1 /* $OpenBSD: in_pcb.h,v 1.164 2025/01/09 16:47:24 bluhm Exp $ */
2 /* $NetBSD: in_pcb.h,v 1.14 1996/02/13 23:42:00 christos Exp $ */
3
4 /*
5 * Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project.
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 * 1. Redistributions of source code must retain the above copyright
12 * notice, this list of conditions and the following disclaimer.
13 * 2. Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in the
15 * documentation and/or other materials provided with the distribution.
16 * 3. Neither the name of the project nor the names of its contributors
17 * may be used to endorse or promote products derived from this software
18 * without specific prior written permission.
19 *
20 * THIS SOFTWARE IS PROVIDED BY THE PROJECT AND CONTRIBUTORS ``AS IS'' AND
21 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23 * ARE DISCLAIMED. IN NO EVENT SHALL THE PROJECT OR CONTRIBUTORS BE LIABLE
24 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
25 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
26 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
27 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
28 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
29 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
30 * SUCH DAMAGE.
31 */
32
33 /*
34 * Copyright (c) 1982, 1986, 1990, 1993
35 * The Regents of the University of California. All rights reserved.
36 *
37 * Redistribution and use in source and binary forms, with or without
38 * modification, are permitted provided that the following conditions
39 * are met:
40 * 1. Redistributions of source code must retain the above copyright
41 * notice, this list of conditions and the following disclaimer.
42 * 2. Redistributions in binary form must reproduce the above copyright
43 * notice, this list of conditions and the following disclaimer in the
44 * documentation and/or other materials provided with the distribution.
45 * 3. Neither the name of the University nor the names of its contributors
46 * may be used to endorse or promote products derived from this software
47 * without specific prior written permission.
48 *
49 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
50 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
51 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
52 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
53 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
54 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
55 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
56 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
57 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
58 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
59 * SUCH DAMAGE.
60 *
61 * @(#)in_pcb.h 8.1 (Berkeley) 6/10/93
62 */
63
64 #ifndef _NETINET_IN_PCB_H_
65 #define _NETINET_IN_PCB_H_
66
67 #include <sys/queue.h>
68 #include <sys/mutex.h>
69 #include <sys/rwlock.h>
70 #include <sys/refcnt.h>
71 #include <netinet/ip6.h>
72 #include <netinet/icmp6.h>
73 #include <netinet/ip_ipsp.h>
74
75 #include <crypto/siphash.h>
76
77 /*
78 * Locks used to protect struct members in this file:
79 * I immutable after creation
80 * N net lock
81 * t inpt_mtx pcb table mutex
82 * L pf_inp_mtx link pf to inp mutex
83 * s so_lock socket rwlock
84 * f inp_sofree_mtx socket detach and lock
85 */
86
87 /*
88 * The pcb table mutex guarantees that all inpcb are consistent and
89 * that bind(2) and connect(2) create unique combinations of
90 * laddr/faddr/lport/fport/rtalbleid. This mutex is used to protect
91 * both address consistency and inpcb lookup during protocol input.
92 * All writes to inp_[lf]addr take table mutex. A per socket lock is
93 * needed, so that socket layer input have a consistent view at these
94 * values.
95 *
96 * In soconnect() and sosend() a per pcb mutex cannot be used. They
97 * eventually call IP output which takes pf lock which is a sleeping lock.
98 * Also connect(2) does a route lookup for source selection. There
99 * route resolve happens, which creates a route, which sends a route
100 * message, which needs route lock, which is a rw-lock.
101 *
102 * On the other hand a mutex should be used in protocol input. It
103 * does not make sense to do a process switch per packet. Better spin
104 * until the packet can be processed.
105 *
106 * So there are three locks. Table mutex is for writing inp_[lf]addr/port
107 * and lookup, socket rw-lock to separate sockets in system calls, and
108 * socket buffer mutex to protect socket receive buffer. Changing
109 * inp_[lf]addr/port takes both per socket rw-lock and global table mutex.
110 * Protocol input only reads inp_[lf]addr/port during lookup and is safe.
111 */
112
113 struct pf_state_key;
114
115 union inpaddru {
116 struct in_addr iau_addr;
117 struct in6_addr iau_addr6;
118 };
119
120 /*
121 * Common structure pcb for internet protocol implementation.
122 * Here are stored pointers to local and foreign host table
123 * entries, local and foreign socket numbers, and pointers
124 * up (to a socket structure) and down (to a protocol-specific)
125 * control block.
126 */
127 struct inpcb {
128 struct inpcbtable *inp_table; /* [I] inet queue/hash table */
129 TAILQ_ENTRY(inpcb) inp_queue; /* [t] inet PCB queue */
130 /* keep fields above in sync with struct inpcb_iterator */
131 LIST_ENTRY(inpcb) inp_hash; /* [t] local and foreign hash */
132 LIST_ENTRY(inpcb) inp_lhash; /* [t] local port hash */
133 union inpaddru inp_faddru; /* [t] Foreign address. */
134 union inpaddru inp_laddru; /* [t] Local address. */
135 #define inp_faddr inp_faddru.iau_addr
136 #define inp_faddr6 inp_faddru.iau_addr6
137 #define inp_laddr inp_laddru.iau_addr
138 #define inp_laddr6 inp_laddru.iau_addr6
139 u_int16_t inp_fport; /* [t] foreign port */
140 u_int16_t inp_lport; /* [t] local port */
141 struct socket *inp_socket; /* [f] back pointer to socket */
142 struct mutex inp_sofree_mtx; /* protect socket free */
143 caddr_t inp_ppcb; /* pointer to per-protocol pcb */
144 struct route inp_route; /* [s] cached route */
145 struct refcnt inp_refcnt; /* refcount PCB, delay memory free */
146 int inp_flags; /* generic IP/datagram flags */
147 union { /* Header prototype. */
148 struct ip hu_ip;
149 struct ip6_hdr hu_ipv6;
150 } inp_hu;
151 #define inp_ip inp_hu.hu_ip
152 #define inp_ipv6 inp_hu.hu_ipv6
153 union {
154 struct mbuf *inp_options; /* IPv4 options */
155 struct ip6_pktopts *inp_outputopts6; /* IPv6 options */
156 };
157 int inp_hops;
158 union {
159 struct ip_moptions *mou_mo;
160 struct ip6_moptions *mou_mo6;
161 } inp_mou;
162 #define inp_moptions inp_mou.mou_mo /* [N] IPv4 multicast options */
163 #define inp_moptions6 inp_mou.mou_mo6 /* [N] IPv6 multicast options */
164 struct ipsec_level inp_seclevel; /* [N] IPsec level of socket */
165 u_char inp_ip_minttl; /* minimum TTL or drop */
166 #define inp_ip6_minhlim inp_ip_minttl /* minimum Hop Limit or drop */
167 #define inp_flowinfo inp_hu.hu_ipv6.ip6_flow
168
169 int inp_cksum6;
170 struct icmp6_filter *inp_icmp6filt;
171 struct pf_state_key *inp_pf_sk; /* [L] */
172 struct mbuf *(*inp_upcall)(void *, struct mbuf *,
173 struct ip *, struct ip6_hdr *, void *, int);
174 void *inp_upcall_arg;
175 u_int inp_rtableid; /* [t] */
176 int inp_pipex; /* pipex indication */
177 uint16_t inp_flowid; /* [s] */
178 };
179
180 LIST_HEAD(inpcbhead, inpcb);
181
182 struct inpcb_iterator {
183 struct inpcbtable *inp_table; /* [I] always NULL */
184 TAILQ_ENTRY(inpcb) inp_queue; /* [t] inet PCB queue */
185 /* keep fields above in sync with struct inpcb */
186 };
187
188 static inline int
in_pcb_is_iterator(struct inpcb * inp)189 in_pcb_is_iterator(struct inpcb *inp)
190 {
191 return (inp->inp_table == NULL ? 1 : 0);
192 }
193
194 struct inpcbtable {
195 struct mutex inpt_mtx; /* protect queue and hash */
196 TAILQ_HEAD(inpthead, inpcb) inpt_queue; /* [t] inet PCB queue */
197 struct inpcbhead *inpt_hashtbl; /* [t] local and foreign hash */
198 struct inpcbhead *inpt_lhashtbl; /* [t] local port hash */
199 SIPHASH_KEY inpt_key, inpt_lkey; /* [I] secrets for hashes */
200 u_long inpt_mask, inpt_lmask; /* [t] hash masks */
201 int inpt_count, inpt_size; /* [t] queue count, hash size */
202 };
203
204 /* flags in inp_flags: */
205 #define INP_RECVOPTS 0x001 /* receive incoming IP options */
206 #define INP_RECVRETOPTS 0x002 /* receive IP options for reply */
207 #define INP_RECVDSTADDR 0x004 /* receive IP dst address */
208
209 #define INP_RXDSTOPTS INP_RECVOPTS
210 #define INP_RXHOPOPTS INP_RECVRETOPTS
211 #define INP_RXINFO INP_RECVDSTADDR
212 #define INP_RXSRCRT 0x010
213 #define INP_HOPLIMIT 0x020
214
215 #define INP_HDRINCL 0x008 /* user supplies entire IP header */
216 #define INP_HIGHPORT 0x010 /* user wants "high" port binding */
217 #define INP_LOWPORT 0x020 /* user wants "low" port binding */
218 #define INP_RECVIF 0x080 /* receive incoming interface */
219 #define INP_RECVTTL 0x040 /* receive incoming IP TTL */
220 #define INP_RECVDSTPORT 0x200 /* receive IP dst addr before rdr */
221 #define INP_RECVRTABLE 0x400 /* receive routing table */
222 #define INP_IPSECFLOWINFO 0x800 /* receive IPsec flow info */
223
224 #define INP_CONTROLOPTS (INP_RECVOPTS|INP_RECVRETOPTS|INP_RECVDSTADDR| \
225 INP_RXSRCRT|INP_HOPLIMIT|INP_RECVIF|INP_RECVTTL|INP_RECVDSTPORT| \
226 INP_RECVRTABLE)
227
228 /*
229 * These flags' values should be determined by either the transport
230 * protocol at PRU_BIND, PRU_LISTEN, PRU_CONNECT, etc, or by in_pcb*().
231 */
232 #define INP_IPV6 0x100 /* socket, proto, domain, family is PF_INET6 */
233
234 /*
235 * Flags in inp_flags for IPV6
236 */
237 #define IN6P_HIGHPORT INP_HIGHPORT /* user wants "high" port */
238 #define IN6P_LOWPORT INP_LOWPORT /* user wants "low" port */
239 #define IN6P_RECVDSTPORT INP_RECVDSTPORT /* receive IP dst addr before rdr */
240 #define IN6P_PKTINFO 0x010000 /* receive IP6 dst and I/F */
241 #define IN6P_HOPLIMIT 0x020000 /* receive hoplimit */
242 #define IN6P_HOPOPTS 0x040000 /* receive hop-by-hop options */
243 #define IN6P_DSTOPTS 0x080000 /* receive dst options after rthdr */
244 #define IN6P_RTHDR 0x100000 /* receive routing header */
245 #define IN6P_TCLASS 0x400000 /* receive traffic class value */
246 #define IN6P_AUTOFLOWLABEL 0x800000 /* attach flowlabel automatically */
247
248 #define IN6P_ANONPORT 0x4000000 /* port chosen for user */
249 #define IN6P_RFC2292 0x40000000 /* used RFC2292 API on the socket */
250 #define IN6P_MTU 0x80000000 /* receive path MTU */
251
252 #define IN6P_MINMTU 0x20000000 /* use minimum MTU */
253
254 #define IN6P_CONTROLOPTS (IN6P_PKTINFO|IN6P_HOPLIMIT|IN6P_HOPOPTS|\
255 IN6P_DSTOPTS|IN6P_RTHDR|\
256 IN6P_TCLASS|IN6P_AUTOFLOWLABEL|IN6P_RFC2292|\
257 IN6P_MTU|IN6P_RECVDSTPORT)
258
259 #define INPLOOKUP_WILDCARD 1
260 #define INPLOOKUP_SETLOCAL 2
261 #define INPLOOKUP_IPV6 4
262
263 #define sotoinpcb(so) ((struct inpcb *)(so)->so_pcb)
264
265 /* macros for handling bitmap of ports not to allocate dynamically */
266 #define DP_MAPBITS (sizeof(u_int32_t) * NBBY)
267 #define DP_MAPSIZE (howmany(65536, DP_MAPBITS))
268 #define DP_SET(m, p) ((m)[(p) / DP_MAPBITS] |= (1U << ((p) % DP_MAPBITS)))
269 #define DP_CLR(m, p) ((m)[(p) / DP_MAPBITS] &= ~(1U << ((p) % DP_MAPBITS)))
270 #define DP_ISSET(m, p) ((m)[(p) / DP_MAPBITS] & (1U << ((p) % DP_MAPBITS)))
271
272 /* default values for baddynamicports [see ip_init()] */
273 #define DEFBADDYNAMICPORTS_TCP { \
274 587, 749, 750, 751, 853, 871, 2049, \
275 6000, 6001, 6002, 6003, 6004, 6005, 6006, 6007, 6008, 6009, 6010, \
276 0 }
277 #define DEFBADDYNAMICPORTS_UDP { 623, 664, 749, 750, 751, 2049, \
278 3784, 3785, 7784, /* BFD/S-BFD ports */ \
279 0 }
280
281 #define DEFROOTONLYPORTS_TCP { \
282 2049, \
283 0 }
284 #define DEFROOTONLYPORTS_UDP { \
285 2049, \
286 0 }
287
288 struct baddynamicports {
289 u_int32_t tcp[DP_MAPSIZE];
290 u_int32_t udp[DP_MAPSIZE];
291 };
292
293 #ifdef _KERNEL
294
295 #define IN_PCBLOCK_HOLD 1
296 #define IN_PCBLOCK_GRAB 2
297
298 extern struct inpcbtable rawcbtable, rawin6pcbtable;
299 extern struct baddynamicports baddynamicports;
300 extern struct baddynamicports rootonlyports;
301 extern int in_pcbnotifymiss;
302
303 void in_init(void);
304 void in_losing(struct inpcb *);
305 int in_pcballoc(struct socket *, struct inpcbtable *, int);
306 int in_pcbbind_locked(struct inpcb *, struct mbuf *, const void *,
307 struct proc *);
308 int in_pcbbind(struct inpcb *, struct mbuf *, struct proc *);
309 int in_pcbaddrisavail(const struct inpcb *, struct sockaddr_in *, int,
310 struct proc *);
311 int in_pcbconnect(struct inpcb *, struct mbuf *);
312 void in_pcbdetach(struct inpcb *);
313 struct socket *
314 in_pcbsolock_ref(struct inpcb *);
315 void in_pcbsounlock_rele(struct inpcb *, struct socket *);
316 struct inpcb *
317 in_pcbref(struct inpcb *);
318 void in_pcbunref(struct inpcb *);
319 void in_pcbdisconnect(struct inpcb *);
320 struct inpcb *
321 in_pcb_iterator(struct inpcbtable *, struct inpcb *,
322 struct inpcb_iterator *);
323 void in_pcb_iterator_abort(struct inpcbtable *, struct inpcb *,
324 struct inpcb_iterator *);
325 struct inpcb *
326 in_pcblookup(struct inpcbtable *, struct in_addr,
327 u_int, struct in_addr, u_int, u_int);
328 struct inpcb *
329 in_pcblookup_listen(struct inpcbtable *, struct in_addr, u_int,
330 struct mbuf *, u_int);
331 #ifdef INET6
332 uint64_t in6_pcbhash(struct inpcbtable *, u_int, const struct in6_addr *,
333 u_short, const struct in6_addr *, u_short);
334 struct inpcb *
335 in6_pcblookup(struct inpcbtable *, const struct in6_addr *,
336 u_int, const struct in6_addr *, u_int, u_int);
337 struct inpcb *
338 in6_pcblookup_listen(struct inpcbtable *, struct in6_addr *, u_int,
339 struct mbuf *, u_int);
340 int in6_pcbaddrisavail_lock(const struct inpcb *, struct sockaddr_in6 *,
341 int, struct proc *, int);
342 int in6_pcbaddrisavail(const struct inpcb *, struct sockaddr_in6 *, int,
343 struct proc *);
344 int in6_pcbconnect(struct inpcb *, struct mbuf *);
345 void in6_setsockaddr(struct inpcb *, struct mbuf *);
346 void in6_setpeeraddr(struct inpcb *, struct mbuf *);
347 int in6_sockaddr(struct socket *, struct mbuf *);
348 int in6_peeraddr(struct socket *, struct mbuf *);
349 #endif /* INET6 */
350 void in_pcbinit(struct inpcbtable *, int);
351 struct inpcb *
352 in_pcblookup_local_lock(struct inpcbtable *, const void *, u_int, int,
353 u_int, int);
354 void in_pcbnotifyall(struct inpcbtable *, const struct sockaddr_in *,
355 u_int, int, void (*)(struct inpcb *, int));
356 void in_pcbrehash(struct inpcb *);
357 void in_rtchange(struct inpcb *, int);
358 void in_setpeeraddr(struct inpcb *, struct mbuf *);
359 void in_setsockaddr(struct inpcb *, struct mbuf *);
360 int in_sockaddr(struct socket *, struct mbuf *);
361 int in_peeraddr(struct socket *, struct mbuf *);
362 int in_baddynamic(u_int16_t, u_int16_t);
363 int in_rootonly(u_int16_t, u_int16_t);
364 int in_pcbselsrc(struct in_addr *, struct sockaddr_in *, struct inpcb *);
365 struct rtentry *
366 in_pcbrtentry(struct inpcb *);
367
368 /* INET6 stuff */
369 struct rtentry *
370 in6_pcbrtentry(struct inpcb *);
371 void in6_pcbnotify(struct inpcbtable *, const struct sockaddr_in6 *,
372 u_int, const struct sockaddr_in6 *, u_int, u_int, int, void *,
373 void (*)(struct inpcb *, int));
374 int in6_selecthlim(const struct inpcb *);
375 int in_pcbset_rtableid(struct inpcb *, u_int);
376 void in_pcbset_laddr(struct inpcb *, const struct sockaddr *, u_int);
377 void in_pcbunset_faddr(struct inpcb *);
378 void in_pcbunset_laddr(struct inpcb *);
379
380 #endif /* _KERNEL */
381 #endif /* _NETINET_IN_PCB_H_ */
382