workqueue.h>
14#include <linux/mm.h>
15#include <linux/module.h>
16#include <linux/slab.h>
17#include <linux/netdevice.h>
18#include <linux/skbuff.h>
19#include <linux/string.h>
20#include <linux/types.h>
21#include <net/net_namespace.h>
22#include <linux/sched.h>
23#include <linux/prefetch.h>
24#include <net/lwtunnel.h>
25#include <net/xfrm.h>
26
27#include <net/dst.h>
28#include <net/dst_metadata.h>
29
30int dst_discard_out(struct net *net, struct sock *sk, struct sk_buff *skb)
31{
32 kfree_skb(skb);
33 return 0;
34}
35EXPORT_SYMBOL(dst_discard_out);
36
37const struct dst_metrics dst_default_metrics = {
38 /* This initializer is needed to force linker to place this variable
39 * into const section. Otherwise it might end into bss section.
40 * We really want to avoid false sharing on this variable, and catch
41 * any writes on it.
42 */
43 .refcnt = REFCOUNT_INIT(1),
44};
45EXPORT_SYMBOL(dst_default_metrics);
46
47void dst_init(struct dst_entry *dst, struct dst_ops *ops,
48 struct net_device *dev, int initial_obsolete,
49 unsigned short flags)
50{
51 dst->dev = dev;
52 netdev_hold(dev, &dst->dev_tracker, GFP_ATOMIC);
53 dst->ops = ops;
54 dst_init_metrics(dst, dst_default_metrics.metrics, true);
55 dst->expires = 0UL;
56#ifdef CONFIG_XFRM
57 dst->xfrm = NULL;
58#endif
59 dst->input = dst_discard;
60 dst->output = dst_discard_out;
61 dst->error = 0;
62 dst->obsolete = initial_obsolete;
63 dst->header_len = 0;
64 dst->trailer_len = 0;
65#ifdef CONFIG_IP_ROUTE_CLASSID
66 dst->tclassid = 0;
67#endif
68 dst->lwtstate = NULL;
69 rcuref_init(&dst->__rcuref, 1);
70 INIT_LIST_HEAD(&dst->rt_uncached);
71 dst->__use = 0;
72 dst->lastuse = jiffies;
73 dst->flags = flags;
74 if (!(flags & DST_NOCOUNT))
75 dst_entries_add(ops, 1);
76}
77EXPORT_SYMBOL(dst_init);
78
79void *dst_alloc(struct dst_ops *ops, struct net_device *dev,
80 int initial_obsolete, unsigned short flags)
81{
82 struct dst_entry *dst;
83
84 if (ops->gc &&
85 !(flags & DST_NOCOUNT) &&
86 dst_entries_get_fast(ops) > ops->gc_thresh)
87 ops->gc(ops);
88
89 dst = kmem_cache_alloc(ops->kmem_cachep, GFP_ATOMIC);
90 if (!dst)
91 return NULL;
92
93 dst_init(dst, ops, dev, initial_obsolete, flags);
94
95 return dst;
96}
97EXPORT_SYMBOL(dst_alloc);
98
99static void dst_destroy(struct dst_entry *dst)
100{
101 struct dst_entry *child = NULL;
102
103 smp_rmb();
104
105#ifdef CONFIG_XFRM
106 if (dst->xfrm) {
107 struct xfrm_dst *xdst = (struct xfrm_dst *) dst;
108
109 child = xdst->child;
110 }
111#endif
112 if (!(dst->flags & DST_NOCOUNT))
113 dst_entries_add(dst->ops, -1);
114
115 if (dst->ops->destroy)
116 dst->ops->destroy(dst);
117 netdev_put(dst->dev, &dst->dev_tracker);
118
119 lwtstate_put(dst->lwtstate);
120
121 if (dst->flags & DST_METADATA)
122 metadata_dst_free((struct metadata_dst *)dst);
123 else
124 kmem_cache_free(dst->ops->kmem_cachep, dst);
125
126 dst = child;
127 if (dst)
128 dst_release_immediate(dst);
129}
130
131static void dst_destroy_rcu(struct rcu_head *head)
132{
133 struct dst_entry *dst = container_of(head, struct dst_entry, rcu_head);
134
135 dst_destroy(dst);
136}
137
138/* Operations to mark dst as DEAD and clean up the net device referenced
139 * by dst:
140 * 1. put the dst under blackhole interface and discard all tx/rx packets
141 * on this route.
142 * 2. release the net_device
143 * This function should be called when removing routes from the fib tree
144 * in preparation for a NETDEV_DOWN/NETDEV_UNREGISTER event and also to
145 * make the next dst_ops->check() fail.
146 */
147void dst_dev_put(struct dst_entry *dst)
148{
149 struct net_device *dev = dst->dev;
150
151 dst->obsolete = DST_OBSOLETE_DEAD;
152 if (dst->ops->ifdown)
153 dst->ops->ifdown(dst, dev);
154 dst->input = dst_discard;
155 dst->output = dst_discard_out;
156 dst->dev = blackhole_netdev;
157 netdev_ref_replace(dev, blackhole_netdev, &dst->dev_tracker,
158 GFP_ATOMIC);
159}
160EXPORT_SYMBOL(dst_dev_put);
161
162void dst_release(struct dst_entry *dst)
163{
164 if (dst && rcuref_put(&dst->__rcuref))
165 call_rcu_hurry(&dst->rcu_head, dst_destroy_rcu);
166}
167EXPORT_SYMBOL(dst_release);
168
169void dst_release_immediate(struct dst_entry *dst)
170{
171 if (dst && rcuref_put(&dst->__rcuref))
172 dst_destroy(dst);
173}
174EXPORT_SYMBOL(dst_release_immediate);
175
176u32 *dst_cow_metrics_generic(struct dst_entry *dst, unsigned long old)
177{
178 struct dst_metrics *p = kmalloc(sizeof(*p), GFP_ATOMIC);
179
180 if (p) {
181 struct dst_metrics *old_p = (struct dst_metrics *)__DST_METRICS_PTR(old);
182 unsigned long prev, new;
183
184 refcount_set(&p->refcnt, 1);
185 memcpy(p->metrics, old_p->metrics, sizeof(p->metrics));
186
187 new = (unsigned long) p;
188 prev = cmpxchg(&dst->_metrics, old, new);
189
190 if (prev != old) {
191 kfree(p);
192 p = (struct dst_metrics *)__DST_METRICS_PTR(prev);
193 if (prev & DST_METRICS_READ_ONLY)
194 p = NULL;
195 } else if (prev & DST_METRICS_REFCOUNTED) {
196 if (refcount_dec_and_test(&old_p->refcnt))
197 kfree(old_p);
198 }
199 }
200 BUILD_BUG_ON(offsetof(struct dst_metrics, metrics) != 0);
201 return (u32 *)p;
202}
203EXPORT_SYMBOL(dst_cow_metrics_generic);
204
205/* Caller asserts that dst_metrics_read_only(dst) is false. */
206void __dst_destroy_metrics_generic(struct dst_entry *dst, unsigned long old)
207{
208 unsigned long prev, new;
209
210 new = ((unsigned long) &dst_default_metrics) | DST_METRICS_READ_ONLY;
211 prev = cmpxchg(&dst->_metrics, old, new);
212 if (prev == old)
213 kfree(__DST_METRICS_PTR(old));
214}
215EXPORT_SYMBOL(__dst_destroy_metrics_generic);
216
217struct dst_entry *dst_blackhole_check(struct dst_entry *dst, u32 cookie)
218{
219 return NULL;
220}
221
222u32 *dst_blackhole_cow_metrics(struct dst_entry *dst, unsigned long old)
223{
224 return NULL;
225}
226
227struct neighbour *dst_blackhole_neigh_lookup(const struct dst_entry *dst,
228 struct sk_buff *skb,
229 const void *daddr)
230{
231 return NULL;
232}
233
234void dst_blackhole_update_pmtu(struct dst_entry *dst, struct sock *sk,
235 struct sk_buff *skb, u32 mtu,
236 bool confirm_neigh)
237{
238}
239EXPORT_SYMBOL_GPL(dst_blackhole_update_pmtu);
240
241void dst_blackhole_redirect(struct dst_entry *dst, struct sock *sk,
242 struct sk_buff *skb)
243{
244}
245EXPORT_SYMBOL_GPL(dst_blackhole_redirect);
246
247unsigned int dst_blackhole_mtu(const struct dst_entry *dst)
248{
249 unsigned int mtu = dst_metric_raw(dst, RTAX_MTU);
250
251 return mtu ? : dst->dev->mtu;
252}
253EXPORT_SYMBOL_GPL(dst_blackhole_mtu);
254
255static struct dst_ops dst_blackhole_ops = {
256 .family = AF_UNSPEC,
257 .neigh_lookup = dst_blackhole_neigh_lookup,
258 .check = dst_blackhole_check,
259 .cow_metrics = dst_blackhole_cow_metrics,
260 .update_pmtu = dst_blackhole_update_pmtu,
261 .redirect = dst_blackhole_redirect,
262 .mtu = dst_blackhole_mtu,
263};
264
265static void __metadata_dst_init(struct metadata_dst *md_dst,
266 enum metadata_type type, u8 optslen)
267{
268 struct dst_entry *dst;
269
270 dst = &md_dst->dst;
271 dst_init(dst, &dst_blackhole_ops, NULL, DST_OBSOLETE_NONE,
272 DST_METADATA | DST_NOCOUNT);
273 memset(dst + 1, 0, sizeof(*md_dst) + optslen - sizeof(*dst));
274 md_dst->type = type;
275}
276
277struct metadata_dst *metadata_dst_alloc(u8 optslen, enum metadata_type type,
278 gfp_t flags)
279{
280 struct metadata_dst *md_dst;
281
282 md_dst = kmalloc(sizeof(*md_dst) + optslen, flags);
283 if (!md_dst)
284 return NULL;
285
286 __metadata_dst_init(md_dst, type, optslen);
287
288 return md_dst;
289}
290EXPORT_SYMBOL_GPL(metadata_dst_alloc);
291
292void metadata_dst_free(struct metadata_dst *md_dst)
293{
294#ifdef CONFIG_DST_CACHE
295 if (md_dst->type == METADATA_IP_TUNNEL)
296 dst_cache_destroy(&md_dst->u.tun_info.dst_cache);
297#endif
298 if (md_dst->type == METADATA_XFRM)
299 dst_release(md_dst->u.xfrm_info.dst_orig);
300 kfree(md_dst);
301}
302EXPORT_SYMBOL_GPL(metadata_dst_free);
303
304struct metadata_dst __percpu *
305metadata_dst_alloc_percpu(u8 optslen, enum metadata_type type, gfp_t flags)
306{
307 int cpu;
308 struct metadata_dst __percpu *md_dst;
309
310 md_dst = __alloc_percpu_gfp(sizeof(struct metadata_dst) + optslen,
311 __alignof__(struct metadata_dst), flags);
312 if (!md_dst)
313 return NULL;
314
315 for_each_possible_cpu(cpu)
316 __metadata_dst_init(per_cpu_ptr(md_dst, cpu), type, optslen);
317
318 return md_dst;
319}
320EXPORT_SYMBOL_GPL(metadata_dst_alloc_percpu);
321
322void metadata_dst_free_percpu(struct metadata_dst __percpu *md_dst)
323{
324 int cpu;
325
326 for_each_possible_cpu(cpu) {
327 struct metadata_dst *one_md_dst = per_cpu_ptr(md_dst, cpu);
328
329#ifdef CONFIG_DST_CACHE
330 if (one_md_dst->type == METADATA_IP_TUNNEL)
331 dst_cache_destroy(&one_md_dst->u.tun_info.dst_cache);
332#endif
333 if (one_md_dst->type == METADATA_XFRM)
334 dst_release(one_md_dst->u.xfrm_info.dst_orig);
335 }
336 free_percpu(md_dst);
337}
338EXPORT_SYMBOL_GPL(metadata_dst_free_percpu);
1/*
2 * net/core/dst.c Protocol independent destination cache.
3 *
4 * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
5 *
6 */
7
8#include <linux/bitops.h>
9#include <linux/errno.h>
10#include <linux/init.h>
11#include <linux/kernel.h>
12#include <linux/workqueue.h>
13#include <linux/mm.h>
14#include <linux/module.h>
15#include <linux/slab.h>
16#include <linux/netdevice.h>
17#include <linux/skbuff.h>
18#include <linux/string.h>
19#include <linux/types.h>
20#include <net/net_namespace.h>
21#include <linux/sched.h>
22#include <linux/prefetch.h>
23#include <net/lwtunnel.h>
24#include <net/xfrm.h>
25
26#include <net/dst.h>
27#include <net/dst_metadata.h>
28
29/*
30 * Theory of operations:
31 * 1) We use a list, protected by a spinlock, to add
32 * new entries from both BH and non-BH context.
33 * 2) In order to keep spinlock held for a small delay,
34 * we use a second list where are stored long lived
35 * entries, that are handled by the garbage collect thread
36 * fired by a workqueue.
37 * 3) This list is guarded by a mutex,
38 * so that the gc_task and dst_dev_event() can be synchronized.
39 */
40
41/*
42 * We want to keep lock & list close together
43 * to dirty as few cache lines as possible in __dst_free().
44 * As this is not a very strong hint, we dont force an alignment on SMP.
45 */
46int dst_discard_out(struct net *net, struct sock *sk, struct sk_buff *skb)
47{
48 kfree_skb(skb);
49 return 0;
50}
51EXPORT_SYMBOL(dst_discard_out);
52
53const struct dst_metrics dst_default_metrics = {
54 /* This initializer is needed to force linker to place this variable
55 * into const section. Otherwise it might end into bss section.
56 * We really want to avoid false sharing on this variable, and catch
57 * any writes on it.
58 */
59 .refcnt = REFCOUNT_INIT(1),
60};
61
62void dst_init(struct dst_entry *dst, struct dst_ops *ops,
63 struct net_device *dev, int initial_ref, int initial_obsolete,
64 unsigned short flags)
65{
66 dst->dev = dev;
67 if (dev)
68 dev_hold(dev);
69 dst->ops = ops;
70 dst_init_metrics(dst, dst_default_metrics.metrics, true);
71 dst->expires = 0UL;
72#ifdef CONFIG_XFRM
73 dst->xfrm = NULL;
74#endif
75 dst->input = dst_discard;
76 dst->output = dst_discard_out;
77 dst->error = 0;
78 dst->obsolete = initial_obsolete;
79 dst->header_len = 0;
80 dst->trailer_len = 0;
81#ifdef CONFIG_IP_ROUTE_CLASSID
82 dst->tclassid = 0;
83#endif
84 dst->lwtstate = NULL;
85 atomic_set(&dst->__refcnt, initial_ref);
86 dst->__use = 0;
87 dst->lastuse = jiffies;
88 dst->flags = flags;
89 if (!(flags & DST_NOCOUNT))
90 dst_entries_add(ops, 1);
91}
92EXPORT_SYMBOL(dst_init);
93
94void *dst_alloc(struct dst_ops *ops, struct net_device *dev,
95 int initial_ref, int initial_obsolete, unsigned short flags)
96{
97 struct dst_entry *dst;
98
99 if (ops->gc && dst_entries_get_fast(ops) > ops->gc_thresh) {
100 if (ops->gc(ops))
101 return NULL;
102 }
103
104 dst = kmem_cache_alloc(ops->kmem_cachep, GFP_ATOMIC);
105 if (!dst)
106 return NULL;
107
108 dst_init(dst, ops, dev, initial_ref, initial_obsolete, flags);
109
110 return dst;
111}
112EXPORT_SYMBOL(dst_alloc);
113
114struct dst_entry *dst_destroy(struct dst_entry * dst)
115{
116 struct dst_entry *child = NULL;
117
118 smp_rmb();
119
120#ifdef CONFIG_XFRM
121 if (dst->xfrm) {
122 struct xfrm_dst *xdst = (struct xfrm_dst *) dst;
123
124 child = xdst->child;
125 }
126#endif
127 if (!(dst->flags & DST_NOCOUNT))
128 dst_entries_add(dst->ops, -1);
129
130 if (dst->ops->destroy)
131 dst->ops->destroy(dst);
132 if (dst->dev)
133 dev_put(dst->dev);
134
135 lwtstate_put(dst->lwtstate);
136
137 if (dst->flags & DST_METADATA)
138 metadata_dst_free((struct metadata_dst *)dst);
139 else
140 kmem_cache_free(dst->ops->kmem_cachep, dst);
141
142 dst = child;
143 if (dst)
144 dst_release_immediate(dst);
145 return NULL;
146}
147EXPORT_SYMBOL(dst_destroy);
148
149static void dst_destroy_rcu(struct rcu_head *head)
150{
151 struct dst_entry *dst = container_of(head, struct dst_entry, rcu_head);
152
153 dst = dst_destroy(dst);
154}
155
156/* Operations to mark dst as DEAD and clean up the net device referenced
157 * by dst:
158 * 1. put the dst under loopback interface and discard all tx/rx packets
159 * on this route.
160 * 2. release the net_device
161 * This function should be called when removing routes from the fib tree
162 * in preparation for a NETDEV_DOWN/NETDEV_UNREGISTER event and also to
163 * make the next dst_ops->check() fail.
164 */
165void dst_dev_put(struct dst_entry *dst)
166{
167 struct net_device *dev = dst->dev;
168
169 dst->obsolete = DST_OBSOLETE_DEAD;
170 if (dst->ops->ifdown)
171 dst->ops->ifdown(dst, dev, true);
172 dst->input = dst_discard;
173 dst->output = dst_discard_out;
174 dst->dev = dev_net(dst->dev)->loopback_dev;
175 dev_hold(dst->dev);
176 dev_put(dev);
177}
178EXPORT_SYMBOL(dst_dev_put);
179
180void dst_release(struct dst_entry *dst)
181{
182 if (dst) {
183 int newrefcnt;
184
185 newrefcnt = atomic_dec_return(&dst->__refcnt);
186 if (unlikely(newrefcnt < 0))
187 net_warn_ratelimited("%s: dst:%p refcnt:%d\n",
188 __func__, dst, newrefcnt);
189 if (!newrefcnt)
190 call_rcu(&dst->rcu_head, dst_destroy_rcu);
191 }
192}
193EXPORT_SYMBOL(dst_release);
194
195void dst_release_immediate(struct dst_entry *dst)
196{
197 if (dst) {
198 int newrefcnt;
199
200 newrefcnt = atomic_dec_return(&dst->__refcnt);
201 if (unlikely(newrefcnt < 0))
202 net_warn_ratelimited("%s: dst:%p refcnt:%d\n",
203 __func__, dst, newrefcnt);
204 if (!newrefcnt)
205 dst_destroy(dst);
206 }
207}
208EXPORT_SYMBOL(dst_release_immediate);
209
210u32 *dst_cow_metrics_generic(struct dst_entry *dst, unsigned long old)
211{
212 struct dst_metrics *p = kmalloc(sizeof(*p), GFP_ATOMIC);
213
214 if (p) {
215 struct dst_metrics *old_p = (struct dst_metrics *)__DST_METRICS_PTR(old);
216 unsigned long prev, new;
217
218 refcount_set(&p->refcnt, 1);
219 memcpy(p->metrics, old_p->metrics, sizeof(p->metrics));
220
221 new = (unsigned long) p;
222 prev = cmpxchg(&dst->_metrics, old, new);
223
224 if (prev != old) {
225 kfree(p);
226 p = (struct dst_metrics *)__DST_METRICS_PTR(prev);
227 if (prev & DST_METRICS_READ_ONLY)
228 p = NULL;
229 } else if (prev & DST_METRICS_REFCOUNTED) {
230 if (refcount_dec_and_test(&old_p->refcnt))
231 kfree(old_p);
232 }
233 }
234 BUILD_BUG_ON(offsetof(struct dst_metrics, metrics) != 0);
235 return (u32 *)p;
236}
237EXPORT_SYMBOL(dst_cow_metrics_generic);
238
239/* Caller asserts that dst_metrics_read_only(dst) is false. */
240void __dst_destroy_metrics_generic(struct dst_entry *dst, unsigned long old)
241{
242 unsigned long prev, new;
243
244 new = ((unsigned long) &dst_default_metrics) | DST_METRICS_READ_ONLY;
245 prev = cmpxchg(&dst->_metrics, old, new);
246 if (prev == old)
247 kfree(__DST_METRICS_PTR(old));
248}
249EXPORT_SYMBOL(__dst_destroy_metrics_generic);
250
251static struct dst_ops md_dst_ops = {
252 .family = AF_UNSPEC,
253};
254
255static int dst_md_discard_out(struct net *net, struct sock *sk, struct sk_buff *skb)
256{
257 WARN_ONCE(1, "Attempting to call output on metadata dst\n");
258 kfree_skb(skb);
259 return 0;
260}
261
262static int dst_md_discard(struct sk_buff *skb)
263{
264 WARN_ONCE(1, "Attempting to call input on metadata dst\n");
265 kfree_skb(skb);
266 return 0;
267}
268
269static void __metadata_dst_init(struct metadata_dst *md_dst,
270 enum metadata_type type, u8 optslen)
271
272{
273 struct dst_entry *dst;
274
275 dst = &md_dst->dst;
276 dst_init(dst, &md_dst_ops, NULL, 1, DST_OBSOLETE_NONE,
277 DST_METADATA | DST_NOCOUNT);
278
279 dst->input = dst_md_discard;
280 dst->output = dst_md_discard_out;
281
282 memset(dst + 1, 0, sizeof(*md_dst) + optslen - sizeof(*dst));
283 md_dst->type = type;
284}
285
286struct metadata_dst *metadata_dst_alloc(u8 optslen, enum metadata_type type,
287 gfp_t flags)
288{
289 struct metadata_dst *md_dst;
290
291 md_dst = kmalloc(sizeof(*md_dst) + optslen, flags);
292 if (!md_dst)
293 return NULL;
294
295 __metadata_dst_init(md_dst, type, optslen);
296
297 return md_dst;
298}
299EXPORT_SYMBOL_GPL(metadata_dst_alloc);
300
301void metadata_dst_free(struct metadata_dst *md_dst)
302{
303#ifdef CONFIG_DST_CACHE
304 if (md_dst->type == METADATA_IP_TUNNEL)
305 dst_cache_destroy(&md_dst->u.tun_info.dst_cache);
306#endif
307 kfree(md_dst);
308}
309
310struct metadata_dst __percpu *
311metadata_dst_alloc_percpu(u8 optslen, enum metadata_type type, gfp_t flags)
312{
313 int cpu;
314 struct metadata_dst __percpu *md_dst;
315
316 md_dst = __alloc_percpu_gfp(sizeof(struct metadata_dst) + optslen,
317 __alignof__(struct metadata_dst), flags);
318 if (!md_dst)
319 return NULL;
320
321 for_each_possible_cpu(cpu)
322 __metadata_dst_init(per_cpu_ptr(md_dst, cpu), type, optslen);
323
324 return md_dst;
325}
326EXPORT_SYMBOL_GPL(metadata_dst_alloc_percpu);
327
328void metadata_dst_free_percpu(struct metadata_dst __percpu *md_dst)
329{
330#ifdef CONFIG_DST_CACHE
331 int cpu;
332
333 for_each_possible_cpu(cpu) {
334 struct metadata_dst *one_md_dst = per_cpu_ptr(md_dst, cpu);
335
336 if (one_md_dst->type == METADATA_IP_TUNNEL)
337 dst_cache_destroy(&one_md_dst->u.tun_info.dst_cache);
338 }
339#endif
340 free_percpu(md_dst);
341}
342EXPORT_SYMBOL_GPL(metadata_dst_free_percpu);