1/*-
2 * SPDX-License-Identifier: BSD-3-Clause
3 *
4 * Copyright (c) 1982, 1986, 1993
5 * The Regents of the University of California. All rights reserved.
6 *
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
9 * are met:
10 * 1. Redistributions of source code must retain the above copyright
11 * notice, this list of conditions and the following disclaimer.
12 * 2. Redistributions in binary form must reproduce the above copyright
13 * notice, this list of conditions and the following disclaimer in the
14 * documentation and/or other materials provided with the distribution.
15 * 3. Neither the name of the University nor the names of its contributors
16 * may be used to endorse or promote products derived from this software
17 * without specific prior written permission.
18 *
19 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
20 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
21 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
22 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
23 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
24 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
25 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
26 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
27 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
28 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
29 * SUCH DAMAGE.
30 */
31
32#ifndef _NETINET_IP_VAR_H_
33#define _NETINET_IP_VAR_H_
34
35#include <sys/epoch.h>
36#include <sys/queue.h>
37#include <sys/types.h>
38
39#include <netinet/in.h>
40
41/*
42 * Overlay for ip header used by other protocols (tcp, udp).
43 */
44struct ipovly {
45 u_char ih_x1[9]; /* (unused) */
46 u_char ih_pr; /* protocol */
47 u_short ih_len; /* protocol length */
48 struct in_addr ih_src; /* source internet address */
49 struct in_addr ih_dst; /* destination internet address */
50} __packed;
51
52#ifdef _KERNEL
53/*
54 * Ip reassembly queue structure. Each fragment
55 * being reassembled is attached to one of these structures.
56 * They are timed out after net.inet.ip.fragttl seconds, and may also be
57 * reclaimed if memory becomes tight.
58 */
59struct ipq {
60 TAILQ_ENTRY(ipq) ipq_list; /* to other reass headers */
61 time_t ipq_expire; /* time_uptime when ipq expires */
62 u_char ipq_nfrags; /* # frags in this packet */
63 u_char ipq_p; /* protocol of this fragment */
64 u_short ipq_id; /* sequence id for reassembly */
65 int ipq_maxoff; /* total length of packet */
66 struct mbuf *ipq_frags; /* to ip headers of fragments */
67 struct in_addr ipq_src,ipq_dst;
68 struct label *ipq_label; /* MAC label */
69};
70#endif /* _KERNEL */
71
72/*
73 * Structure stored in mbuf in inpcb.ip_options
74 * and passed to ip_output when ip options are in use.
75 * The actual length of the options (including ipopt_dst)
76 * is in m_len.
77 */
78#define MAX_IPOPTLEN 40
79
80struct ipoption {
81 struct in_addr ipopt_dst; /* first-hop dst if source routed */
82 char ipopt_list[MAX_IPOPTLEN]; /* options proper */
83};
84
85#if defined(_NETINET_IN_VAR_H_) && defined(_KERNEL)
86/*
87 * Structure attached to inpcb.ip_moptions and
88 * passed to ip_output when IP multicast options are in use.
89 * This structure is lazy-allocated.
90 */
91struct ip_moptions {
92 struct ifnet *imo_multicast_ifp; /* ifp for outgoing multicasts */
93 struct in_addr imo_multicast_addr; /* ifindex/addr on MULTICAST_IF */
94 u_long imo_multicast_vif; /* vif num outgoing multicasts */
95 u_char imo_multicast_ttl; /* TTL for outgoing multicasts */
96 u_char imo_multicast_loop; /* 1 => hear sends if a member */
97 struct ip_mfilter_head imo_head; /* group membership list */
98};
99#else
100struct ip_moptions;
101#endif
102
103struct ipstat {
104 uint64_t ips_total; /* total packets received */
105 uint64_t ips_badsum; /* checksum bad */
106 uint64_t ips_tooshort; /* packet too short */
107 uint64_t ips_toosmall; /* not enough data */
108 uint64_t ips_badhlen; /* ip header length < data size */
109 uint64_t ips_badlen; /* ip length < ip header length */
110 uint64_t ips_fragments; /* fragments received */
111 uint64_t ips_fragdropped; /* frags dropped (dups, out of space) */
112 uint64_t ips_fragtimeout; /* fragments timed out */
113 uint64_t ips_forward; /* packets forwarded */
114 uint64_t ips_fastforward; /* packets fast forwarded */
115 uint64_t ips_cantforward; /* packets rcvd for unreachable dest */
116 uint64_t ips_redirectsent; /* packets forwarded on same net */
117 uint64_t ips_noproto; /* unknown or unsupported protocol */
118 uint64_t ips_delivered; /* datagrams delivered to upper level*/
119 uint64_t ips_localout; /* total ip packets generated here */
120 uint64_t ips_odropped; /* lost packets due to nobufs, etc. */
121 uint64_t ips_reassembled; /* total packets reassembled ok */
122 uint64_t ips_fragmented; /* datagrams successfully fragmented */
123 uint64_t ips_ofragments; /* output fragments created */
124 uint64_t ips_cantfrag; /* don't fragment flag was set, etc. */
125 uint64_t ips_badoptions; /* error in option processing */
126 uint64_t ips_noroute; /* packets discarded due to no route */
127 uint64_t ips_badvers; /* ip version != 4 */
128 uint64_t ips_rawout; /* total raw ip packets generated */
129 uint64_t ips_toolong; /* ip length > max ip packet size */
130 uint64_t ips_notmember; /* multicasts for unregistered grps */
131 uint64_t ips_nogif; /* no match gif found */
132 uint64_t ips_badaddr; /* invalid address on header */
133};
134
135#ifdef _KERNEL
136
137#include <sys/counter.h>
138#include <net/vnet.h>
139#include <netinet/in_kdtrace.h>
140
141VNET_PCPUSTAT_DECLARE(struct ipstat, ipstat);
142/*
143 * In-kernel consumers can use these accessor macros directly to update
144 * stats.
145 */
146#define IPSTAT_ADD(name, val) \
147 do { \
148 MIB_SDT_PROBE1(ip, count, name, (val)); \
149 VNET_PCPUSTAT_ADD(struct ipstat, ipstat, name, (val)); \
150 } while (0)
151#define IPSTAT_SUB(name, val) IPSTAT_ADD(name, -(val))
152#define IPSTAT_INC(name) IPSTAT_ADD(name, 1)
153#define IPSTAT_DEC(name) IPSTAT_SUB(name, 1)
154
155/*
156 * Kernel module consumers must use this accessor macro.
157 */
158void kmod_ipstat_inc(int statnum);
159#define KMOD_IPSTAT_INC(name) \
160 do { \
161 MIB_SDT_PROBE1(ip, count, name, 1); \
162 kmod_ipstat_inc( \
163 offsetof(struct ipstat, name) / sizeof(uint64_t)); \
164 } while (0)
165void kmod_ipstat_dec(int statnum);
166#define KMOD_IPSTAT_DEC(name) \
167 do { \
168 MIB_SDT_PROBE1(ip, count, name, -1); \
169 kmod_ipstat_dec( \
170 offsetof(struct ipstat, name) / sizeof(uint64_t)); \
171 } while (0)
172
173/* flags passed to ip_output as last parameter */
174#define IP_FORWARDING 0x1 /* most of ip header exists */
175#define IP_RAWOUTPUT 0x2 /* raw ip header exists */
176#define IP_SENDONES 0x4 /* send all-ones broadcast */
177#define IP_SENDTOIF 0x8 /* send on specific ifnet */
178#define IP_ROUTETOIF SO_DONTROUTE /* 0x10 bypass routing tables */
179#define IP_ALLOWBROADCAST SO_BROADCAST /* 0x20 can send broadcast packets */
180#define IP_NODEFAULTFLOWID 0x40 /* Don't set the flowid from inp */
181#define IP_NO_SND_TAG_RL 0x80 /* Don't send down the ratelimit tag */
182
183#ifdef __NO_STRICT_ALIGNMENT
184#define IP_HDR_ALIGNED_P(ip) 1
185#else
186#define IP_HDR_ALIGNED_P(ip) ((((intptr_t) (ip)) & 3) == 0)
187#endif
188
189struct ip;
190struct inpcb;
191struct route;
192struct sockopt;
193struct inpcbinfo;
194
195VNET_DECLARE(int, ip_defttl); /* default IP ttl */
196VNET_DECLARE(int, ipforwarding); /* ip forwarding */
197VNET_DECLARE(int, ipsendredirects);
198#ifdef IPSTEALTH
199VNET_DECLARE(int, ipstealth); /* stealth forwarding */
200#endif
201VNET_DECLARE(struct socket *, ip_rsvpd); /* reservation protocol daemon*/
202VNET_DECLARE(struct socket *, ip_mrouter); /* multicast routing daemon */
203extern int (*legal_vif_num)(int);
204extern u_long (*ip_mcast_src)(int);
205VNET_DECLARE(int, rsvp_on);
206VNET_DECLARE(int, drop_redirect);
207VNET_DECLARE(int, ip_random_id);
208
209#define V_ip_id VNET(ip_id)
210#define V_ip_defttl VNET(ip_defttl)
211#define V_ipforwarding VNET(ipforwarding)
212#define V_ipsendredirects VNET(ipsendredirects)
213#ifdef IPSTEALTH
214#define V_ipstealth VNET(ipstealth)
215#endif
216#define V_ip_rsvpd VNET(ip_rsvpd)
217#define V_ip_mrouter VNET(ip_mrouter)
218#define V_rsvp_on VNET(rsvp_on)
219#define V_drop_redirect VNET(drop_redirect)
220#define V_ip_random_id VNET(ip_random_id)
221
222void inp_freemoptions(struct ip_moptions *);
223int inp_getmoptions(struct inpcb *, struct sockopt *);
224int inp_setmoptions(struct inpcb *, struct sockopt *);
225
226int ip_ctloutput(struct socket *, struct sockopt *sopt);
227int ip_fragment(struct ip *ip, struct mbuf **m_frag, int mtu,
228 u_long if_hwassist_flags);
229void ip_forward(struct mbuf *m, int srcrt);
230extern int
231 (*ip_mforward)(struct ip *, struct ifnet *, struct mbuf *,
232 struct ip_moptions *);
233int ip_output(struct mbuf *,
234 struct mbuf *, struct route *, int, struct ip_moptions *,
235 struct inpcb *);
236struct mbuf *
237 ip_reass(struct mbuf *);
238void ip_savecontrol(struct inpcb *, struct mbuf **, struct ip *,
239 struct mbuf *);
240void ip_fillid(struct ip *, bool);
241int rip_ctloutput(struct socket *, struct sockopt *);
242int ipip_input(struct mbuf **, int *, int);
243int rsvp_input(struct mbuf **, int *, int);
244
245int ip_rsvp_init(struct socket *);
246int ip_rsvp_done(void);
247extern int (*ip_rsvp_vif)(struct socket *, struct sockopt *);
248extern void (*ip_rsvp_force_done)(struct socket *);
249extern int (*rsvp_input_p)(struct mbuf **, int *, int);
250
251typedef int ipproto_input_t(struct mbuf **, int *, int);
252struct icmp;
253typedef void ipproto_ctlinput_t(struct icmp *);
254int ipproto_register(uint8_t, ipproto_input_t, ipproto_ctlinput_t);
255int ipproto_unregister(uint8_t);
256#define IPPROTO_REGISTER(prot, input, ctl) do { \
257 int error __diagused; \
258 error = ipproto_register(prot, input, ctl); \
259 MPASS(error == 0); \
260} while (0)
261
262ipproto_input_t rip_input;
263ipproto_ctlinput_t rip_ctlinput;
264
265VNET_DECLARE(struct pfil_head *, inet_pfil_head);
266#define V_inet_pfil_head VNET(inet_pfil_head)
267#define PFIL_INET_NAME "inet"
268
269VNET_DECLARE(struct pfil_head *, inet_local_pfil_head);
270#define V_inet_local_pfil_head VNET(inet_local_pfil_head)
271#define PFIL_INET_LOCAL_NAME "inet-local"
272
273void in_delayed_cksum(struct mbuf *m);
274void in_delayed_cksum_o(struct mbuf *m, uint16_t o);
275
276/* Hooks for ipfw, dummynet, divert etc. Most are declared in raw_ip.c */
277/*
278 * Reference to an ipfw or packet filter rule that can be carried
279 * outside critical sections.
280 * A rule is identified by rulenum:rule_id which is ordered.
281 * In version chain_id the rule can be found in slot 'slot', so
282 * we don't need a lookup if chain_id == chain->id.
283 *
284 * On exit from the firewall this structure refers to the rule after
285 * the matching one (slot points to the new rule; rulenum:rule_id-1
286 * is the matching rule), and additional info (e.g. info often contains
287 * the insn argument or tablearg in the low 16 bits, in host format).
288 * On entry, the structure is valid if slot>0, and refers to the starting
289 * rules. 'info' contains the reason for reinject, e.g. divert port,
290 * divert direction, and so on.
291 *
292 * Packet Mark is an analogue to ipfw tags with O(1) lookup from mbuf while
293 * regular tags require a single-linked list traversal. Mark is a 32-bit
294 * number that can be looked up in a table [with 'number' table-type], matched
295 * or compared with a number with optional mask applied before comparison.
296 * Having generic nature, Mark can be used in a variety of needs.
297 * For example, it could be used as a security group: mark will hold a
298 * security group id and represent a group of packet flows that shares same
299 * access control policy.
300 * O_MASK opcode can match mark value bitwise so one can build a hierarchical
301 * model designating different meanings for a bit range(s).
302 */
303struct ipfw_rule_ref {
304/* struct m_tag spans 24 bytes above this point, see mbuf_tags(9) */
305 /* spare space just to be save in case struct m_tag grows */
306/* -- 32 bytes -- */
307 uint32_t slot; /* slot for matching rule */
308 uint32_t rulenum; /* matching rule number */
309 uint32_t rule_id; /* matching rule id */
310 uint32_t chain_id; /* ruleset id */
311 uint32_t info; /* see below */
312 uint32_t pkt_mark; /* packet mark */
313 uint32_t spare[2];
314/* -- 64 bytes -- */
315};
316
317enum {
318 IPFW_INFO_MASK = 0x0000ffff,
319 IPFW_INFO_OUT = 0x00000000, /* outgoing, just for convenience */
320 IPFW_INFO_IN = 0x80000000, /* incoming, overloads dir */
321 IPFW_ONEPASS = 0x40000000, /* One-pass, do not reinject */
322 IPFW_IS_MASK = 0x30000000, /* which source ? */
323 IPFW_IS_DIVERT = 0x20000000,
324 IPFW_IS_DUMMYNET =0x10000000,
325 IPFW_IS_PIPE = 0x08000000, /* pipe=1, queue = 0 */
326};
327#define MTAG_IPFW 1148380143 /* IPFW-tagged cookie */
328#define MTAG_IPFW_RULE 1262273568 /* rule reference */
329#define MTAG_IPFW_CALL 1308397630 /* call stack */
330
331struct ip_fw_args;
332typedef int (*ip_fw_ctl_ptr_t)(struct sockopt *);
333VNET_DECLARE(ip_fw_ctl_ptr_t, ip_fw_ctl_ptr);
334#define V_ip_fw_ctl_ptr VNET(ip_fw_ctl_ptr)
335
336/* Divert hooks. */
337extern void (*ip_divert_ptr)(struct mbuf *m, bool incoming);
338/* ng_ipfw hooks -- XXX make it the same as divert and dummynet */
339extern int (*ng_ipfw_input_p)(struct mbuf **, struct ip_fw_args *, bool);
340extern int (*ip_dn_ctl_ptr)(struct sockopt *);
341extern int (*ip_dn_io_ptr)(struct mbuf **, struct ip_fw_args *);
342
343/* pf specific mtag for divert(4) support */
344__enum_uint8_decl(pf_mtag_dir) {
345 PF_DIVERT_MTAG_DIR_IN = 1,
346 PF_DIVERT_MTAG_DIR_OUT = 2
347};
348struct pf_divert_mtag {
349 __enum_uint8(pf_mtag_dir) idir; /* initial pkt direction */
350 union {
351 __enum_uint8(pf_mtag_dir) ndir; /* new dir after re-enter */
352 uint16_t port; /* initial divert(4) port */
353 };
354};
355#define MTAG_PF_DIVERT 1262273569
356
357#endif /* _KERNEL */
358
359#endif /* !_NETINET_IP_VAR_H_ */