raw_ip.c revision 17269
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 1.32 1996/07/20 00:16:20 alex Exp $ 35 */ 36 37#include <sys/param.h> 38#include <sys/queue.h> 39#include <sys/malloc.h> 40#include <sys/mbuf.h> 41#include <sys/socket.h> 42#include <sys/protosw.h> 43#include <sys/socketvar.h> 44#include <sys/errno.h> 45#include <sys/systm.h> 46 47#include <net/if.h> 48#include <net/route.h> 49 50#define _IP_VHL 51#include <netinet/in.h> 52#include <netinet/in_systm.h> 53#include <netinet/ip.h> 54#include <netinet/in_pcb.h> 55#include <netinet/in_var.h> 56#include <netinet/ip_var.h> 57#include <netinet/ip_mroute.h> 58 59#include <netinet/ip_fw.h> 60 61#if !defined(COMPAT_IPFW) || COMPAT_IPFW == 1 62#undef COMPAT_IPFW 63#define COMPAT_IPFW 1 64#else 65#undef COMPAT_IPFW 66#endif 67 68static struct inpcbhead ripcb; 69static struct inpcbinfo ripcbinfo; 70 71/* 72 * Nominal space allocated to a raw ip socket. 73 */ 74#define RIPSNDQ 8192 75#define RIPRCVQ 8192 76 77/* 78 * Raw interface to IP protocol. 79 */ 80 81/* 82 * Initialize raw connection block q. 83 */ 84void 85rip_init() 86{ 87 LIST_INIT(&ripcb); 88 ripcbinfo.listhead = &ripcb; 89 /* 90 * XXX We don't use the hash list for raw IP, but it's easier 91 * to allocate a one entry hash list than it is to check all 92 * over the place for hashbase == NULL. 93 */ 94 ripcbinfo.hashbase = phashinit(1, M_PCB, &ripcbinfo.hashsize); 95} 96 97static struct sockaddr_in ripsrc = { sizeof(ripsrc), AF_INET }; 98/* 99 * Setup generic address and protocol structures 100 * for raw_input routine, then pass them along with 101 * mbuf chain. 102 */ 103void 104rip_input(m, iphlen) 105 struct mbuf *m; 106 int iphlen; 107{ 108 register struct ip *ip = mtod(m, struct ip *); 109 register struct inpcb *inp; 110 struct socket *last = 0; 111 112 ripsrc.sin_addr = ip->ip_src; 113 for (inp = ripcb.lh_first; inp != NULL; inp = inp->inp_list.le_next) { 114 if (inp->inp_ip.ip_p && inp->inp_ip.ip_p != ip->ip_p) 115 continue; 116 if (inp->inp_laddr.s_addr && 117 inp->inp_laddr.s_addr != ip->ip_dst.s_addr) 118 continue; 119 if (inp->inp_faddr.s_addr && 120 inp->inp_faddr.s_addr != ip->ip_src.s_addr) 121 continue; 122 if (last) { 123 struct mbuf *n = m_copy(m, 0, (int)M_COPYALL); 124 if (n) { 125 if (sbappendaddr(&last->so_rcv, 126 (struct sockaddr *)&ripsrc, n, 127 (struct mbuf *)0) == 0) 128 /* should notify about lost packet */ 129 m_freem(n); 130 else 131 sorwakeup(last); 132 } 133 } 134 last = inp->inp_socket; 135 } 136 if (last) { 137 if (sbappendaddr(&last->so_rcv, (struct sockaddr *)&ripsrc, 138 m, (struct mbuf *)0) == 0) 139 m_freem(m); 140 else 141 sorwakeup(last); 142 } else { 143 m_freem(m); 144 ipstat.ips_noproto++; 145 ipstat.ips_delivered--; 146 } 147} 148 149/* 150 * Generate IP header and pass packet to ip_output. 151 * Tack on options user may have setup with control call. 152 */ 153int 154rip_output(m, so, dst) 155 register struct mbuf *m; 156 struct socket *so; 157 u_long dst; 158{ 159 register struct ip *ip; 160 register struct inpcb *inp = sotoinpcb(so); 161 int flags = (so->so_options & SO_DONTROUTE) | IP_ALLOWBROADCAST; 162 163 /* 164 * If the user handed us a complete IP packet, use it. 165 * Otherwise, allocate an mbuf for a header and fill it in. 166 */ 167 if ((inp->inp_flags & INP_HDRINCL) == 0) { 168 M_PREPEND(m, sizeof(struct ip), M_WAIT); 169 ip = mtod(m, struct ip *); 170 ip->ip_tos = 0; 171 ip->ip_off = 0; 172 ip->ip_p = inp->inp_ip.ip_p; 173 ip->ip_len = m->m_pkthdr.len; 174 ip->ip_src = inp->inp_laddr; 175 ip->ip_dst.s_addr = dst; 176 ip->ip_ttl = MAXTTL; 177 } else { 178 ip = mtod(m, struct ip *); 179 /* don't allow both user specified and setsockopt options, 180 and don't allow packet length sizes that will crash */ 181 if (((IP_VHL_HL(ip->ip_vhl) != (sizeof (*ip) >> 2)) 182 && inp->inp_options) 183 || (ip->ip_len > m->m_pkthdr.len)) { 184 m_freem(m); 185 return EINVAL; 186 } 187 if (ip->ip_id == 0) 188 ip->ip_id = htons(ip_id++); 189 /* XXX prevent ip_output from overwriting header fields */ 190 flags |= IP_RAWOUTPUT; 191 ipstat.ips_rawout++; 192 } 193 return (ip_output(m, inp->inp_options, &inp->inp_route, flags, 194 inp->inp_moptions)); 195} 196 197/* 198 * Raw IP socket option processing. 199 */ 200int 201rip_ctloutput(op, so, level, optname, m) 202 int op; 203 struct socket *so; 204 int level, optname; 205 struct mbuf **m; 206{ 207 register struct inpcb *inp = sotoinpcb(so); 208 register int error; 209 210 if (level != IPPROTO_IP) { 211 if (op == PRCO_SETOPT && *m) 212 (void)m_free(*m); 213 return (EINVAL); 214 } 215 216 switch (optname) { 217 218 case IP_HDRINCL: 219 error = 0; 220 if (op == PRCO_SETOPT) { 221 if (m == 0 || *m == 0 || (*m)->m_len < sizeof (int)) 222 error = EINVAL; 223 else if (*mtod(*m, int *)) 224 inp->inp_flags |= INP_HDRINCL; 225 else 226 inp->inp_flags &= ~INP_HDRINCL; 227 if (*m) 228 (void)m_free(*m); 229 } else { 230 *m = m_get(M_WAIT, MT_SOOPTS); 231 (*m)->m_len = sizeof (int); 232 *mtod(*m, int *) = inp->inp_flags & INP_HDRINCL; 233 } 234 return (error); 235 236#ifdef COMPAT_IPFW 237 case IP_FW_GET: 238 if (ip_fw_ctl_ptr==NULL || op == PRCO_SETOPT) { 239 if (*m) (void)m_free(*m); 240 return(EINVAL); 241 } 242 return (*ip_fw_ctl_ptr)(optname, m); 243 case IP_FW_ADD: 244 case IP_FW_DEL: 245 case IP_FW_FLUSH: 246 case IP_FW_ZERO: 247 if (ip_fw_ctl_ptr==NULL || op != PRCO_SETOPT) { 248 if (*m) (void)m_free(*m); 249 return(EINVAL); 250 } 251 252 return (*ip_fw_ctl_ptr)(optname, m); 253#endif 254 255 case IP_RSVP_ON: 256 return ip_rsvp_init(so); 257 break; 258 259 case IP_RSVP_OFF: 260 return ip_rsvp_done(); 261 break; 262 263 case IP_RSVP_VIF_ON: 264 return ip_rsvp_vif_init(so, *m); 265 266 case IP_RSVP_VIF_OFF: 267 return ip_rsvp_vif_done(so, *m); 268 269 case MRT_INIT: 270 case MRT_DONE: 271 case MRT_ADD_VIF: 272 case MRT_DEL_VIF: 273 case MRT_ADD_MFC: 274 case MRT_DEL_MFC: 275 case MRT_VERSION: 276 case MRT_ASSERT: 277 if (op == PRCO_SETOPT) { 278 error = ip_mrouter_set(optname, so, *m); 279 if (*m) 280 (void)m_free(*m); 281 } else if (op == PRCO_GETOPT) { 282 error = ip_mrouter_get(optname, so, m); 283 } else 284 error = EINVAL; 285 return (error); 286 } 287 return (ip_ctloutput(op, so, level, optname, m)); 288} 289 290static u_long rip_sendspace = RIPSNDQ; /* XXX sysctl ? */ 291static u_long rip_recvspace = RIPRCVQ; /* XXX sysctl ? */ 292 293/*ARGSUSED*/ 294int 295rip_usrreq(so, req, m, nam, control) 296 register struct socket *so; 297 int req; 298 struct mbuf *m, *nam, *control; 299{ 300 register int error = 0; 301 register struct inpcb *inp = sotoinpcb(so); 302 303 if (req == PRU_CONTROL) 304 return (in_control(so, (u_long)m, (caddr_t)nam, 305 (struct ifnet *)control)); 306 307 switch (req) { 308 309 case PRU_ATTACH: 310 if (inp) 311 panic("rip_attach"); 312 if ((so->so_state & SS_PRIV) == 0) { 313 error = EACCES; 314 break; 315 } 316 if ((error = soreserve(so, rip_sendspace, rip_recvspace)) || 317 (error = in_pcballoc(so, &ripcbinfo))) 318 break; 319 inp = (struct inpcb *)so->so_pcb; 320 inp->inp_ip.ip_p = (int)nam; 321 break; 322 323 case PRU_DISCONNECT: 324 if ((so->so_state & SS_ISCONNECTED) == 0) { 325 error = ENOTCONN; 326 break; 327 } 328 /* FALLTHROUGH */ 329 case PRU_ABORT: 330 soisdisconnected(so); 331 /* FALLTHROUGH */ 332 case PRU_DETACH: 333 if (inp == 0) 334 panic("rip_detach"); 335 if (so == ip_mrouter) 336 ip_mrouter_done(); 337 ip_rsvp_force_done(so); 338 if (so == ip_rsvpd) 339 ip_rsvp_done(); 340 in_pcbdetach(inp); 341 break; 342 343 case PRU_BIND: 344 { 345 struct sockaddr_in *addr = mtod(nam, struct sockaddr_in *); 346 347 if (nam->m_len != sizeof(*addr)) { 348 error = EINVAL; 349 break; 350 } 351 if ((ifnet == 0) || 352 ((addr->sin_family != AF_INET) && 353 (addr->sin_family != AF_IMPLINK)) || 354 (addr->sin_addr.s_addr && 355 ifa_ifwithaddr((struct sockaddr *)addr) == 0)) { 356 error = EADDRNOTAVAIL; 357 break; 358 } 359 inp->inp_laddr = addr->sin_addr; 360 break; 361 } 362 case PRU_CONNECT: 363 { 364 struct sockaddr_in *addr = mtod(nam, struct sockaddr_in *); 365 366 if (nam->m_len != sizeof(*addr)) { 367 error = EINVAL; 368 break; 369 } 370 if (ifnet == 0) { 371 error = EADDRNOTAVAIL; 372 break; 373 } 374 if ((addr->sin_family != AF_INET) && 375 (addr->sin_family != AF_IMPLINK)) { 376 error = EAFNOSUPPORT; 377 break; 378 } 379 inp->inp_faddr = addr->sin_addr; 380 soisconnected(so); 381 break; 382 } 383 384 case PRU_CONNECT2: 385 error = EOPNOTSUPP; 386 break; 387 388 /* 389 * Mark the connection as being incapable of further input. 390 */ 391 case PRU_SHUTDOWN: 392 socantsendmore(so); 393 break; 394 395 /* 396 * Ship a packet out. The appropriate raw output 397 * routine handles any massaging necessary. 398 */ 399 case PRU_SEND: 400 { 401 register u_long dst; 402 403 if (so->so_state & SS_ISCONNECTED) { 404 if (nam) { 405 error = EISCONN; 406 break; 407 } 408 dst = inp->inp_faddr.s_addr; 409 } else { 410 if (nam == NULL) { 411 error = ENOTCONN; 412 break; 413 } 414 dst = mtod(nam, struct sockaddr_in *)->sin_addr.s_addr; 415 } 416 error = rip_output(m, so, dst); 417 m = NULL; 418 break; 419 } 420 421 case PRU_SENSE: 422 /* 423 * stat: don't bother with a blocksize. 424 */ 425 return (0); 426 427 /* 428 * Not supported. 429 */ 430 case PRU_RCVOOB: 431 case PRU_RCVD: 432 case PRU_LISTEN: 433 case PRU_ACCEPT: 434 case PRU_SENDOOB: 435 error = EOPNOTSUPP; 436 break; 437 438 case PRU_SOCKADDR: 439 in_setsockaddr(inp, nam); 440 break; 441 442 case PRU_PEERADDR: 443 in_setpeeraddr(inp, nam); 444 break; 445 446 default: 447 panic("rip_usrreq"); 448 } 449 if (m != NULL) 450 m_freem(m); 451 return (error); 452} 453