raw_ip.c revision 7684
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 1.16 1995/03/16 16:25:43 wollman 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 139void rip_ip_input(mm, ip_mrouter, src) 140 struct mbuf *mm; 141 register struct socket *ip_mrouter; 142 struct sockaddr *src; 143{ 144 if (ip_mrouter) 145 { 146 if (sbappendaddr(&ip_mrouter->so_rcv, src, 147 mm, (struct mbuf *) 0) == 0) 148 m_freem(mm); 149 else 150 sorwakeup(ip_mrouter); 151 } 152 else 153 { 154 m_freem(mm); 155 ipstat.ips_noproto++; 156 ipstat.ips_delivered--; 157 } 158} 159 160/* 161 * Generate IP header and pass packet to ip_output. 162 * Tack on options user may have setup with control call. 163 */ 164int 165rip_output(m, so, dst) 166 register struct mbuf *m; 167 struct socket *so; 168 u_long dst; 169{ 170 register struct ip *ip; 171 register struct inpcb *inp = sotoinpcb(so); 172 struct mbuf *opts; 173 int flags = (so->so_options & SO_DONTROUTE) | IP_ALLOWBROADCAST; 174 175 /* 176 * If the user handed us a complete IP packet, use it. 177 * Otherwise, allocate an mbuf for a header and fill it in. 178 */ 179 if ((inp->inp_flags & INP_HDRINCL) == 0) { 180 M_PREPEND(m, sizeof(struct ip), M_WAIT); 181 ip = mtod(m, struct ip *); 182 ip->ip_tos = 0; 183 ip->ip_off = 0; 184 ip->ip_p = inp->inp_ip.ip_p; 185 ip->ip_len = m->m_pkthdr.len; 186 ip->ip_src = inp->inp_laddr; 187 ip->ip_dst.s_addr = dst; 188 ip->ip_ttl = MAXTTL; 189 opts = inp->inp_options; 190 } else { 191 ip = mtod(m, struct ip *); 192 if (ip->ip_id == 0) 193 ip->ip_id = htons(ip_id++); 194 opts = NULL; 195 /* XXX prevent ip_output from overwriting header fields */ 196 flags |= IP_RAWOUTPUT; 197 ipstat.ips_rawout++; 198 } 199 return (ip_output(m, opts, &inp->inp_route, flags, inp->inp_moptions)); 200} 201 202/* 203 * Raw IP socket option processing. 204 */ 205int 206rip_ctloutput(op, so, level, optname, m) 207 int op; 208 struct socket *so; 209 int level, optname; 210 struct mbuf **m; 211{ 212 register struct inpcb *inp = sotoinpcb(so); 213 register int error; 214 215 if (level != IPPROTO_IP) { 216 if (op == PRCO_SETOPT && *m) 217 (void)m_free(*m); 218 return (EINVAL); 219 } 220 221 switch (optname) { 222 223 case IP_HDRINCL: 224 if (op == PRCO_SETOPT || op == PRCO_GETOPT) { 225 if (m == 0 || *m == 0 || (*m)->m_len < sizeof (int)) 226 return (EINVAL); 227 if (op == PRCO_SETOPT) { 228 if (*mtod(*m, int *)) 229 inp->inp_flags |= INP_HDRINCL; 230 else 231 inp->inp_flags &= ~INP_HDRINCL; 232 (void)m_free(*m); 233 } else { 234 (*m)->m_len = sizeof (int); 235 *mtod(*m, int *) = inp->inp_flags & INP_HDRINCL; 236 } 237 return (0); 238 } 239 break; 240 241 case IP_FW_ADD: 242 case IP_FW_DEL: 243 case IP_FW_FLUSH: 244 case IP_FW_POLICY: 245 if (ip_fw_ctl_ptr==NULL) { 246 if (*m) 247 (void)m_free(*m); 248 return(EINVAL); 249 } 250 251 if (op == PRCO_SETOPT) { 252 error=(*ip_fw_ctl_ptr)(optname, *m); 253 if (*m) 254 (void)m_free(*m); 255 } 256 else 257 error=EINVAL; 258 return(error); 259 260 case IP_ACCT_DEL: 261 case IP_ACCT_ADD: 262 case IP_ACCT_CLR: 263 case IP_ACCT_FLUSH: 264 case IP_ACCT_ZERO: 265 if (ip_acct_ctl_ptr==NULL) { 266 if (*m) 267 (void)m_free(*m); 268 return(EINVAL); 269 } 270 271 op = PRCO_SETOPT; 272 if (op) { 273 error=(*ip_acct_ctl_ptr)(optname, *m); 274 if (*m) 275 (void)m_free(*m); 276 } 277 else 278 error=EINVAL; 279 return(error); 280 281 case IP_RSVP_ON: 282 return ip_rsvp_init(so); 283 break; 284 285 case IP_RSVP_OFF: 286 return ip_rsvp_done(); 287 break; 288 289 case DVMRP_INIT: 290 case DVMRP_DONE: 291 case DVMRP_ADD_VIF: 292 case DVMRP_DEL_VIF: 293 case DVMRP_ADD_MFC: 294 case DVMRP_DEL_MFC: 295 if (op == PRCO_SETOPT) { 296 error = ip_mrouter_cmd(optname, so, *m); 297 if (*m) 298 (void)m_free(*m); 299 } else 300 error = EINVAL; 301 return (error); 302 } 303 return (ip_ctloutput(op, so, level, optname, m)); 304} 305 306u_long rip_sendspace = RIPSNDQ; 307u_long rip_recvspace = RIPRCVQ; 308 309/*ARGSUSED*/ 310int 311rip_usrreq(so, req, m, nam, control) 312 register struct socket *so; 313 int req; 314 struct mbuf *m, *nam, *control; 315{ 316 register int error = 0; 317 register struct inpcb *inp = sotoinpcb(so); 318 switch (req) { 319 320 case PRU_ATTACH: 321 if (inp) 322 panic("rip_attach"); 323 if ((so->so_state & SS_PRIV) == 0) { 324 error = EACCES; 325 break; 326 } 327 if ((error = soreserve(so, rip_sendspace, rip_recvspace)) || 328 (error = in_pcballoc(so, &ripcbinfo))) 329 break; 330 inp = (struct inpcb *)so->so_pcb; 331 inp->inp_ip.ip_p = (int)nam; 332 break; 333 334 case PRU_DISCONNECT: 335 if ((so->so_state & SS_ISCONNECTED) == 0) { 336 error = ENOTCONN; 337 break; 338 } 339 /* FALLTHROUGH */ 340 case PRU_ABORT: 341 soisdisconnected(so); 342 /* FALLTHROUGH */ 343 case PRU_DETACH: 344 if (inp == 0) 345 panic("rip_detach"); 346 if (so == ip_mrouter) 347 ip_mrouter_done(); 348 if (so == ip_rsvpd) 349 ip_rsvp_done(); 350 in_pcbdetach(inp); 351 break; 352 353 case PRU_BIND: 354 { 355 struct sockaddr_in *addr = mtod(nam, struct sockaddr_in *); 356 357 if (nam->m_len != sizeof(*addr)) { 358 error = EINVAL; 359 break; 360 } 361 if ((ifnet == 0) || 362 ((addr->sin_family != AF_INET) && 363 (addr->sin_family != AF_IMPLINK)) || 364 (addr->sin_addr.s_addr && 365 ifa_ifwithaddr((struct sockaddr *)addr) == 0)) { 366 error = EADDRNOTAVAIL; 367 break; 368 } 369 inp->inp_laddr = addr->sin_addr; 370 break; 371 } 372 case PRU_CONNECT: 373 { 374 struct sockaddr_in *addr = mtod(nam, struct sockaddr_in *); 375 376 if (nam->m_len != sizeof(*addr)) { 377 error = EINVAL; 378 break; 379 } 380 if (ifnet == 0) { 381 error = EADDRNOTAVAIL; 382 break; 383 } 384 if ((addr->sin_family != AF_INET) && 385 (addr->sin_family != AF_IMPLINK)) { 386 error = EAFNOSUPPORT; 387 break; 388 } 389 inp->inp_faddr = addr->sin_addr; 390 soisconnected(so); 391 break; 392 } 393 394 case PRU_CONNECT2: 395 error = EOPNOTSUPP; 396 break; 397 398 /* 399 * Mark the connection as being incapable of further input. 400 */ 401 case PRU_SHUTDOWN: 402 socantsendmore(so); 403 break; 404 405 /* 406 * Ship a packet out. The appropriate raw output 407 * routine handles any massaging necessary. 408 */ 409 case PRU_SEND: 410 { 411 register u_long dst; 412 413 if (so->so_state & SS_ISCONNECTED) { 414 if (nam) { 415 error = EISCONN; 416 break; 417 } 418 dst = inp->inp_faddr.s_addr; 419 } else { 420 if (nam == NULL) { 421 error = ENOTCONN; 422 break; 423 } 424 dst = mtod(nam, struct sockaddr_in *)->sin_addr.s_addr; 425 } 426 error = rip_output(m, so, dst); 427 m = NULL; 428 break; 429 } 430 431 case PRU_SENSE: 432 /* 433 * stat: don't bother with a blocksize. 434 */ 435 return (0); 436 437 /* 438 * Not supported. 439 */ 440 case PRU_RCVOOB: 441 case PRU_RCVD: 442 case PRU_LISTEN: 443 case PRU_ACCEPT: 444 case PRU_SENDOOB: 445 error = EOPNOTSUPP; 446 break; 447 448 case PRU_SOCKADDR: 449 in_setsockaddr(inp, nam); 450 break; 451 452 case PRU_PEERADDR: 453 in_setpeeraddr(inp, nam); 454 break; 455 456 default: 457 panic("rip_usrreq"); 458 } 459 if (m != NULL) 460 m_freem(m); 461 return (error); 462} 463