raw_ip.c revision 9209 
1/*
2 * Copyright (c) 1982, 1986, 1988, 1993
3 *	The Regents of the University of California.  All rights reserved.
4 *
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
7 * are met:
8 * 1. Redistributions of source code must retain the above copyright
9 *    notice, this list of conditions and the following disclaimer.
10 * 2. Redistributions in binary form must reproduce the above copyright
11 *    notice, this list of conditions and the following disclaimer in the
12 *    documentation and/or other materials provided with the distribution.
13 * 3. All advertising materials mentioning features or use of this software
14 *    must display the following acknowledgement:
15 *	This product includes software developed by the University of
16 *	California, Berkeley and its contributors.
17 * 4. Neither the name of the University nor the names of its contributors
18 *    may be used to endorse or promote products derived from this software
19 *    without specific prior written permission.
20 *
21 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
22 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
23 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
24 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
25 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
26 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
27 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
28 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
29 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
30 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
31 * SUCH DAMAGE.
32 *
33 *	@(#)raw_ip.c	8.2 (Berkeley) 1/4/94
34 * $Id: raw_ip.c,v 1.18 1995/05/12 20:00:21 ache Exp $
35 */
36
37#include <sys/param.h>
38#include <sys/malloc.h>
39#include <sys/mbuf.h>
40#include <sys/socket.h>
41#include <sys/protosw.h>
42#include <sys/socketvar.h>
43#include <sys/errno.h>
44#include <sys/systm.h>
45#include <sys/queue.h>
46
47#include <net/if.h>
48#include <net/route.h>
49
50#include <netinet/in.h>
51#include <netinet/in_systm.h>
52#include <netinet/ip.h>
53#include <netinet/ip_var.h>
54#include <netinet/ip_mroute.h>
55#include <netinet/in_pcb.h>
56
57#include <netinet/ip_fw.h>
58
59struct inpcbhead ripcb;
60struct inpcbinfo ripcbinfo;
61
62/*
63 * Nominal space allocated to a raw ip socket.
64 */
65#define	RIPSNDQ		8192
66#define	RIPRCVQ		8192
67
68/*
69 * Raw interface to IP protocol.
70 */
71
72/*
73 * Initialize raw connection block q.
74 */
75void
76rip_init()
77{
78	LIST_INIT(&ripcb);
79	ripcbinfo.listhead = &ripcb;
80	/*
81	 * XXX We don't use the hash list for raw IP, but it's easier
82	 * to allocate a one entry hash list than it is to check all
83	 * over the place for hashbase == NULL.
84	 */
85	ripcbinfo.hashbase = phashinit(1, M_PCB, &ripcbinfo.hashsize);
86}
87
88struct	sockaddr_in ripsrc = { sizeof(ripsrc), AF_INET };
89/*
90 * Setup generic address and protocol structures
91 * for raw_input routine, then pass them along with
92 * mbuf chain.
93 */
94void
95rip_input(m)
96	struct mbuf *m;
97{
98	register struct ip *ip = mtod(m, struct ip *);
99	register struct inpcb *inp;
100	struct socket *last = 0;
101
102	ripsrc.sin_addr = ip->ip_src;
103	for (inp = ripcb.lh_first; inp != NULL; inp = inp->inp_list.le_next) {
104		if (inp->inp_ip.ip_p && inp->inp_ip.ip_p != ip->ip_p)
105			continue;
106		if (inp->inp_laddr.s_addr &&
107                  inp->inp_laddr.s_addr != ip->ip_dst.s_addr)
108			continue;
109		if (inp->inp_faddr.s_addr &&
110                  inp->inp_faddr.s_addr != ip->ip_src.s_addr)
111			continue;
112		if (last) {
113			struct mbuf *n = m_copy(m, 0, (int)M_COPYALL);
114			if (n) {
115				if (sbappendaddr(&last->so_rcv,
116				    (struct sockaddr *)&ripsrc, n,
117				    (struct mbuf *)0) == 0)
118					/* should notify about lost packet */
119					m_freem(n);
120				else
121					sorwakeup(last);
122			}
123		}
124		last = inp->inp_socket;
125	}
126	if (last) {
127		if (sbappendaddr(&last->so_rcv, (struct sockaddr *)&ripsrc,
128		    m, (struct mbuf *)0) == 0)
129			m_freem(m);
130		else
131			sorwakeup(last);
132	} else {
133		m_freem(m);
134              ipstat.ips_noproto++;
135              ipstat.ips_delivered--;
136      }
137}
138
139/*
140 * Generate IP header and pass packet to ip_output.
141 * Tack on options user may have setup with control call.
142 */
143int
144rip_output(m, so, dst)
145	register struct mbuf *m;
146	struct socket *so;
147	u_long dst;
148{
149	register struct ip *ip;
150	register struct inpcb *inp = sotoinpcb(so);
151	struct mbuf *opts;
152	int flags = (so->so_options & SO_DONTROUTE) | IP_ALLOWBROADCAST;
153
154	/*
155	 * If the user handed us a complete IP packet, use it.
156	 * Otherwise, allocate an mbuf for a header and fill it in.
157	 */
158	if ((inp->inp_flags & INP_HDRINCL) == 0) {
159		M_PREPEND(m, sizeof(struct ip), M_WAIT);
160		ip = mtod(m, struct ip *);
161		ip->ip_tos = 0;
162		ip->ip_off = 0;
163		ip->ip_p = inp->inp_ip.ip_p;
164		ip->ip_len = m->m_pkthdr.len;
165		ip->ip_src = inp->inp_laddr;
166		ip->ip_dst.s_addr = dst;
167		ip->ip_ttl = MAXTTL;
168		opts = inp->inp_options;
169	} else {
170		ip = mtod(m, struct ip *);
171		if (ip->ip_id == 0)
172			ip->ip_id = htons(ip_id++);
173		opts = NULL;
174		/* XXX prevent ip_output from overwriting header fields */
175		flags |= IP_RAWOUTPUT;
176		ipstat.ips_rawout++;
177	}
178	return (ip_output(m, opts, &inp->inp_route, flags, inp->inp_moptions));
179}
180
181/*
182 * Raw IP socket option processing.
183 */
184int
185rip_ctloutput(op, so, level, optname, m)
186	int op;
187	struct socket *so;
188	int level, optname;
189	struct mbuf **m;
190{
191	register struct inpcb *inp = sotoinpcb(so);
192	register int error;
193
194	if (level != IPPROTO_IP) {
195		if (op == PRCO_SETOPT && *m)
196			(void)m_free(*m);
197		return (EINVAL);
198	}
199
200	switch (optname) {
201
202	case IP_HDRINCL:
203		if (op == PRCO_SETOPT || op == PRCO_GETOPT) {
204			if (m == 0 || *m == 0 || (*m)->m_len < sizeof (int))
205				return (EINVAL);
206			if (op == PRCO_SETOPT) {
207				if (*mtod(*m, int *))
208					inp->inp_flags |= INP_HDRINCL;
209				else
210					inp->inp_flags &= ~INP_HDRINCL;
211				(void)m_free(*m);
212			} else {
213				(*m)->m_len = sizeof (int);
214				*mtod(*m, int *) = inp->inp_flags & INP_HDRINCL;
215			}
216			return (0);
217		}
218		break;
219
220	case IP_FW_ADD:
221	case IP_FW_DEL:
222	case IP_FW_FLUSH:
223	case IP_FW_POLICY:
224		if (ip_fw_ctl_ptr==NULL) {
225			if (*m)
226				(void)m_free(*m);
227			return(EINVAL);
228		}
229
230		if (op == PRCO_SETOPT) {
231			error=(*ip_fw_ctl_ptr)(optname, *m);
232			if (*m)
233				(void)m_free(*m);
234		}
235		else
236			error=EINVAL;
237		return(error);
238
239	case IP_ACCT_DEL:
240	case IP_ACCT_ADD:
241	case IP_ACCT_CLR:
242	case IP_ACCT_FLUSH:
243	case IP_ACCT_ZERO:
244		if (ip_acct_ctl_ptr==NULL) {
245			if (*m)
246				(void)m_free(*m);
247			return(EINVAL);
248		}
249
250		if (op == PRCO_SETOPT) {
251			error=(*ip_acct_ctl_ptr)(optname, *m);
252			if (*m)
253				(void)m_free(*m);
254		}
255		else
256			error=EINVAL;
257		return(error);
258
259	case IP_RSVP_ON:
260		return ip_rsvp_init(so);
261		break;
262
263	case IP_RSVP_OFF:
264		return ip_rsvp_done();
265		break;
266
267	case IP_RSVP_VIF_ON:
268		return ip_rsvp_vif_init(so, *m);
269
270	case IP_RSVP_VIF_OFF:
271		return ip_rsvp_vif_done(so, *m);
272
273	case MRT_INIT:
274	case MRT_DONE:
275	case MRT_ADD_VIF:
276	case MRT_DEL_VIF:
277	case MRT_ADD_MFC:
278	case MRT_DEL_MFC:
279	case MRT_VERSION:
280	case MRT_ASSERT:
281		if (op == PRCO_SETOPT) {
282			error = ip_mrouter_set(optname, so, *m);
283			if (*m)
284				(void)m_free(*m);
285		} else if (op == PRCO_GETOPT) {
286			error = ip_mrouter_get(optname, so, m);
287		} else
288			error = EINVAL;
289		return (error);
290	}
291	return (ip_ctloutput(op, so, level, optname, m));
292}
293
294u_long	rip_sendspace = RIPSNDQ;
295u_long	rip_recvspace = RIPRCVQ;
296
297/*ARGSUSED*/
298int
299rip_usrreq(so, req, m, nam, control)
300	register struct socket *so;
301	int req;
302	struct mbuf *m, *nam, *control;
303{
304	register int error = 0;
305	register struct inpcb *inp = sotoinpcb(so);
306	switch (req) {
307
308	case PRU_ATTACH:
309		if (inp)
310			panic("rip_attach");
311		if ((so->so_state & SS_PRIV) == 0) {
312			error = EACCES;
313			break;
314		}
315		if ((error = soreserve(so, rip_sendspace, rip_recvspace)) ||
316		    (error = in_pcballoc(so, &ripcbinfo)))
317			break;
318		inp = (struct inpcb *)so->so_pcb;
319		inp->inp_ip.ip_p = (int)nam;
320		break;
321
322	case PRU_DISCONNECT:
323		if ((so->so_state & SS_ISCONNECTED) == 0) {
324			error = ENOTCONN;
325			break;
326		}
327		/* FALLTHROUGH */
328	case PRU_ABORT:
329		soisdisconnected(so);
330		/* FALLTHROUGH */
331	case PRU_DETACH:
332		if (inp == 0)
333			panic("rip_detach");
334		if (so == ip_mrouter)
335			ip_mrouter_done();
336		if (so == ip_rsvpd)
337			ip_rsvp_done();
338		in_pcbdetach(inp);
339		break;
340
341	case PRU_BIND:
342	    {
343		struct sockaddr_in *addr = mtod(nam, struct sockaddr_in *);
344
345		if (nam->m_len != sizeof(*addr)) {
346			error = EINVAL;
347			break;
348		}
349		if ((ifnet == 0) ||
350		    ((addr->sin_family != AF_INET) &&
351		     (addr->sin_family != AF_IMPLINK)) ||
352		    (addr->sin_addr.s_addr &&
353		     ifa_ifwithaddr((struct sockaddr *)addr) == 0)) {
354			error = EADDRNOTAVAIL;
355			break;
356		}
357		inp->inp_laddr = addr->sin_addr;
358		break;
359	    }
360	case PRU_CONNECT:
361	    {
362		struct sockaddr_in *addr = mtod(nam, struct sockaddr_in *);
363
364		if (nam->m_len != sizeof(*addr)) {
365			error = EINVAL;
366			break;
367		}
368		if (ifnet == 0) {
369			error = EADDRNOTAVAIL;
370			break;
371		}
372		if ((addr->sin_family != AF_INET) &&
373		     (addr->sin_family != AF_IMPLINK)) {
374			error = EAFNOSUPPORT;
375			break;
376		}
377		inp->inp_faddr = addr->sin_addr;
378		soisconnected(so);
379		break;
380	    }
381
382	case PRU_CONNECT2:
383		error = EOPNOTSUPP;
384		break;
385
386	/*
387	 * Mark the connection as being incapable of further input.
388	 */
389	case PRU_SHUTDOWN:
390		socantsendmore(so);
391		break;
392
393	/*
394	 * Ship a packet out.  The appropriate raw output
395	 * routine handles any massaging necessary.
396	 */
397	case PRU_SEND:
398	    {
399		register u_long dst;
400
401		if (so->so_state & SS_ISCONNECTED) {
402			if (nam) {
403				error = EISCONN;
404				break;
405			}
406			dst = inp->inp_faddr.s_addr;
407		} else {
408			if (nam == NULL) {
409				error = ENOTCONN;
410				break;
411			}
412			dst = mtod(nam, struct sockaddr_in *)->sin_addr.s_addr;
413		}
414		error = rip_output(m, so, dst);
415		m = NULL;
416		break;
417	    }
418
419	case PRU_SENSE:
420		/*
421		 * stat: don't bother with a blocksize.
422		 */
423		return (0);
424
425	/*
426	 * Not supported.
427	 */
428	case PRU_RCVOOB:
429	case PRU_RCVD:
430	case PRU_LISTEN:
431	case PRU_ACCEPT:
432	case PRU_SENDOOB:
433		error = EOPNOTSUPP;
434		break;
435
436	case PRU_SOCKADDR:
437		in_setsockaddr(inp, nam);
438		break;
439
440	case PRU_PEERADDR:
441		in_setpeeraddr(inp, nam);
442		break;
443
444	default:
445		panic("rip_usrreq");
446	}
447	if (m != NULL)
448		m_freem(m);
449	return (error);
450}
451