1 /* SPDX-License-Identifier: GPL-2.0 */
2 #ifndef __NET_UDP_TUNNEL_H
3 #define __NET_UDP_TUNNEL_H
4
5 #include <net/ip_tunnels.h>
6 #include <net/udp.h>
7
8 #if IS_ENABLED(CONFIG_IPV6)
9 #include <net/ipv6.h>
10 #include <net/ipv6_stubs.h>
11 #endif
12
13 struct udp_port_cfg {
14 u8 family;
15
16 /* Used only for kernel-created sockets */
17 union {
18 struct in_addr local_ip;
19 #if IS_ENABLED(CONFIG_IPV6)
20 struct in6_addr local_ip6;
21 #endif
22 };
23
24 union {
25 struct in_addr peer_ip;
26 #if IS_ENABLED(CONFIG_IPV6)
27 struct in6_addr peer_ip6;
28 #endif
29 };
30
31 __be16 local_udp_port;
32 __be16 peer_udp_port;
33 int bind_ifindex;
34 unsigned int use_udp_checksums:1,
35 use_udp6_tx_checksums:1,
36 use_udp6_rx_checksums:1,
37 ipv6_v6only:1;
38 };
39
40 int udp_sock_create4(struct net *net, struct udp_port_cfg *cfg,
41 struct socket **sockp);
42
43 #if IS_ENABLED(CONFIG_IPV6)
44 int udp_sock_create6(struct net *net, struct udp_port_cfg *cfg,
45 struct socket **sockp);
46 #else
udp_sock_create6(struct net * net,struct udp_port_cfg * cfg,struct socket ** sockp)47 static inline int udp_sock_create6(struct net *net, struct udp_port_cfg *cfg,
48 struct socket **sockp)
49 {
50 return 0;
51 }
52 #endif
53
udp_sock_create(struct net * net,struct udp_port_cfg * cfg,struct socket ** sockp)54 static inline int udp_sock_create(struct net *net,
55 struct udp_port_cfg *cfg,
56 struct socket **sockp)
57 {
58 if (cfg->family == AF_INET)
59 return udp_sock_create4(net, cfg, sockp);
60
61 if (cfg->family == AF_INET6)
62 return udp_sock_create6(net, cfg, sockp);
63
64 return -EPFNOSUPPORT;
65 }
66
67 typedef int (*udp_tunnel_encap_rcv_t)(struct sock *sk, struct sk_buff *skb);
68 typedef int (*udp_tunnel_encap_err_lookup_t)(struct sock *sk,
69 struct sk_buff *skb);
70 typedef void (*udp_tunnel_encap_err_rcv_t)(struct sock *sk,
71 struct sk_buff *skb, int err,
72 __be16 port, u32 info, u8 *payload);
73 typedef void (*udp_tunnel_encap_destroy_t)(struct sock *sk);
74 typedef struct sk_buff *(*udp_tunnel_gro_receive_t)(struct sock *sk,
75 struct list_head *head,
76 struct sk_buff *skb);
77 typedef int (*udp_tunnel_gro_complete_t)(struct sock *sk, struct sk_buff *skb,
78 int nhoff);
79
80 struct udp_tunnel_sock_cfg {
81 void *sk_user_data; /* user data used by encap_rcv call back */
82 /* Used for setting up udp_sock fields, see udp.h for details */
83 __u8 encap_type;
84 udp_tunnel_encap_rcv_t encap_rcv;
85 udp_tunnel_encap_err_lookup_t encap_err_lookup;
86 udp_tunnel_encap_err_rcv_t encap_err_rcv;
87 udp_tunnel_encap_destroy_t encap_destroy;
88 udp_tunnel_gro_receive_t gro_receive;
89 udp_tunnel_gro_complete_t gro_complete;
90 };
91
92 /* Setup the given (UDP) sock to receive UDP encapsulated packets */
93 void setup_udp_tunnel_sock(struct net *net, struct socket *sock,
94 struct udp_tunnel_sock_cfg *sock_cfg);
95
96 /* -- List of parsable UDP tunnel types --
97 *
98 * Adding to this list will result in serious debate. The main issue is
99 * that this list is essentially a list of workarounds for either poorly
100 * designed tunnels, or poorly designed device offloads.
101 *
102 * The parsing supported via these types should really be used for Rx
103 * traffic only as the network stack will have already inserted offsets for
104 * the location of the headers in the skb. In addition any ports that are
105 * pushed should be kept within the namespace without leaking to other
106 * devices such as VFs or other ports on the same device.
107 *
108 * It is strongly encouraged to use CHECKSUM_COMPLETE for Rx to avoid the
109 * need to use this for Rx checksum offload. It should not be necessary to
110 * call this function to perform Tx offloads on outgoing traffic.
111 */
112 enum udp_parsable_tunnel_type {
113 UDP_TUNNEL_TYPE_VXLAN = BIT(0), /* RFC 7348 */
114 UDP_TUNNEL_TYPE_GENEVE = BIT(1), /* draft-ietf-nvo3-geneve */
115 UDP_TUNNEL_TYPE_VXLAN_GPE = BIT(2), /* draft-ietf-nvo3-vxlan-gpe */
116 };
117
118 struct udp_tunnel_info {
119 unsigned short type;
120 sa_family_t sa_family;
121 __be16 port;
122 u8 hw_priv;
123 };
124
125 /* Notify network devices of offloadable types */
126 void udp_tunnel_push_rx_port(struct net_device *dev, struct socket *sock,
127 unsigned short type);
128 void udp_tunnel_drop_rx_port(struct net_device *dev, struct socket *sock,
129 unsigned short type);
130 void udp_tunnel_notify_add_rx_port(struct socket *sock, unsigned short type);
131 void udp_tunnel_notify_del_rx_port(struct socket *sock, unsigned short type);
132
udp_tunnel_get_rx_info(struct net_device * dev)133 static inline void udp_tunnel_get_rx_info(struct net_device *dev)
134 {
135 ASSERT_RTNL();
136 if (!(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT))
137 return;
138 call_netdevice_notifiers(NETDEV_UDP_TUNNEL_PUSH_INFO, dev);
139 }
140
udp_tunnel_drop_rx_info(struct net_device * dev)141 static inline void udp_tunnel_drop_rx_info(struct net_device *dev)
142 {
143 ASSERT_RTNL();
144 if (!(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT))
145 return;
146 call_netdevice_notifiers(NETDEV_UDP_TUNNEL_DROP_INFO, dev);
147 }
148
149 /* Transmit the skb using UDP encapsulation. */
150 void udp_tunnel_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *skb,
151 __be32 src, __be32 dst, __u8 tos, __u8 ttl,
152 __be16 df, __be16 src_port, __be16 dst_port,
153 bool xnet, bool nocheck);
154
155 int udp_tunnel6_xmit_skb(struct dst_entry *dst, struct sock *sk,
156 struct sk_buff *skb,
157 struct net_device *dev, struct in6_addr *saddr,
158 struct in6_addr *daddr,
159 __u8 prio, __u8 ttl, __be32 label,
160 __be16 src_port, __be16 dst_port, bool nocheck);
161
162 void udp_tunnel_sock_release(struct socket *sock);
163
164 struct metadata_dst *udp_tun_rx_dst(struct sk_buff *skb, unsigned short family,
165 __be16 flags, __be64 tunnel_id,
166 int md_size);
167
168 #ifdef CONFIG_INET
udp_tunnel_handle_offloads(struct sk_buff * skb,bool udp_csum)169 static inline int udp_tunnel_handle_offloads(struct sk_buff *skb, bool udp_csum)
170 {
171 int type = udp_csum ? SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL;
172
173 return iptunnel_handle_offloads(skb, type);
174 }
175 #endif
176
udp_tunnel_encap_enable(struct socket * sock)177 static inline void udp_tunnel_encap_enable(struct socket *sock)
178 {
179 struct udp_sock *up = udp_sk(sock->sk);
180
181 if (up->encap_enabled)
182 return;
183
184 up->encap_enabled = 1;
185 #if IS_ENABLED(CONFIG_IPV6)
186 if (sock->sk->sk_family == PF_INET6)
187 ipv6_stub->udpv6_encap_enable();
188 #endif
189 udp_encap_enable();
190 }
191
192 #define UDP_TUNNEL_NIC_MAX_TABLES 4
193
194 enum udp_tunnel_nic_info_flags {
195 /* Device callbacks may sleep */
196 UDP_TUNNEL_NIC_INFO_MAY_SLEEP = BIT(0),
197 /* Device only supports offloads when it's open, all ports
198 * will be removed before close and re-added after open.
199 */
200 UDP_TUNNEL_NIC_INFO_OPEN_ONLY = BIT(1),
201 /* Device supports only IPv4 tunnels */
202 UDP_TUNNEL_NIC_INFO_IPV4_ONLY = BIT(2),
203 /* Device has hard-coded the IANA VXLAN port (4789) as VXLAN.
204 * This port must not be counted towards n_entries of any table.
205 * Driver will not receive any callback associated with port 4789.
206 */
207 UDP_TUNNEL_NIC_INFO_STATIC_IANA_VXLAN = BIT(3),
208 };
209
210 struct udp_tunnel_nic;
211
212 #define UDP_TUNNEL_NIC_MAX_SHARING_DEVICES (U16_MAX / 2)
213
214 struct udp_tunnel_nic_shared {
215 struct udp_tunnel_nic *udp_tunnel_nic_info;
216
217 struct list_head devices;
218 };
219
220 struct udp_tunnel_nic_shared_node {
221 struct net_device *dev;
222 struct list_head list;
223 };
224
225 /**
226 * struct udp_tunnel_nic_info - driver UDP tunnel offload information
227 * @set_port: callback for adding a new port
228 * @unset_port: callback for removing a port
229 * @sync_table: callback for syncing the entire port table at once
230 * @shared: reference to device global state (optional)
231 * @flags: device flags from enum udp_tunnel_nic_info_flags
232 * @tables: UDP port tables this device has
233 * @tables.n_entries: number of entries in this table
234 * @tables.tunnel_types: types of tunnels this table accepts
235 *
236 * Drivers are expected to provide either @set_port and @unset_port callbacks
237 * or the @sync_table callback. Callbacks are invoked with rtnl lock held.
238 *
239 * Devices which (misguidedly) share the UDP tunnel port table across multiple
240 * netdevs should allocate an instance of struct udp_tunnel_nic_shared and
241 * point @shared at it.
242 * There must never be more than %UDP_TUNNEL_NIC_MAX_SHARING_DEVICES devices
243 * sharing a table.
244 *
245 * Known limitations:
246 * - UDP tunnel port notifications are fundamentally best-effort -
247 * it is likely the driver will both see skbs which use a UDP tunnel port,
248 * while not being a tunneled skb, and tunnel skbs from other ports -
249 * drivers should only use these ports for non-critical RX-side offloads,
250 * e.g. the checksum offload;
251 * - none of the devices care about the socket family at present, so we don't
252 * track it. Please extend this code if you care.
253 */
254 struct udp_tunnel_nic_info {
255 /* one-by-one */
256 int (*set_port)(struct net_device *dev,
257 unsigned int table, unsigned int entry,
258 struct udp_tunnel_info *ti);
259 int (*unset_port)(struct net_device *dev,
260 unsigned int table, unsigned int entry,
261 struct udp_tunnel_info *ti);
262
263 /* all at once */
264 int (*sync_table)(struct net_device *dev, unsigned int table);
265
266 struct udp_tunnel_nic_shared *shared;
267
268 unsigned int flags;
269
270 struct udp_tunnel_nic_table_info {
271 unsigned int n_entries;
272 unsigned int tunnel_types;
273 } tables[UDP_TUNNEL_NIC_MAX_TABLES];
274 };
275
276 /* UDP tunnel module dependencies
277 *
278 * Tunnel drivers are expected to have a hard dependency on the udp_tunnel
279 * module. NIC drivers are not, they just attach their
280 * struct udp_tunnel_nic_info to the netdev and wait for callbacks to come.
281 * Loading a tunnel driver will cause the udp_tunnel module to be loaded
282 * and only then will all the required state structures be allocated.
283 * Since we want a weak dependency from the drivers and the core to udp_tunnel
284 * we call things through the following stubs.
285 */
286 struct udp_tunnel_nic_ops {
287 void (*get_port)(struct net_device *dev, unsigned int table,
288 unsigned int idx, struct udp_tunnel_info *ti);
289 void (*set_port_priv)(struct net_device *dev, unsigned int table,
290 unsigned int idx, u8 priv);
291 void (*add_port)(struct net_device *dev, struct udp_tunnel_info *ti);
292 void (*del_port)(struct net_device *dev, struct udp_tunnel_info *ti);
293 void (*reset_ntf)(struct net_device *dev);
294
295 size_t (*dump_size)(struct net_device *dev, unsigned int table);
296 int (*dump_write)(struct net_device *dev, unsigned int table,
297 struct sk_buff *skb);
298 };
299
300 #ifdef CONFIG_INET
301 extern const struct udp_tunnel_nic_ops *udp_tunnel_nic_ops;
302 #else
303 #define udp_tunnel_nic_ops ((struct udp_tunnel_nic_ops *)NULL)
304 #endif
305
306 static inline void
udp_tunnel_nic_get_port(struct net_device * dev,unsigned int table,unsigned int idx,struct udp_tunnel_info * ti)307 udp_tunnel_nic_get_port(struct net_device *dev, unsigned int table,
308 unsigned int idx, struct udp_tunnel_info *ti)
309 {
310 /* This helper is used from .sync_table, we indicate empty entries
311 * by zero'ed @ti. Drivers which need to know the details of a port
312 * when it gets deleted should use the .set_port / .unset_port
313 * callbacks.
314 * Zero out here, otherwise !CONFIG_INET causes uninitilized warnings.
315 */
316 memset(ti, 0, sizeof(*ti));
317
318 if (udp_tunnel_nic_ops)
319 udp_tunnel_nic_ops->get_port(dev, table, idx, ti);
320 }
321
322 static inline void
udp_tunnel_nic_set_port_priv(struct net_device * dev,unsigned int table,unsigned int idx,u8 priv)323 udp_tunnel_nic_set_port_priv(struct net_device *dev, unsigned int table,
324 unsigned int idx, u8 priv)
325 {
326 if (udp_tunnel_nic_ops)
327 udp_tunnel_nic_ops->set_port_priv(dev, table, idx, priv);
328 }
329
330 static inline void
udp_tunnel_nic_add_port(struct net_device * dev,struct udp_tunnel_info * ti)331 udp_tunnel_nic_add_port(struct net_device *dev, struct udp_tunnel_info *ti)
332 {
333 if (!(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT))
334 return;
335 if (udp_tunnel_nic_ops)
336 udp_tunnel_nic_ops->add_port(dev, ti);
337 }
338
339 static inline void
udp_tunnel_nic_del_port(struct net_device * dev,struct udp_tunnel_info * ti)340 udp_tunnel_nic_del_port(struct net_device *dev, struct udp_tunnel_info *ti)
341 {
342 if (!(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT))
343 return;
344 if (udp_tunnel_nic_ops)
345 udp_tunnel_nic_ops->del_port(dev, ti);
346 }
347
348 /**
349 * udp_tunnel_nic_reset_ntf() - device-originating reset notification
350 * @dev: network interface device structure
351 *
352 * Called by the driver to inform the core that the entire UDP tunnel port
353 * state has been lost, usually due to device reset. Core will assume device
354 * forgot all the ports and issue .set_port and .sync_table callbacks as
355 * necessary.
356 *
357 * This function must be called with rtnl lock held, and will issue all
358 * the callbacks before returning.
359 */
udp_tunnel_nic_reset_ntf(struct net_device * dev)360 static inline void udp_tunnel_nic_reset_ntf(struct net_device *dev)
361 {
362 if (udp_tunnel_nic_ops)
363 udp_tunnel_nic_ops->reset_ntf(dev);
364 }
365
366 static inline size_t
udp_tunnel_nic_dump_size(struct net_device * dev,unsigned int table)367 udp_tunnel_nic_dump_size(struct net_device *dev, unsigned int table)
368 {
369 if (!udp_tunnel_nic_ops)
370 return 0;
371 return udp_tunnel_nic_ops->dump_size(dev, table);
372 }
373
374 static inline int
udp_tunnel_nic_dump_write(struct net_device * dev,unsigned int table,struct sk_buff * skb)375 udp_tunnel_nic_dump_write(struct net_device *dev, unsigned int table,
376 struct sk_buff *skb)
377 {
378 if (!udp_tunnel_nic_ops)
379 return 0;
380 return udp_tunnel_nic_ops->dump_write(dev, table, skb);
381 }
382 #endif
383