raw_ip.c revision 14232
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	8.7 (Berkeley) 5/15/95
34 *	$Id: raw_ip.c,v 1.26 1996/02/23 15:47:58 phk 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/in_pcb.h>
54#include <netinet/in_var.h>
55#include <netinet/ip_var.h>
56#include <netinet/ip_mroute.h>
57
58#include <netinet/ip_fw.h>
59
60static struct inpcbhead ripcb;
61static struct inpcbinfo ripcbinfo;
62
63/*
64 * Nominal space allocated to a raw ip socket.
65 */
66#define	RIPSNDQ		8192
67#define	RIPRCVQ		8192
68
69/*
70 * Raw interface to IP protocol.
71 */
72
73/*
74 * Initialize raw connection block q.
75 */
76void
77rip_init()
78{
79	LIST_INIT(&ripcb);
80	ripcbinfo.listhead = &ripcb;
81	/*
82	 * XXX We don't use the hash list for raw IP, but it's easier
83	 * to allocate a one entry hash list than it is to check all
84	 * over the place for hashbase == NULL.
85	 */
86	ripcbinfo.hashbase = phashinit(1, M_PCB, &ripcbinfo.hashsize);
87}
88
89static struct	sockaddr_in ripsrc = { sizeof(ripsrc), AF_INET };
90/*
91 * Setup generic address and protocol structures
92 * for raw_input routine, then pass them along with
93 * mbuf chain.
94 */
95void
96rip_input(m)
97	struct mbuf *m;
98{
99	register struct ip *ip = mtod(m, struct ip *);
100	register struct inpcb *inp;
101	struct socket *last = 0;
102
103	ripsrc.sin_addr = ip->ip_src;
104	for (inp = ripcb.lh_first; inp != NULL; inp = inp->inp_list.le_next) {
105		if (inp->inp_ip.ip_p && inp->inp_ip.ip_p != ip->ip_p)
106			continue;
107		if (inp->inp_laddr.s_addr &&
108                  inp->inp_laddr.s_addr != ip->ip_dst.s_addr)
109			continue;
110		if (inp->inp_faddr.s_addr &&
111                  inp->inp_faddr.s_addr != ip->ip_src.s_addr)
112			continue;
113		if (last) {
114			struct mbuf *n = m_copy(m, 0, (int)M_COPYALL);
115			if (n) {
116				if (sbappendaddr(&last->so_rcv,
117				    (struct sockaddr *)&ripsrc, n,
118				    (struct mbuf *)0) == 0)
119					/* should notify about lost packet */
120					m_freem(n);
121				else
122					sorwakeup(last);
123			}
124		}
125		last = inp->inp_socket;
126	}
127	if (last) {
128		if (sbappendaddr(&last->so_rcv, (struct sockaddr *)&ripsrc,
129		    m, (struct mbuf *)0) == 0)
130			m_freem(m);
131		else
132			sorwakeup(last);
133	} else {
134		m_freem(m);
135              ipstat.ips_noproto++;
136              ipstat.ips_delivered--;
137      }
138}
139
140/*
141 * Generate IP header and pass packet to ip_output.
142 * Tack on options user may have setup with control call.
143 */
144int
145rip_output(m, so, dst)
146	register struct mbuf *m;
147	struct socket *so;
148	u_long dst;
149{
150	register struct ip *ip;
151	register struct inpcb *inp = sotoinpcb(so);
152	struct mbuf *opts;
153	int flags = (so->so_options & SO_DONTROUTE) | IP_ALLOWBROADCAST;
154
155	/*
156	 * If the user handed us a complete IP packet, use it.
157	 * Otherwise, allocate an mbuf for a header and fill it in.
158	 */
159	if ((inp->inp_flags & INP_HDRINCL) == 0) {
160		M_PREPEND(m, sizeof(struct ip), M_WAIT);
161		ip = mtod(m, struct ip *);
162		ip->ip_tos = 0;
163		ip->ip_off = 0;
164		ip->ip_p = inp->inp_ip.ip_p;
165		ip->ip_len = m->m_pkthdr.len;
166		ip->ip_src = inp->inp_laddr;
167		ip->ip_dst.s_addr = dst;
168		ip->ip_ttl = MAXTTL;
169		opts = inp->inp_options;
170	} else {
171		ip = mtod(m, struct ip *);
172		if (ip->ip_id == 0)
173			ip->ip_id = htons(ip_id++);
174		opts = NULL;
175		/* XXX prevent ip_output from overwriting header fields */
176		flags |= IP_RAWOUTPUT;
177		ipstat.ips_rawout++;
178	}
179	return (ip_output(m, opts, &inp->inp_route, flags, inp->inp_moptions));
180}
181
182/*
183 * Raw IP socket option processing.
184 */
185int
186rip_ctloutput(op, so, level, optname, m)
187	int op;
188	struct socket *so;
189	int level, optname;
190	struct mbuf **m;
191{
192	register struct inpcb *inp = sotoinpcb(so);
193	register int error;
194
195	if (level != IPPROTO_IP) {
196		if (op == PRCO_SETOPT && *m)
197			(void)m_free(*m);
198		return (EINVAL);
199	}
200
201	switch (optname) {
202
203	case IP_HDRINCL:
204		error = 0;
205		if (op == PRCO_SETOPT) {
206			if (m == 0 || *m == 0 || (*m)->m_len < sizeof (int))
207				error = EINVAL;
208			else if (*mtod(*m, int *))
209				inp->inp_flags |= INP_HDRINCL;
210			else
211				inp->inp_flags &= ~INP_HDRINCL;
212			if (*m)
213				(void)m_free(*m);
214		} else {
215			*m = m_get(M_WAIT, MT_SOOPTS);
216			(*m)->m_len = sizeof (int);
217			*mtod(*m, int *) = inp->inp_flags & INP_HDRINCL;
218		}
219		return (error);
220
221	case IP_FW_GET:
222		if (ip_fw_ctl_ptr==NULL || op == PRCO_SETOPT) {
223			if (*m) (void)m_free(*m);
224			return(EINVAL);
225		}
226		return (*ip_fw_ctl_ptr)(optname, m);
227	case IP_FW_ADD:
228	case IP_FW_DEL:
229	case IP_FW_FLUSH:
230	case IP_FW_ZERO:
231		if (ip_fw_ctl_ptr==NULL || op != PRCO_SETOPT) {
232			if (*m) (void)m_free(*m);
233			return(EINVAL);
234		}
235
236		return (*ip_fw_ctl_ptr)(optname, m);
237		return(error);
238
239	case IP_RSVP_ON:
240		return ip_rsvp_init(so);
241		break;
242
243	case IP_RSVP_OFF:
244		return ip_rsvp_done();
245		break;
246
247	case IP_RSVP_VIF_ON:
248		return ip_rsvp_vif_init(so, *m);
249
250	case IP_RSVP_VIF_OFF:
251		return ip_rsvp_vif_done(so, *m);
252
253	case MRT_INIT:
254	case MRT_DONE:
255	case MRT_ADD_VIF:
256	case MRT_DEL_VIF:
257	case MRT_ADD_MFC:
258	case MRT_DEL_MFC:
259	case MRT_VERSION:
260	case MRT_ASSERT:
261		if (op == PRCO_SETOPT) {
262			error = ip_mrouter_set(optname, so, *m);
263			if (*m)
264				(void)m_free(*m);
265		} else if (op == PRCO_GETOPT) {
266			error = ip_mrouter_get(optname, so, m);
267		} else
268			error = EINVAL;
269		return (error);
270	}
271	return (ip_ctloutput(op, so, level, optname, m));
272}
273
274static u_long	rip_sendspace = RIPSNDQ; /* XXX sysctl ? */
275static u_long	rip_recvspace = RIPRCVQ; /* XXX sysctl ? */
276
277/*ARGSUSED*/
278int
279rip_usrreq(so, req, m, nam, control)
280	register struct socket *so;
281	int req;
282	struct mbuf *m, *nam, *control;
283{
284	register int error = 0;
285	register struct inpcb *inp = sotoinpcb(so);
286
287	if (req == PRU_CONTROL)
288		return (in_control(so, (u_long)m, (caddr_t)nam,
289			(struct ifnet *)control));
290
291	switch (req) {
292
293	case PRU_ATTACH:
294		if (inp)
295			panic("rip_attach");
296		if ((so->so_state & SS_PRIV) == 0) {
297			error = EACCES;
298			break;
299		}
300		if ((error = soreserve(so, rip_sendspace, rip_recvspace)) ||
301		    (error = in_pcballoc(so, &ripcbinfo)))
302			break;
303		inp = (struct inpcb *)so->so_pcb;
304		inp->inp_ip.ip_p = (int)nam;
305		break;
306
307	case PRU_DISCONNECT:
308		if ((so->so_state & SS_ISCONNECTED) == 0) {
309			error = ENOTCONN;
310			break;
311		}
312		/* FALLTHROUGH */
313	case PRU_ABORT:
314		soisdisconnected(so);
315		/* FALLTHROUGH */
316	case PRU_DETACH:
317		if (inp == 0)
318			panic("rip_detach");
319		if (so == ip_mrouter)
320			ip_mrouter_done();
321		ip_rsvp_force_done(so);
322		if (so == ip_rsvpd)
323			ip_rsvp_done();
324		in_pcbdetach(inp);
325		break;
326
327	case PRU_BIND:
328	    {
329		struct sockaddr_in *addr = mtod(nam, struct sockaddr_in *);
330
331		if (nam->m_len != sizeof(*addr)) {
332			error = EINVAL;
333			break;
334		}
335		if ((ifnet == 0) ||
336		    ((addr->sin_family != AF_INET) &&
337		     (addr->sin_family != AF_IMPLINK)) ||
338		    (addr->sin_addr.s_addr &&
339		     ifa_ifwithaddr((struct sockaddr *)addr) == 0)) {
340			error = EADDRNOTAVAIL;
341			break;
342		}
343		inp->inp_laddr = addr->sin_addr;
344		break;
345	    }
346	case PRU_CONNECT:
347	    {
348		struct sockaddr_in *addr = mtod(nam, struct sockaddr_in *);
349
350		if (nam->m_len != sizeof(*addr)) {
351			error = EINVAL;
352			break;
353		}
354		if (ifnet == 0) {
355			error = EADDRNOTAVAIL;
356			break;
357		}
358		if ((addr->sin_family != AF_INET) &&
359		     (addr->sin_family != AF_IMPLINK)) {
360			error = EAFNOSUPPORT;
361			break;
362		}
363		inp->inp_faddr = addr->sin_addr;
364		soisconnected(so);
365		break;
366	    }
367
368	case PRU_CONNECT2:
369		error = EOPNOTSUPP;
370		break;
371
372	/*
373	 * Mark the connection as being incapable of further input.
374	 */
375	case PRU_SHUTDOWN:
376		socantsendmore(so);
377		break;
378
379	/*
380	 * Ship a packet out.  The appropriate raw output
381	 * routine handles any massaging necessary.
382	 */
383	case PRU_SEND:
384	    {
385		register u_long dst;
386
387		if (so->so_state & SS_ISCONNECTED) {
388			if (nam) {
389				error = EISCONN;
390				break;
391			}
392			dst = inp->inp_faddr.s_addr;
393		} else {
394			if (nam == NULL) {
395				error = ENOTCONN;
396				break;
397			}
398			dst = mtod(nam, struct sockaddr_in *)->sin_addr.s_addr;
399		}
400		error = rip_output(m, so, dst);
401		m = NULL;
402		break;
403	    }
404
405	case PRU_SENSE:
406		/*
407		 * stat: don't bother with a blocksize.
408		 */
409		return (0);
410
411	/*
412	 * Not supported.
413	 */
414	case PRU_RCVOOB:
415	case PRU_RCVD:
416	case PRU_LISTEN:
417	case PRU_ACCEPT:
418	case PRU_SENDOOB:
419		error = EOPNOTSUPP;
420		break;
421
422	case PRU_SOCKADDR:
423		in_setsockaddr(inp, nam);
424		break;
425
426	case PRU_PEERADDR:
427		in_setpeeraddr(inp, nam);
428		break;
429
430	default:
431		panic("rip_usrreq");
432	}
433	if (m != NULL)
434		m_freem(m);
435	return (error);
436}
437