| /* |
| * Copyright (c) 1982, 1986, 1988, 1993 |
| * The Regents of the University of California. All rights reserved. |
| * |
| * Redistribution and use in source and binary forms, with or without |
| * modification, are permitted provided that the following conditions |
| * are met: |
| * 1. Redistributions of source code must retain the above copyright |
| * notice, this list of conditions and the following disclaimer. |
| * 2. Redistributions in binary form must reproduce the above copyright |
| * notice, this list of conditions and the following disclaimer in the |
| * documentation and/or other materials provided with the distribution. |
| * 3. Neither the name of the University nor the names of its contributors |
| * may be used to endorse or promote products derived from this software |
| * without specific prior written permission. |
| * |
| * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND |
| * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
| * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE |
| * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE |
| * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL |
| * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS |
| * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) |
| * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT |
| * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY |
| * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF |
| * SUCH DAMAGE. |
| * |
| * @(#)ip_icmp.c 8.2 (Berkeley) 1/4/94 |
| * ip_icmp.c,v 1.7 1995/05/30 08:09:42 rgrimes Exp |
| */ |
| |
| #include "slirp.h" |
| #include "ip_icmp.h" |
| |
| /* The message sent when emulating PING */ |
| /* Be nice and tell them it's just a pseudo-ping packet */ |
| static const char icmp_ping_msg[] = |
| "This is a pseudo-PING packet used by Slirp to emulate ICMP ECHO-REQUEST " |
| "packets.\n"; |
| |
| /* list of actions for icmp_error() on RX of an icmp message */ |
| static const int icmp_flush[19] = { |
| /* ECHO REPLY (0) */ 0, |
| 1, |
| 1, |
| /* DEST UNREACH (3) */ 1, |
| /* SOURCE QUENCH (4)*/ 1, |
| /* REDIRECT (5) */ 1, |
| 1, |
| 1, |
| /* ECHO (8) */ 0, |
| /* ROUTERADVERT (9) */ 1, |
| /* ROUTERSOLICIT (10) */ 1, |
| /* TIME EXCEEDED (11) */ 1, |
| /* PARAMETER PROBLEM (12) */ 1, |
| /* TIMESTAMP (13) */ 0, |
| /* TIMESTAMP REPLY (14) */ 0, |
| /* INFO (15) */ 0, |
| /* INFO REPLY (16) */ 0, |
| /* ADDR MASK (17) */ 0, |
| /* ADDR MASK REPLY (18) */ 0 |
| }; |
| |
| void icmp_init(Slirp *slirp) |
| { |
| slirp->icmp.so_next = slirp->icmp.so_prev = &slirp->icmp; |
| slirp->icmp_last_so = &slirp->icmp; |
| } |
| |
| void icmp_cleanup(Slirp *slirp) |
| { |
| while (slirp->icmp.so_next != &slirp->icmp) { |
| icmp_detach(slirp->icmp.so_next); |
| } |
| } |
| |
| static int icmp_send(struct socket *so, struct mbuf *m, int hlen) |
| { |
| struct ip *ip = mtod(m, struct ip *); |
| struct sockaddr_in addr; |
| |
| so->s = qemu_socket(AF_INET, SOCK_DGRAM, IPPROTO_ICMP); |
| if (so->s == -1) { |
| return -1; |
| } |
| |
| so->so_m = m; |
| so->so_faddr = ip->ip_dst; |
| so->so_laddr = ip->ip_src; |
| so->so_iptos = ip->ip_tos; |
| so->so_type = IPPROTO_ICMP; |
| so->so_state = SS_ISFCONNECTED; |
| so->so_expire = curtime + SO_EXPIRE; |
| |
| addr.sin_family = AF_INET; |
| addr.sin_addr = so->so_faddr; |
| |
| insque(so, &so->slirp->icmp); |
| |
| if (sendto(so->s, m->m_data + hlen, m->m_len - hlen, 0, |
| (struct sockaddr *)&addr, sizeof(addr)) == -1) { |
| DEBUG_MISC((dfd, "icmp_input icmp sendto tx errno = %d-%s\n", errno, |
| strerror(errno))); |
| icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_NET, 0, strerror(errno)); |
| icmp_detach(so); |
| } |
| |
| return 0; |
| } |
| |
| void icmp_detach(struct socket *so) |
| { |
| closesocket(so->s); |
| sofree(so); |
| } |
| |
| /* |
| * Process a received ICMP message. |
| */ |
| void icmp_input(struct mbuf *m, int hlen) |
| { |
| register struct icmp *icp; |
| register struct ip *ip = mtod(m, struct ip *); |
| int icmplen = ip->ip_len; |
| Slirp *slirp = m->slirp; |
| |
| DEBUG_CALL("icmp_input"); |
| DEBUG_ARG("m = %lx", (long)m); |
| DEBUG_ARG("m_len = %d", m->m_len); |
| |
| /* |
| * Locate icmp structure in mbuf, and check |
| * that its not corrupted and of at least minimum length. |
| */ |
| if (icmplen < ICMP_MINLEN) { /* min 8 bytes payload */ |
| freeit: |
| m_free(m); |
| goto end_error; |
| } |
| |
| m->m_len -= hlen; |
| m->m_data += hlen; |
| icp = mtod(m, struct icmp *); |
| if (cksum(m, icmplen)) { |
| goto freeit; |
| } |
| m->m_len += hlen; |
| m->m_data -= hlen; |
| |
| DEBUG_ARG("icmp_type = %d", icp->icmp_type); |
| switch (icp->icmp_type) { |
| case ICMP_ECHO: |
| ip->ip_len += hlen; /* since ip_input subtracts this */ |
| if (ip->ip_dst.s_addr == slirp->vhost_addr.s_addr) { |
| icmp_reflect(m); |
| } else if (slirp->restricted) { |
| goto freeit; |
| } else { |
| struct socket *so; |
| struct sockaddr_in addr; |
| if ((so = socreate(slirp)) == NULL) |
| goto freeit; |
| if (icmp_send(so, m, hlen) == 0) { |
| return; |
| } |
| if (udp_attach(so) == -1) { |
| DEBUG_MISC((dfd, "icmp_input udp_attach errno = %d-%s\n", errno, |
| strerror(errno))); |
| sofree(so); |
| m_free(m); |
| goto end_error; |
| } |
| so->so_m = m; |
| so->so_faddr = ip->ip_dst; |
| so->so_fport = htons(7); |
| so->so_laddr = ip->ip_src; |
| so->so_lport = htons(9); |
| so->so_iptos = ip->ip_tos; |
| so->so_type = IPPROTO_ICMP; |
| so->so_state = SS_ISFCONNECTED; |
| |
| /* Send the packet */ |
| addr.sin_family = AF_INET; |
| if ((so->so_faddr.s_addr & slirp->vnetwork_mask.s_addr) == |
| slirp->vnetwork_addr.s_addr) { |
| /* It's an alias */ |
| if (so->so_faddr.s_addr == slirp->vnameserver_addr.s_addr) { |
| if (get_dns_addr(&addr.sin_addr) < 0) |
| addr.sin_addr = loopback_addr; |
| } else { |
| addr.sin_addr = loopback_addr; |
| } |
| } else { |
| addr.sin_addr = so->so_faddr; |
| } |
| addr.sin_port = so->so_fport; |
| if (sendto(so->s, icmp_ping_msg, strlen(icmp_ping_msg), 0, |
| (struct sockaddr *)&addr, sizeof(addr)) == -1) { |
| DEBUG_MISC((dfd, "icmp_input udp sendto tx errno = %d-%s\n", |
| errno, strerror(errno))); |
| icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_NET, 0, |
| strerror(errno)); |
| udp_detach(so); |
| } |
| } /* if ip->ip_dst.s_addr == alias_addr.s_addr */ |
| break; |
| case ICMP_UNREACH: |
| /* XXX? report error? close socket? */ |
| case ICMP_TIMXCEED: |
| case ICMP_PARAMPROB: |
| case ICMP_SOURCEQUENCH: |
| case ICMP_TSTAMP: |
| case ICMP_MASKREQ: |
| case ICMP_REDIRECT: |
| m_free(m); |
| break; |
| |
| default: |
| m_free(m); |
| } /* swith */ |
| |
| end_error: |
| /* m is m_free()'d xor put in a socket xor or given to ip_send */ |
| return; |
| } |
| |
| |
| /* |
| * Send an ICMP message in response to a situation |
| * |
| * RFC 1122: 3.2.2 MUST send at least the IP header and 8 bytes of header. |
| *MAY send more (we do). MUST NOT change this header information. MUST NOT reply |
| *to a multicast/broadcast IP address. MUST NOT reply to a multicast/broadcast |
| *MAC address. MUST reply to only the first fragment. |
| */ |
| /* |
| * Send ICMP_UNREACH back to the source regarding msrc. |
| * mbuf *msrc is used as a template, but is NOT m_free()'d. |
| * It is reported as the bad ip packet. The header should |
| * be fully correct and in host byte order. |
| * ICMP fragmentation is illegal. All machines must accept 576 bytes in one |
| * packet. The maximum payload is 576-20(ip hdr)-8(icmp hdr)=548 |
| */ |
| |
| #define ICMP_MAXDATALEN (IP_MSS - 28) |
| void icmp_error(struct mbuf *msrc, u_char type, u_char code, int minsize, |
| const char *message) |
| { |
| unsigned hlen, shlen, s_ip_len; |
| register struct ip *ip; |
| register struct icmp *icp; |
| register struct mbuf *m; |
| |
| DEBUG_CALL("icmp_error"); |
| DEBUG_ARG("msrc = %lx", (long)msrc); |
| DEBUG_ARG("msrc_len = %d", msrc->m_len); |
| |
| if (type != ICMP_UNREACH && type != ICMP_TIMXCEED) |
| goto end_error; |
| |
| /* check msrc */ |
| if (!msrc) |
| goto end_error; |
| ip = mtod(msrc, struct ip *); |
| #ifdef DEBUG |
| { |
| char bufa[20], bufb[20]; |
| strcpy(bufa, inet_ntoa(ip->ip_src)); |
| strcpy(bufb, inet_ntoa(ip->ip_dst)); |
| DEBUG_MISC((dfd, " %.16s to %.16s\n", bufa, bufb)); |
| } |
| #endif |
| if (ip->ip_off & IP_OFFMASK) |
| goto end_error; /* Only reply to fragment 0 */ |
| |
| /* Do not reply to source-only IPs */ |
| if ((ip->ip_src.s_addr & htonl(~(0xf << 28))) == 0) { |
| goto end_error; |
| } |
| |
| shlen = ip->ip_hl << 2; |
| s_ip_len = ip->ip_len; |
| if (ip->ip_p == IPPROTO_ICMP) { |
| icp = (struct icmp *)((char *)ip + shlen); |
| /* |
| * Assume any unknown ICMP type is an error. This isn't |
| * specified by the RFC, but think about it.. |
| */ |
| if (icp->icmp_type > 18 || icmp_flush[icp->icmp_type]) |
| goto end_error; |
| } |
| |
| /* make a copy */ |
| m = m_get(msrc->slirp); |
| if (!m) { |
| goto end_error; |
| } |
| |
| { |
| int new_m_size; |
| new_m_size = |
| sizeof(struct ip) + ICMP_MINLEN + msrc->m_len + ICMP_MAXDATALEN; |
| if (new_m_size > m->m_size) |
| m_inc(m, new_m_size); |
| } |
| memcpy(m->m_data, msrc->m_data, msrc->m_len); |
| m->m_len = msrc->m_len; /* copy msrc to m */ |
| |
| /* make the header of the reply packet */ |
| ip = mtod(m, struct ip *); |
| hlen = sizeof(struct ip); /* no options in reply */ |
| |
| /* fill in icmp */ |
| m->m_data += hlen; |
| m->m_len -= hlen; |
| |
| icp = mtod(m, struct icmp *); |
| |
| if (minsize) |
| s_ip_len = shlen + ICMP_MINLEN; /* return header+8b only */ |
| else if (s_ip_len > ICMP_MAXDATALEN) /* maximum size */ |
| s_ip_len = ICMP_MAXDATALEN; |
| |
| m->m_len = ICMP_MINLEN + s_ip_len; /* 8 bytes ICMP header */ |
| |
| /* min. size = 8+sizeof(struct ip)+8 */ |
| |
| icp->icmp_type = type; |
| icp->icmp_code = code; |
| icp->icmp_id = 0; |
| icp->icmp_seq = 0; |
| |
| memcpy(&icp->icmp_ip, msrc->m_data, s_ip_len); /* report the ip packet */ |
| HTONS(icp->icmp_ip.ip_len); |
| HTONS(icp->icmp_ip.ip_id); |
| HTONS(icp->icmp_ip.ip_off); |
| |
| #ifdef DEBUG |
| if (message) { /* DEBUG : append message to ICMP packet */ |
| int message_len; |
| char *cpnt; |
| message_len = strlen(message); |
| if (message_len > ICMP_MAXDATALEN) |
| message_len = ICMP_MAXDATALEN; |
| cpnt = (char *)m->m_data + m->m_len; |
| memcpy(cpnt, message, message_len); |
| m->m_len += message_len; |
| } |
| #endif |
| |
| icp->icmp_cksum = 0; |
| icp->icmp_cksum = cksum(m, m->m_len); |
| |
| m->m_data -= hlen; |
| m->m_len += hlen; |
| |
| /* fill in ip */ |
| ip->ip_hl = hlen >> 2; |
| ip->ip_len = m->m_len; |
| |
| ip->ip_tos = ((ip->ip_tos & 0x1E) | 0xC0); /* high priority for errors */ |
| |
| ip->ip_ttl = MAXTTL; |
| ip->ip_p = IPPROTO_ICMP; |
| ip->ip_dst = ip->ip_src; /* ip addresses */ |
| ip->ip_src = m->slirp->vhost_addr; |
| |
| (void)ip_output((struct socket *)NULL, m); |
| |
| end_error: |
| return; |
| } |
| #undef ICMP_MAXDATALEN |
| |
| /* |
| * Reflect the ip packet back to the source |
| */ |
| void icmp_reflect(struct mbuf *m) |
| { |
| register struct ip *ip = mtod(m, struct ip *); |
| int hlen = ip->ip_hl << 2; |
| int optlen = hlen - sizeof(struct ip); |
| register struct icmp *icp; |
| |
| /* |
| * Send an icmp packet back to the ip level, |
| * after supplying a checksum. |
| */ |
| m->m_data += hlen; |
| m->m_len -= hlen; |
| icp = mtod(m, struct icmp *); |
| |
| icp->icmp_type = ICMP_ECHOREPLY; |
| icp->icmp_cksum = 0; |
| icp->icmp_cksum = cksum(m, ip->ip_len - hlen); |
| |
| m->m_data -= hlen; |
| m->m_len += hlen; |
| |
| /* fill in ip */ |
| if (optlen > 0) { |
| /* |
| * Strip out original options by copying rest of first |
| * mbuf's data back, and adjust the IP length. |
| */ |
| memmove((caddr_t)(ip + 1), (caddr_t)ip + hlen, |
| (unsigned)(m->m_len - hlen)); |
| hlen -= optlen; |
| ip->ip_hl = hlen >> 2; |
| ip->ip_len -= optlen; |
| m->m_len -= optlen; |
| } |
| |
| ip->ip_ttl = MAXTTL; |
| { /* swap */ |
| struct in_addr icmp_dst; |
| icmp_dst = ip->ip_dst; |
| ip->ip_dst = ip->ip_src; |
| ip->ip_src = icmp_dst; |
| } |
| |
| (void)ip_output((struct socket *)NULL, m); |
| } |
| |
| void icmp_receive(struct socket *so) |
| { |
| struct mbuf *m = so->so_m; |
| struct ip *ip = mtod(m, struct ip *); |
| int hlen = ip->ip_hl << 2; |
| u_char error_code; |
| struct icmp *icp; |
| int id, len; |
| |
| m->m_data += hlen; |
| m->m_len -= hlen; |
| icp = mtod(m, struct icmp *); |
| |
| id = icp->icmp_id; |
| len = qemu_recv(so->s, icp, m->m_len, 0); |
| icp->icmp_id = id; |
| |
| m->m_data -= hlen; |
| m->m_len += hlen; |
| |
| if (len == -1 || len == 0) { |
| if (errno == ENETUNREACH) { |
| error_code = ICMP_UNREACH_NET; |
| } else { |
| error_code = ICMP_UNREACH_HOST; |
| } |
| DEBUG_MISC( |
| (dfd, " udp icmp rx errno = %d-%s\n", errno, strerror(errno))); |
| icmp_error(so->so_m, ICMP_UNREACH, error_code, 0, strerror(errno)); |
| } else { |
| icmp_reflect(so->so_m); |
| so->so_m = NULL; /* Don't m_free() it again! */ |
| } |
| icmp_detach(so); |
| } |