/*
* Read from so's socket into sb_snd, updating all relevant sbuf fields
* NOTE: This will only be called if it is select()ed for reading, so
* a read() of 0 (or less) means it's disconnected
*/
int
soread(struct socket *so)
{
int n, nn;
struct sbuf *sb = &so->so_snd;
struct iovec iov[2];
DEBUG_CALL("soread");
DEBUG_ARG("so = %lx", (long )so);
/*
* No need to check if there's enough room to read.
* soread wouldn't have been called if there weren't
*/
sopreprbuf(so, iov, &n);
#ifdef HAVE_READV
nn = readv(so->s, (struct iovec *)iov, n);
DEBUG_MISC((dfd, " ... read nn = %d bytes\n", nn));
#else
nn = qemu_recv(so->s, iov[0].iov_base, iov[0].iov_len,0);
#endif
if (nn <= 0) {
if (nn < 0 && (errno == EINTR || errno == EAGAIN))
return 0;
else {
DEBUG_MISC((dfd, " --- soread() disconnected, nn = %d, errno = %d-%s\n", nn, errno,strerror(errno)));
sofcantrcvmore(so);
tcp_sockclosed(sototcpcb(so));
return -1;
}
}
#ifndef HAVE_READV
/*
* If there was no error, try and read the second time round
* We read again if n = 2 (ie, there's another part of the buffer)
* and we read as much as we could in the first read
* We don't test for <= 0 this time, because there legitimately
* might not be any more data (since the socket is non-blocking),
* a close will be detected on next iteration.
* A return of -1 wont (shouldn't) happen, since it didn't happen above
*/
if (n == 2 && nn == iov[0].iov_len) {
int ret;
ret = qemu_recv(so->s, iov[1].iov_base, iov[1].iov_len,0);
if (ret > 0)
nn += ret;
}
DEBUG_MISC((dfd, " ... read nn = %d bytes\n", nn));
#endif
/* Update fields */
sb->sb_cc += nn;
sb->sb_wptr += nn;
if (sb->sb_wptr >= (sb->sb_data + sb->sb_datalen))
sb->sb_wptr -= sb->sb_datalen;
return nn;
}
/**
* Analyze the info request message sent by the client to see what data it
* provided and what it wants to have. The information is gathered in the
* "requested_infos" struct. Note that client_id (if provided) points into
* the odata region, thus the caller must keep odata valid as long as it
* needs to access the requested_infos struct.
*/
static int dhcpv6_parse_info_request(uint8_t *odata, int olen,
struct requested_infos *ri)
{
int i, req_opt;
while (olen > 4) {
/* Parse one option */
int option = odata[0] << 8 | odata[1];
int len = odata[2] << 8 | odata[3];
if (len + 4 > olen) {
qemu_log_mask(LOG_GUEST_ERROR, "Guest sent bad DHCPv6 packet!\n");
return -E2BIG;
}
switch (option) {
case OPTION_IAADDR:
/* According to RFC3315, we must discard requests with IA option */
return -EINVAL;
case OPTION_CLIENTID:
if (len > 256) {
/* Avoid very long IDs which could cause problems later */
return -E2BIG;
}
ri->client_id = odata + 4;
ri->client_id_len = len;
break;
case OPTION_ORO: /* Option request option */
if (len & 1) {
return -EINVAL;
}
/* Check which options the client wants to have */
for (i = 0; i < len; i += 2) {
req_opt = odata[4 + i] << 8 | odata[4 + i + 1];
switch (req_opt) {
case OPTION_DNS_SERVERS:
ri->want_dns = true;
break;
case OPTION_BOOTFILE_URL:
ri->want_boot_url = true;
break;
default:
DEBUG_MISC((dfd, "dhcpv6: Unsupported option request %d\n",
req_opt));
}
}
break;
default:
DEBUG_MISC((dfd, "dhcpv6 info req: Unsupported option %d, len=%d\n",
option, len));
}
odata += len + 4;
olen -= len + 4;
}
return 0;
}
/*
* Send urgent data
* There's a lot duplicated code here, but...
*/
int
sosendoob(struct socket *so)
{
struct sbuf *sb = &so->so_rcv;
char buff[2048]; /* XXX Shouldn't be sending more oob data than this */
int n, len;
DEBUG_CALL("sosendoob");
DEBUG_ARG("so = %lx", (long)so);
DEBUG_ARG("sb->sb_cc = %d", sb->sb_cc);
if (so->so_urgc > 2048)
so->so_urgc = 2048; /* XXXX */
if (sb->sb_rptr < sb->sb_wptr) {
/* We can send it directly */
n = slirp_send(so, sb->sb_rptr, so->so_urgc, (MSG_OOB)); /* |MSG_DONTWAIT)); */
so->so_urgc -= n;
DEBUG_MISC((dfd, " --- sent %d bytes urgent data, %d urgent bytes left\n", n, so->so_urgc));
} else {
/*
* Since there's no sendv or sendtov like writev,
* we must copy all data to a linear buffer then
* send it all
*/
len = (sb->sb_data + sb->sb_datalen) - sb->sb_rptr;
if (len > so->so_urgc) len = so->so_urgc;
memcpy(buff, sb->sb_rptr, len);
so->so_urgc -= len;
if (so->so_urgc) {
n = sb->sb_wptr - sb->sb_data;
if (n > so->so_urgc) n = so->so_urgc;
memcpy((buff + len), sb->sb_data, n);
so->so_urgc -= n;
len += n;
}
n = slirp_send(so, buff, len, (MSG_OOB)); /* |MSG_DONTWAIT)); */
#ifdef DEBUG
if (n != len)
DEBUG_ERROR((dfd, "Didn't send all data urgently XXXXX\n"));
#endif
DEBUG_MISC((dfd, " ---2 sent %d bytes urgent data, %d urgent bytes left\n", n, so->so_urgc));
}
sb->sb_cc -= n;
sb->sb_rptr += n;
if (sb->sb_rptr >= (sb->sb_data + sb->sb_datalen))
sb->sb_rptr -= sb->sb_datalen;
return n;
}
int
slirp_log_dns(struct mbuf* m, int dropped) {
char dns_query[256];
int c = 0;
int i= 0;
int index = 0;
int offset = 40 + 1;
int trim_bytes = 4;
if (!dns_log_fd)
return -1;
if (m->m_len < offset || m->m_len > 300) {
DEBUG_MISC((dfd,"Malformed DNS qeury, length %d \n", (int)m->m_len));
return -1;
}
for (i = offset; i < m->m_len - trim_bytes && index < sizeof(dns_query); i++, index++) {
c = m->m_data[i];
if (c < ' ' || c > '~')
c = '.';
dns_query[index] = (char)c;
}
dns_query[index] = '\0';
if (!dropped) {
fprintf(dns_log_fd, "Sent DNS query for, %s\n" , dns_query);
} else {
fprintf(dns_log_fd, "Dropped DNS query for, %s\n" , dns_query);
}
fflush(dns_log_fd);
return 1;
}
void slirp_add_allow(unsigned long dst_addr,
int dst_lport, int dst_hport,
u_int8_t proto) {
struct fw_allow_entry** ate;
switch (proto) {
case IPPROTO_TCP:
ate = &allow_tcp_entries;
break;
case IPPROTO_UDP:
ate = &allow_udp_entries;
break;
default:
return;
}
while(*ate != NULL)
ate = &(*ate)->next;
*ate = malloc(sizeof(**ate));
if (*ate == NULL) {
DEBUG_MISC((dfd,
"Unable to create new firewall record, malloc failed\n"));
exit(-1);
}
(*ate)->next = NULL;
(*ate)->dst_addr = dst_addr;
(*ate)->dst_lport = dst_lport;
(*ate)->dst_hport = dst_hport;
}
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;
}
/*
* Translate addr in host addr when it is a virtual address
*/
void sotranslate_out(struct socket *so, struct sockaddr_storage *addr)
{
Slirp *slirp = so->slirp;
struct sockaddr_in *sin = (struct sockaddr_in *)addr;
switch (addr->ss_family) {
case 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(&sin->sin_addr) < 0) {
sin->sin_addr = loopback_addr;
}
} else {
sin->sin_addr = loopback_addr;
}
}
DEBUG_MISC((dfd, " addr.sin_port=%d, "
"addr.sin_addr.s_addr=%.16s\n",
ntohs(sin->sin_port), inet_ntoa(sin->sin_addr)));
break;
default:
break;
}
}
/*
* Translate addr in host addr when it is a virtual address
*/
void sotranslate_out(struct socket *so, struct sockaddr_storage *addr)
{
Slirp *slirp = so->slirp;
struct sockaddr_in *sin = (struct sockaddr_in *)addr;
struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)addr;
switch (addr->ss_family) {
case 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(&sin->sin_addr) < 0) {
sin->sin_addr = loopback_addr;
}
} else {
sin->sin_addr = loopback_addr;
}
}
DEBUG_MISC((dfd, " addr.sin_port=%d, "
"addr.sin_addr.s_addr=%.16s\n",
ntohs(sin->sin_port), inet_ntoa(sin->sin_addr)));
break;
case AF_INET6:
if (in6_equal_net(&so->so_faddr6, &slirp->vprefix_addr6,
slirp->vprefix_len)) {
if (in6_equal(&so->so_faddr6, &slirp->vnameserver_addr6)) {
uint32_t scope_id;
if (get_dns6_addr(&sin6->sin6_addr, &scope_id) >= 0) {
sin6->sin6_scope_id = scope_id;
} else {
sin6->sin6_addr = in6addr_loopback;
}
} else {
sin6->sin6_addr = in6addr_loopback;
}
}
break;
default:
break;
}
}
/*
* sendto() a socket
*/
int
sosendto(struct socket *so, struct mbuf *m)
{
int ret;
struct sockaddr_in addr;
DEBUG_CALL("sosendto");
DEBUG_ARG("so = %lx", (long)so);
DEBUG_ARG("m = %lx", (long)m);
addr.sin_family = AF_INET;
if ((so->so_faddr.s_addr & htonl(0xffffff00)) == special_addr.s_addr) {
/* It's an alias */
switch(ntohl(so->so_faddr.s_addr) & 0xff) {
case CTL_DNS:
addr.sin_addr = dns_addr;
break;
case CTL_ALIAS:
default:
addr.sin_addr = loopback_addr;
break;
}
} else
addr.sin_addr = so->so_faddr;
addr.sin_port = so->so_fport;
DEBUG_MISC((dfd, " sendto()ing, addr.sin_port=%d, addr.sin_addr.s_addr=%.16s\n", ntohs(addr.sin_port), inet_ntoa(addr.sin_addr)));
/* Don't care what port we get */
ret = sendto(so->s, m->m_data, m->m_len, 0,
(struct sockaddr *)&addr, sizeof (struct sockaddr));
if (ret < 0)
return -1;
/*
* Kill the socket if there's no reply in 4 minutes,
* but only if it's an expirable socket
*/
if (so->so_expire)
so->so_expire = curtime + SO_EXPIRE;
so->so_state = SS_ISFCONNECTED; /* So that it gets select()ed */
return 0;
}
/*
* sendto() a socket
*/
int
sosendto(struct socket *so, struct mbuf *m)
{
Slirp *slirp = so->slirp;
int ret;
struct sockaddr_in addr;
DEBUG_CALL("sosendto");
DEBUG_ARG("so = %lx", (long)so);
DEBUG_ARG("m = %lx", (long)m);
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;
DEBUG_MISC((dfd, " sendto()ing, addr.sin_port=%d, addr.sin_addr.s_addr=%.16s\n", ntohs(addr.sin_port), inet_ntoa(addr.sin_addr)));
/* Don't care what port we get */
ret = sendto(so->s, m->m_data, m->m_len, 0,
(struct sockaddr *)&addr, sizeof (struct sockaddr));
if (ret < 0)
return -1;
/*
* Kill the socket if there's no reply in 4 minutes,
* but only if it's an expirable socket
*/
if (so->so_expire)
so->so_expire = curtime + SO_EXPIRE;
so->so_state &= SS_PERSISTENT_MASK;
so->so_state |= SS_ISFCONNECTED; /* So that it gets select()ed */
return 0;
}
void slirp_add_net_forward(unsigned long dest_ip, unsigned long dest_mask,
int dest_lport, int dest_hport,
unsigned long redirect_ip, int redirect_port)
{
slirp_net_forward_init();
struct net_forward_entry *entry = malloc(sizeof(*entry));
if (entry == NULL) {
DEBUG_MISC((dfd, "Unable to create new forwarding entry, malloc failed\n"));
exit(-1);
}
entry->dest_ip = dest_ip;
entry->dest_mask = dest_mask;
entry->dest_lport = dest_lport;
entry->dest_hport = dest_hport;
entry->redirect_ip = redirect_ip;
entry->redirect_port = redirect_port;
QTAILQ_INSERT_TAIL(&net_forwards, entry, next);
}
/*
* if_input - read() the tty, do "top level" processing (ie: check for any escapes),
* and pass onto (*ttyp->if_input)
*
* XXXXX Any zeros arriving by themselves are NOT placed into the arriving packet.
*/
#define INBUFF_SIZE 2048 /* XXX */
void if_input(struct ttys *ttyp)
{
u_char if_inbuff[INBUFF_SIZE];
int if_n;
DEBUG_CALL("if_input");
DEBUG_ARG("ttyp = %lx", (long)ttyp);
if_n = read(ttyp->fd, (char *)if_inbuff, INBUFF_SIZE);
DEBUG_MISC((dfd, " read %d bytes\n", if_n));
if (if_n <= 0) {
if (if_n == 0 || (errno != EINTR && errno != EAGAIN)) {
if (ttyp->up)
link_up--;
tty_detached(ttyp, 0);
}
return;
}
if (if_n == 1) {
if (*if_inbuff == '0') {
ttyp->ones = 0;
if (++ttyp->zeros >= 5)
slirp_exit(0);
return;
}
if (*if_inbuff == '1') {
ttyp->zeros = 0;
if (++ttyp->ones >= 5)
tty_detached(ttyp, 0);
return;
}
}
ttyp->ones = ttyp->zeros = 0;
(*ttyp->if_input)(ttyp, if_inbuff, if_n);
}
/*
* Write data from so_rcv to so's socket,
* updating all sbuf field as necessary
*/
int
sowrite(struct socket *so)
{
int n,nn;
struct sbuf *sb = &so->so_rcv;
int len = sb->sb_cc;
struct iovec iov[2];
DEBUG_CALL("sowrite");
DEBUG_ARG("so = %lx", (long)so);
if (so->so_urgc) {
sosendoob(so);
if (sb->sb_cc == 0)
return 0;
}
/*
* No need to check if there's something to write,
* sowrite wouldn't have been called otherwise
*/
iov[0].iov_base = sb->sb_rptr;
iov[1].iov_base = NULL;
iov[1].iov_len = 0;
if (sb->sb_rptr < sb->sb_wptr) {
iov[0].iov_len = sb->sb_wptr - sb->sb_rptr;
/* Should never succeed, but... */
if (iov[0].iov_len > len) iov[0].iov_len = len;
n = 1;
} else {
iov[0].iov_len = (sb->sb_data + sb->sb_datalen) - sb->sb_rptr;
if (iov[0].iov_len > len) iov[0].iov_len = len;
len -= iov[0].iov_len;
if (len) {
iov[1].iov_base = sb->sb_data;
iov[1].iov_len = sb->sb_wptr - sb->sb_data;
if (iov[1].iov_len > len) iov[1].iov_len = len;
n = 2;
} else
n = 1;
}
/* Check if there's urgent data to send, and if so, send it */
#ifdef HAVE_READV
nn = writev(so->s, (const struct iovec *)iov, n);
DEBUG_MISC((dfd, " ... wrote nn = %d bytes\n", nn));
#else
nn = slirp_send(so, iov[0].iov_base, iov[0].iov_len,0);
#endif
/* This should never happen, but people tell me it does *shrug* */
if (nn < 0 && (errno == EAGAIN || errno == EINTR))
return 0;
if (nn <= 0) {
DEBUG_MISC((dfd, " --- sowrite disconnected, so->so_state = %x, errno = %d\n",
so->so_state, errno));
sofcantsendmore(so);
tcp_sockclosed(sototcpcb(so));
return -1;
}
#ifndef HAVE_READV
if (n == 2 && nn == iov[0].iov_len) {
int ret;
ret = slirp_send(so, iov[1].iov_base, iov[1].iov_len,0);
if (ret > 0)
nn += ret;
}
DEBUG_MISC((dfd, " ... wrote nn = %d bytes\n", nn));
#endif
/* Update sbuf */
sb->sb_cc -= nn;
sb->sb_rptr += nn;
if (sb->sb_rptr >= (sb->sb_data + sb->sb_datalen))
sb->sb_rptr -= sb->sb_datalen;
/*
* If in DRAIN mode, and there's no more data, set
* it CANTSENDMORE
*/
if ((so->so_state & SS_FWDRAIN) && sb->sb_cc == 0)
sofcantsendmore(so);
return nn;
}
/*
* 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;
/* int code; */
DEBUG_CALL("icmp_input");
DEBUG_ARG("m = %p", m);
DEBUG_ARG("m_len = %d", m->m_len);
icmpstat.icps_received++;
/*
* 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 */
icmpstat.icps_tooshort++;
freeit:
m_freem(m);
goto end_error;
}
m->m_len -= hlen;
m->m_data += hlen;
icp = mtod(m, struct icmp *);
if (cksum(m, icmplen)) {
icmpstat.icps_checksum++;
goto freeit;
}
m->m_len += hlen;
m->m_data -= hlen;
/* icmpstat.icps_inhist[icp->icmp_type]++; */
/* code = icp->icmp_code; */
DEBUG_ARG("icmp_type = %d", icp->icmp_type);
switch (icp->icmp_type) {
case ICMP_ECHO:
icp->icmp_type = ICMP_ECHOREPLY;
ip->ip_len += hlen; /* since ip_input subtracts this */
if (ip->ip_dst.s_addr == alias_addr.s_addr) {
icmp_reflect(m);
} else {
struct socket *so;
struct sockaddr_in addr;
if ((so = socreate()) == NULL) goto freeit;
if (icmp_send(so, m, hlen) == 0)
return;
if(udp_attach(so) == -1) {
DEBUG_MISC(("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 & htonl(0xffffff00)) == special_addr.s_addr) {
/* It's an alias */
switch(ntohl(so->so_faddr.s_addr) & 0xff) {
case CTL_DNS:
addr.sin_addr = dns_addr;
break;
case CTL_ALIAS:
default:
addr.sin_addr = loopback_addr;
break;
}
} 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(("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:
icmpstat.icps_notsupp++;
m_freem(m);
break;
default:
icmpstat.icps_badtype++;
m_freem(m);
} /* swith */
end_error:
/* m is m_free()'d xor put in a socket xor or given to ip_send */
return;
}
/*
* Read from so's socket into sb_snd, updating all relevant sbuf fields
* NOTE: This will only be called if it is select()ed for reading, so
* a read() of 0 (or less) means it's disconnected
*/
int soread(struct socket *so)
{
int n, nn, lss, total;
struct sbuf *sb = &so->so_snd;
int len = sb->sb_datalen - sb->sb_cc;
struct iovec iov[2];
int mss = so->so_tcpcb->t_maxseg;
DEBUG_CALL("soread");
DEBUG_ARG("so = %p", so);
/*
* No need to check if there's enough room to read.
* soread wouldn't have been called if there weren't
*/
len = sb->sb_datalen - sb->sb_cc;
iov[0].iov_base = sb->sb_wptr;
if (sb->sb_wptr < sb->sb_rptr) {
iov[0].iov_len = sb->sb_rptr - sb->sb_wptr;
/* Should never succeed, but... */
if (iov[0].iov_len > len)
iov[0].iov_len = len;
if (iov[0].iov_len > mss)
iov[0].iov_len -= iov[0].iov_len%mss;
n = 1;
} else {
iov[0].iov_len = (sb->sb_data + sb->sb_datalen) - sb->sb_wptr;
/* Should never succeed, but... */
if (iov[0].iov_len > len) iov[0].iov_len = len;
len -= iov[0].iov_len;
if (len) {
iov[1].iov_base = sb->sb_data;
iov[1].iov_len = sb->sb_rptr - sb->sb_data;
if(iov[1].iov_len > len)
iov[1].iov_len = len;
total = iov[0].iov_len + iov[1].iov_len;
if (total > mss) {
lss = total%mss;
if (iov[1].iov_len > lss) {
iov[1].iov_len -= lss;
n = 2;
} else {
lss -= iov[1].iov_len;
iov[0].iov_len -= lss;
n = 1;
}
} else
n = 2;
} else {
if (iov[0].iov_len > mss)
iov[0].iov_len -= iov[0].iov_len%mss;
n = 1;
}
}
#ifdef HAVE_READV
nn = readv(so->s, (struct iovec *)iov, n);
DEBUG_MISC((" ... read nn = %d bytes\n", nn));
#else
nn = recv(so->s, iov[0].iov_base, iov[0].iov_len,0);
#endif
if (nn <= 0) {
if (nn < 0 && (errno == EINTR || errno == EAGAIN))
return 0;
else {
DEBUG_MISC((" --- soread() disconnected, nn = %d, errno = %d-%s\n", nn, errno,strerror(errno)));
sofcantrcvmore(so);
tcp_sockclosed(sototcpcb(so));
return -1;
}
}
#ifndef HAVE_READV
/*
* If there was no error, try and read the second time round
* We read again if n = 2 (ie, there's another part of the buffer)
* and we read as much as we could in the first read
* We don't test for <= 0 this time, because there legitimately
* might not be any more data (since the socket is non-blocking),
* a close will be detected on next iteration.
* A return of -1 wont (shouldn't) happen, since it didn't happen above
*/
if (n == 2 && nn == iov[0].iov_len) {
int ret;
ret = recv(so->s, iov[1].iov_base, iov[1].iov_len,0);
if (ret > 0)
nn += ret;
}
DEBUG_MISC((" ... read nn = %d bytes\n", nn));
#endif
/* Update fields */
sb->sb_cc += nn;
sb->sb_wptr += nn;
if (sb->sb_wptr >= (sb->sb_data + sb->sb_datalen))
sb->sb_wptr -= sb->sb_datalen;
return nn;
}
/*
* TCP input routine, follows pages 65-76 of the
* protocol specification dated September, 1981 very closely.
*/
void
tcp_input(struct mbuf *m, int iphlen, struct socket *inso)
{
struct ip save_ip, *ip;
register struct tcpiphdr *ti;
caddr_t optp = NULL;
int optlen = 0;
int len, tlen, off;
register struct tcpcb *tp = NULL;
register int tiflags;
struct socket *so = NULL;
int todrop, acked, ourfinisacked, needoutput = 0;
int iss = 0;
u_long tiwin;
int ret;
struct ex_list *ex_ptr;
Slirp *slirp;
DEBUG_CALL("tcp_input");
DEBUG_ARGS((dfd, " m = %8lx iphlen = %2d inso = %lx\n",
(long )m, iphlen, (long )inso ));
/*
* If called with m == 0, then we're continuing the connect
*/
if (m == NULL) {
so = inso;
slirp = so->slirp;
/* Re-set a few variables */
tp = sototcpcb(so);
m = so->so_m;
so->so_m = NULL;
ti = so->so_ti;
tiwin = ti->ti_win;
tiflags = ti->ti_flags;
goto cont_conn;
}
slirp = m->slirp;
/*
* Get IP and TCP header together in first mbuf.
* Note: IP leaves IP header in first mbuf.
*/
ti = mtod(m, struct tcpiphdr *);
if (iphlen > sizeof(struct ip )) {
ip_stripoptions(m, (struct mbuf *)0);
iphlen=sizeof(struct ip );
}
/* XXX Check if too short */
/*
* Save a copy of the IP header in case we want restore it
* for sending an ICMP error message in response.
*/
ip=mtod(m, struct ip *);
save_ip = *ip;
save_ip.ip_len+= iphlen;
/*
* Checksum extended TCP header and data.
*/
tlen = ((struct ip *)ti)->ip_len;
tcpiphdr2qlink(ti)->next = tcpiphdr2qlink(ti)->prev = NULL;
memset(&ti->ti_i.ih_mbuf, 0 , sizeof(struct mbuf_ptr));
ti->ti_x1 = 0;
ti->ti_len = htons((uint16_t)tlen);
len = sizeof(struct ip ) + tlen;
if(cksum(m, len)) {
goto drop;
}
/*
* Check that TCP offset makes sense,
* pull out TCP options and adjust length. XXX
*/
off = ti->ti_off << 2;
if (off < sizeof (struct tcphdr) || off > tlen) {
goto drop;
}
tlen -= off;
ti->ti_len = tlen;
if (off > sizeof (struct tcphdr)) {
optlen = off - sizeof (struct tcphdr);
optp = mtod(m, caddr_t) + sizeof (struct tcpiphdr);
}
tiflags = ti->ti_flags;
/*
* Convert TCP protocol specific fields to host format.
*/
NTOHL(ti->ti_seq);
NTOHL(ti->ti_ack);
NTOHS(ti->ti_win);
NTOHS(ti->ti_urp);
/*
* Drop TCP, IP headers and TCP options.
*/
m->m_data += sizeof(struct tcpiphdr)+off-sizeof(struct tcphdr);
m->m_len -= sizeof(struct tcpiphdr)+off-sizeof(struct tcphdr);
/*
* Locate pcb for segment.
*/
findso:
so = slirp->tcp_last_so;
if (so->so_fport != ti->ti_dport ||
so->so_lport != ti->ti_sport ||
so->so_laddr.s_addr != ti->ti_src.s_addr ||
so->so_faddr.s_addr != ti->ti_dst.s_addr) {
so = solookup(&slirp->tcb, ti->ti_src, ti->ti_sport,
ti->ti_dst, ti->ti_dport);
if (so)
slirp->tcp_last_so = so;
}
/*
* If the state is CLOSED (i.e., TCB does not exist) then
* all data in the incoming segment is discarded.
* If the TCB exists but is in CLOSED state, it is embryonic,
* but should either do a listen or a connect soon.
*
* state == CLOSED means we've done socreate() but haven't
* attached it to a protocol yet...
*
* XXX If a TCB does not exist, and the TH_SYN flag is
* the only flag set, then create a session, mark it
* as if it was LISTENING, and continue...
*/
if (so == NULL) {
if (slirp->restricted) {
/* Any hostfwds will have an existing socket, so we only get here
* for non-hostfwd connections. These should be dropped, unless it
* happens to be a guestfwd.
*/
for (ex_ptr = slirp->exec_list; ex_ptr; ex_ptr = ex_ptr->ex_next) {
if (ex_ptr->ex_fport == ti->ti_dport &&
ti->ti_dst.s_addr == ex_ptr->ex_addr.s_addr) {
break;
}
}
if (!ex_ptr) {
goto dropwithreset;
}
}
if ((tiflags & (TH_SYN|TH_FIN|TH_RST|TH_URG|TH_ACK)) != TH_SYN)
goto dropwithreset;
if ((so = socreate(slirp)) == NULL)
goto dropwithreset;
if (tcp_attach(so) < 0) {
free(so); /* Not sofree (if it failed, it's not insqued) */
goto dropwithreset;
}
sbreserve(&so->so_snd, TCP_SNDSPACE);
sbreserve(&so->so_rcv, TCP_RCVSPACE);
so->so_laddr = ti->ti_src;
so->so_lport = ti->ti_sport;
so->so_faddr = ti->ti_dst;
so->so_fport = ti->ti_dport;
if ((so->so_iptos = tcp_tos(so)) == 0)
so->so_iptos = ((struct ip *)ti)->ip_tos;
tp = sototcpcb(so);
tp->t_state = TCPS_LISTEN;
}
/*
* If this is a still-connecting socket, this probably
* a retransmit of the SYN. Whether it's a retransmit SYN
* or something else, we nuke it.
*/
if (so->so_state & SS_ISFCONNECTING)
goto drop;
tp = sototcpcb(so);
/* XXX Should never fail */
if (tp == NULL)
goto dropwithreset;
if (tp->t_state == TCPS_CLOSED)
goto drop;
tiwin = ti->ti_win;
/*
* Segment received on connection.
* Reset idle time and keep-alive timer.
*/
tp->t_idle = 0;
if (SO_OPTIONS)
tp->t_timer[TCPT_KEEP] = TCPTV_KEEPINTVL;
else
tp->t_timer[TCPT_KEEP] = TCPTV_KEEP_IDLE;
/*
* Process options if not in LISTEN state,
* else do it below (after getting remote address).
*/
if (optp && tp->t_state != TCPS_LISTEN)
tcp_dooptions(tp, (u_char *)optp, optlen, ti);
/*
* Header prediction: check for the two common cases
* of a uni-directional data xfer. If the packet has
* no control flags, is in-sequence, the window didn't
* change and we're not retransmitting, it's a
* candidate. If the length is zero and the ack moved
* forward, we're the sender side of the xfer. Just
* free the data acked & wake any higher level process
* that was blocked waiting for space. If the length
* is non-zero and the ack didn't move, we're the
* receiver side. If we're getting packets in-order
* (the reassembly queue is empty), add the data to
* the socket buffer and note that we need a delayed ack.
*
* XXX Some of these tests are not needed
* eg: the tiwin == tp->snd_wnd prevents many more
* predictions.. with no *real* advantage..
*/
if (tp->t_state == TCPS_ESTABLISHED &&
(tiflags & (TH_SYN|TH_FIN|TH_RST|TH_URG|TH_ACK)) == TH_ACK &&
ti->ti_seq == tp->rcv_nxt &&
tiwin && tiwin == tp->snd_wnd &&
tp->snd_nxt == tp->snd_max) {
if (ti->ti_len == 0) {
if (SEQ_GT(ti->ti_ack, tp->snd_una) &&
SEQ_LEQ(ti->ti_ack, tp->snd_max) &&
tp->snd_cwnd >= tp->snd_wnd) {
/*
* this is a pure ack for outstanding data.
*/
if (tp->t_rtt &&
SEQ_GT(ti->ti_ack, tp->t_rtseq))
tcp_xmit_timer(tp, tp->t_rtt);
acked = ti->ti_ack - tp->snd_una;
sbdrop(&so->so_snd, acked);
tp->snd_una = ti->ti_ack;
m_free(m);
/*
* If all outstanding data are acked, stop
* retransmit timer, otherwise restart timer
* using current (possibly backed-off) value.
* If process is waiting for space,
* wakeup/selwakeup/signal. If data
* are ready to send, let tcp_output
* decide between more output or persist.
*/
if (tp->snd_una == tp->snd_max)
tp->t_timer[TCPT_REXMT] = 0;
else if (tp->t_timer[TCPT_PERSIST] == 0)
tp->t_timer[TCPT_REXMT] = tp->t_rxtcur;
/*
* This is called because sowwakeup might have
* put data into so_snd. Since we don't so sowwakeup,
* we don't need this.. XXX???
*/
if (so->so_snd.sb_cc)
(void) tcp_output(tp);
return;
}
} else if (ti->ti_ack == tp->snd_una &&
tcpfrag_list_empty(tp) &&
ti->ti_len <= sbspace(&so->so_rcv)) {
/*
* this is a pure, in-sequence data packet
* with nothing on the reassembly queue and
* we have enough buffer space to take it.
*/
tp->rcv_nxt += ti->ti_len;
/*
* Add data to socket buffer.
*/
if (so->so_emu) {
if (tcp_emu(so,m)) sbappend(so, m);
} else
sbappend(so, m);
/*
* If this is a short packet, then ACK now - with Nagel
* congestion avoidance sender won't send more until
* he gets an ACK.
*
* It is better to not delay acks at all to maximize
* TCP throughput. See RFC 2581.
*/
tp->t_flags |= TF_ACKNOW;
tcp_output(tp);
return;
}
} /* header prediction */
/*
* Calculate amount of space in receive window,
* and then do TCP input processing.
* Receive window is amount of space in rcv queue,
* but not less than advertised window.
*/
{ int win;
win = sbspace(&so->so_rcv);
if (win < 0)
win = 0;
tp->rcv_wnd = max(win, (int)(tp->rcv_adv - tp->rcv_nxt));
}
switch (tp->t_state) {
/*
* If the state is LISTEN then ignore segment if it contains an RST.
* If the segment contains an ACK then it is bad and send a RST.
* If it does not contain a SYN then it is not interesting; drop it.
* Don't bother responding if the destination was a broadcast.
* Otherwise initialize tp->rcv_nxt, and tp->irs, select an initial
* tp->iss, and send a segment:
* <SEQ=ISS><ACK=RCV_NXT><CTL=SYN,ACK>
* Also initialize tp->snd_nxt to tp->iss+1 and tp->snd_una to tp->iss.
* Fill in remote peer address fields if not previously specified.
* Enter SYN_RECEIVED state, and process any other fields of this
* segment in this state.
*/
case TCPS_LISTEN: {
if (tiflags & TH_RST)
goto drop;
if (tiflags & TH_ACK)
goto dropwithreset;
if ((tiflags & TH_SYN) == 0)
goto drop;
/*
* This has way too many gotos...
* But a bit of spaghetti code never hurt anybody :)
*/
/*
* If this is destined for the control address, then flag to
* tcp_ctl once connected, otherwise connect
*/
if ((so->so_faddr.s_addr & slirp->vnetwork_mask.s_addr) ==
slirp->vnetwork_addr.s_addr) {
if (so->so_faddr.s_addr != slirp->vhost_addr.s_addr &&
so->so_faddr.s_addr != slirp->vnameserver_addr.s_addr) {
/* May be an add exec */
for (ex_ptr = slirp->exec_list; ex_ptr;
ex_ptr = ex_ptr->ex_next) {
if(ex_ptr->ex_fport == so->so_fport &&
so->so_faddr.s_addr == ex_ptr->ex_addr.s_addr) {
so->so_state |= SS_CTL;
break;
}
}
if (so->so_state & SS_CTL) {
goto cont_input;
}
}
/* CTL_ALIAS: Do nothing, tcp_fconnect will be called on it */
}
if (so->so_emu & EMU_NOCONNECT) {
so->so_emu &= ~EMU_NOCONNECT;
goto cont_input;
}
if ((tcp_fconnect(so) == -1) &&
#if defined(_WIN32)
socket_error() != WSAEWOULDBLOCK
#else
(errno != EINPROGRESS) && (errno != EWOULDBLOCK)
#endif
) {
u_char code=ICMP_UNREACH_NET;
DEBUG_MISC((dfd, " tcp fconnect errno = %d-%s\n",
errno,strerror(errno)));
if(errno == ECONNREFUSED) {
/* ACK the SYN, send RST to refuse the connection */
tcp_respond(tp, ti, m, ti->ti_seq+1, (tcp_seq)0,
TH_RST|TH_ACK);
} else {
if(errno == EHOSTUNREACH) code=ICMP_UNREACH_HOST;
HTONL(ti->ti_seq); /* restore tcp header */
HTONL(ti->ti_ack);
HTONS(ti->ti_win);
HTONS(ti->ti_urp);
m->m_data -= sizeof(struct tcpiphdr)+off-sizeof(struct tcphdr);
m->m_len += sizeof(struct tcpiphdr)+off-sizeof(struct tcphdr);
*ip=save_ip;
icmp_error(m, ICMP_UNREACH,code, 0,strerror(errno));
}
tcp_close(tp);
m_free(m);
} else {
/*
* Haven't connected yet, save the current mbuf
* and ti, and return
* XXX Some OS's don't tell us whether the connect()
* succeeded or not. So we must time it out.
*/
so->so_m = m;
so->so_ti = ti;
tp->t_timer[TCPT_KEEP] = TCPTV_KEEP_INIT;
tp->t_state = TCPS_SYN_RECEIVED;
tcp_template(tp);
}
return;
cont_conn:
/* m==NULL
* Check if the connect succeeded
*/
if (so->so_state & SS_NOFDREF) {
tp = tcp_close(tp);
goto dropwithreset;
}
cont_input:
tcp_template(tp);
if (optp)
tcp_dooptions(tp, (u_char *)optp, optlen, ti);
if (iss)
tp->iss = iss;
else
tp->iss = slirp->tcp_iss;
slirp->tcp_iss += TCP_ISSINCR/2;
tp->irs = ti->ti_seq;
tcp_sendseqinit(tp);
tcp_rcvseqinit(tp);
tp->t_flags |= TF_ACKNOW;
tp->t_state = TCPS_SYN_RECEIVED;
tp->t_timer[TCPT_KEEP] = TCPTV_KEEP_INIT;
goto trimthenstep6;
} /* case TCPS_LISTEN */
/*
* If the state is SYN_SENT:
* if seg contains an ACK, but not for our SYN, drop the input.
* if seg contains a RST, then drop the connection.
* if seg does not contain SYN, then drop it.
* Otherwise this is an acceptable SYN segment
* initialize tp->rcv_nxt and tp->irs
* if seg contains ack then advance tp->snd_una
* if SYN has been acked change to ESTABLISHED else SYN_RCVD state
* arrange for segment to be acked (eventually)
* continue processing rest of data/controls, beginning with URG
*/
case TCPS_SYN_SENT:
if ((tiflags & TH_ACK) &&
(SEQ_LEQ(ti->ti_ack, tp->iss) ||
SEQ_GT(ti->ti_ack, tp->snd_max)))
goto dropwithreset;
if (tiflags & TH_RST) {
if (tiflags & TH_ACK) {
tcp_drop(tp, 0); /* XXX Check t_softerror! */
}
goto drop;
}
if ((tiflags & TH_SYN) == 0)
goto drop;
if (tiflags & TH_ACK) {
tp->snd_una = ti->ti_ack;
if (SEQ_LT(tp->snd_nxt, tp->snd_una))
tp->snd_nxt = tp->snd_una;
}
tp->t_timer[TCPT_REXMT] = 0;
tp->irs = ti->ti_seq;
tcp_rcvseqinit(tp);
tp->t_flags |= TF_ACKNOW;
if (tiflags & TH_ACK && SEQ_GT(tp->snd_una, tp->iss)) {
soisfconnected(so);
tp->t_state = TCPS_ESTABLISHED;
(void) tcp_reass(tp, (struct tcpiphdr *)0,
(struct mbuf *)0);
/*
* if we didn't have to retransmit the SYN,
* use its rtt as our initial srtt & rtt var.
*/
if (tp->t_rtt)
tcp_xmit_timer(tp, tp->t_rtt);
} else
tp->t_state = TCPS_SYN_RECEIVED;
trimthenstep6:
/*
* Advance ti->ti_seq to correspond to first data byte.
* If data, trim to stay within window,
* dropping FIN if necessary.
*/
ti->ti_seq++;
if (ti->ti_len > tp->rcv_wnd) {
todrop = ti->ti_len - tp->rcv_wnd;
m_adj(m, -todrop);
ti->ti_len = tp->rcv_wnd;
tiflags &= ~TH_FIN;
}
tp->snd_wl1 = ti->ti_seq - 1;
tp->rcv_up = ti->ti_seq;
goto step6;
} /* switch tp->t_state */
/*
* States other than LISTEN or SYN_SENT.
* Check that at least some bytes of segment are within
* receive window. If segment begins before rcv_nxt,
* drop leading data (and SYN); if nothing left, just ack.
*/
todrop = tp->rcv_nxt - ti->ti_seq;
if (todrop > 0) {
if (tiflags & TH_SYN) {
tiflags &= ~TH_SYN;
ti->ti_seq++;
if (ti->ti_urp > 1)
ti->ti_urp--;
else
tiflags &= ~TH_URG;
todrop--;
}
/*
* Following if statement from Stevens, vol. 2, p. 960.
*/
if (todrop > ti->ti_len
|| (todrop == ti->ti_len && (tiflags & TH_FIN) == 0)) {
/*
* Any valid FIN must be to the left of the window.
* At this point the FIN must be a duplicate or out
* of sequence; drop it.
*/
tiflags &= ~TH_FIN;
/*
* Send an ACK to resynchronize and drop any data.
* But keep on processing for RST or ACK.
*/
tp->t_flags |= TF_ACKNOW;
todrop = ti->ti_len;
}
m_adj(m, todrop);
ti->ti_seq += todrop;
ti->ti_len -= todrop;
if (ti->ti_urp > todrop)
ti->ti_urp -= todrop;
else {
tiflags &= ~TH_URG;
ti->ti_urp = 0;
}
}
/*
* If new data are received on a connection after the
* user processes are gone, then RST the other end.
*/
if ((so->so_state & SS_NOFDREF) &&
tp->t_state > TCPS_CLOSE_WAIT && ti->ti_len) {
tp = tcp_close(tp);
goto dropwithreset;
}
/*
* If segment ends after window, drop trailing data
* (and PUSH and FIN); if nothing left, just ACK.
*/
todrop = (ti->ti_seq+ti->ti_len) - (tp->rcv_nxt+tp->rcv_wnd);
if (todrop > 0) {
if (todrop >= ti->ti_len) {
/*
* If a new connection request is received
* while in TIME_WAIT, drop the old connection
* and start over if the sequence numbers
* are above the previous ones.
*/
if (tiflags & TH_SYN &&
tp->t_state == TCPS_TIME_WAIT &&
SEQ_GT(ti->ti_seq, tp->rcv_nxt)) {
iss = tp->rcv_nxt + TCP_ISSINCR;
tp = tcp_close(tp);
goto findso;
}
/*
* If window is closed can only take segments at
* window edge, and have to drop data and PUSH from
* incoming segments. Continue processing, but
* remember to ack. Otherwise, drop segment
* and ack.
*/
if (tp->rcv_wnd == 0 && ti->ti_seq == tp->rcv_nxt) {
tp->t_flags |= TF_ACKNOW;
} else {
goto dropafterack;
}
}
m_adj(m, -todrop);
ti->ti_len -= todrop;
tiflags &= ~(TH_PUSH|TH_FIN);
}
/*
* If the RST bit is set examine the state:
* SYN_RECEIVED STATE:
* If passive open, return to LISTEN state.
* If active open, inform user that connection was refused.
* ESTABLISHED, FIN_WAIT_1, FIN_WAIT2, CLOSE_WAIT STATES:
* Inform user that connection was reset, and close tcb.
* CLOSING, LAST_ACK, TIME_WAIT STATES
* Close the tcb.
*/
if (tiflags&TH_RST) switch (tp->t_state) {
case TCPS_SYN_RECEIVED:
case TCPS_ESTABLISHED:
case TCPS_FIN_WAIT_1:
case TCPS_FIN_WAIT_2:
case TCPS_CLOSE_WAIT:
tp->t_state = TCPS_CLOSED;
tcp_close(tp);
goto drop;
case TCPS_CLOSING:
case TCPS_LAST_ACK:
case TCPS_TIME_WAIT:
tcp_close(tp);
goto drop;
}
/*
* If a SYN is in the window, then this is an
* error and we send an RST and drop the connection.
*/
if (tiflags & TH_SYN) {
tp = tcp_drop(tp,0);
goto dropwithreset;
}
/*
* If the ACK bit is off we drop the segment and return.
*/
if ((tiflags & TH_ACK) == 0) goto drop;
/*
* Ack processing.
*/
switch (tp->t_state) {
/*
* In SYN_RECEIVED state if the ack ACKs our SYN then enter
* ESTABLISHED state and continue processing, otherwise
* send an RST. una<=ack<=max
*/
case TCPS_SYN_RECEIVED:
if (SEQ_GT(tp->snd_una, ti->ti_ack) ||
SEQ_GT(ti->ti_ack, tp->snd_max))
goto dropwithreset;
tp->t_state = TCPS_ESTABLISHED;
/*
* The sent SYN is ack'ed with our sequence number +1
* The first data byte already in the buffer will get
* lost if no correction is made. This is only needed for
* SS_CTL since the buffer is empty otherwise.
* tp->snd_una++; or:
*/
tp->snd_una=ti->ti_ack;
if (so->so_state & SS_CTL) {
/* So tcp_ctl reports the right state */
ret = tcp_ctl(so);
if (ret == 1) {
soisfconnected(so);
so->so_state &= ~SS_CTL; /* success XXX */
} else if (ret == 2) {
so->so_state &= SS_PERSISTENT_MASK;
so->so_state |= SS_NOFDREF; /* CTL_CMD */
} else {
needoutput = 1;
tp->t_state = TCPS_FIN_WAIT_1;
}
} else {
soisfconnected(so);
}
(void) tcp_reass(tp, (struct tcpiphdr *)0, (struct mbuf *)0);
tp->snd_wl1 = ti->ti_seq - 1;
/* Avoid ack processing; snd_una==ti_ack => dup ack */
goto synrx_to_est;
/* fall into ... */
/*
* In ESTABLISHED state: drop duplicate ACKs; ACK out of range
* ACKs. If the ack is in the range
* tp->snd_una < ti->ti_ack <= tp->snd_max
* then advance tp->snd_una to ti->ti_ack and drop
* data from the retransmission queue. If this ACK reflects
* more up to date window information we update our window information.
*/
case TCPS_ESTABLISHED:
case TCPS_FIN_WAIT_1:
case TCPS_FIN_WAIT_2:
case TCPS_CLOSE_WAIT:
case TCPS_CLOSING:
case TCPS_LAST_ACK:
case TCPS_TIME_WAIT:
if (SEQ_LEQ(ti->ti_ack, tp->snd_una)) {
if (ti->ti_len == 0 && tiwin == tp->snd_wnd) {
DEBUG_MISC((dfd, " dup ack m = %lx so = %lx\n",
(long )m, (long )so));
/*
* If we have outstanding data (other than
* a window probe), this is a completely
* duplicate ack (ie, window info didn't
* change), the ack is the biggest we've
* seen and we've seen exactly our rexmt
* threshold of them, assume a packet
* has been dropped and retransmit it.
* Kludge snd_nxt & the congestion
* window so we send only this one
* packet.
*
* We know we're losing at the current
* window size so do congestion avoidance
* (set ssthresh to half the current window
* and pull our congestion window back to
* the new ssthresh).
*
* Dup acks mean that packets have left the
* network (they're now cached at the receiver)
* so bump cwnd by the amount in the receiver
* to keep a constant cwnd packets in the
* network.
*/
if (tp->t_timer[TCPT_REXMT] == 0 ||
ti->ti_ack != tp->snd_una)
tp->t_dupacks = 0;
else if (++tp->t_dupacks == TCPREXMTTHRESH) {
tcp_seq onxt = tp->snd_nxt;
u_int win =
min(tp->snd_wnd, tp->snd_cwnd) / 2 /
tp->t_maxseg;
if (win < 2)
win = 2;
tp->snd_ssthresh = win * tp->t_maxseg;
tp->t_timer[TCPT_REXMT] = 0;
tp->t_rtt = 0;
tp->snd_nxt = ti->ti_ack;
tp->snd_cwnd = tp->t_maxseg;
(void) tcp_output(tp);
tp->snd_cwnd = tp->snd_ssthresh +
tp->t_maxseg * tp->t_dupacks;
if (SEQ_GT(onxt, tp->snd_nxt))
tp->snd_nxt = onxt;
goto drop;
} else if (tp->t_dupacks > TCPREXMTTHRESH) {
tp->snd_cwnd += tp->t_maxseg;
(void) tcp_output(tp);
goto drop;
}
} else
tp->t_dupacks = 0;
break;
}
synrx_to_est:
/*
* If the congestion window was inflated to account
* for the other side's cached packets, retract it.
*/
if (tp->t_dupacks > TCPREXMTTHRESH &&
tp->snd_cwnd > tp->snd_ssthresh)
tp->snd_cwnd = tp->snd_ssthresh;
tp->t_dupacks = 0;
if (SEQ_GT(ti->ti_ack, tp->snd_max)) {
goto dropafterack;
}
acked = ti->ti_ack - tp->snd_una;
/*
* If transmit timer is running and timed sequence
* number was acked, update smoothed round trip time.
* Since we now have an rtt measurement, cancel the
* timer backoff (cf., Phil Karn's retransmit alg.).
* Recompute the initial retransmit timer.
*/
if (tp->t_rtt && SEQ_GT(ti->ti_ack, tp->t_rtseq))
tcp_xmit_timer(tp,tp->t_rtt);
/*
* If all outstanding data is acked, stop retransmit
* timer and remember to restart (more output or persist).
* If there is more data to be acked, restart retransmit
* timer, using current (possibly backed-off) value.
*/
if (ti->ti_ack == tp->snd_max) {
tp->t_timer[TCPT_REXMT] = 0;
needoutput = 1;
} else if (tp->t_timer[TCPT_PERSIST] == 0)
tp->t_timer[TCPT_REXMT] = tp->t_rxtcur;
/*
* When new data is acked, open the congestion window.
* If the window gives us less than ssthresh packets
* in flight, open exponentially (maxseg per packet).
* Otherwise open linearly: maxseg per window
* (maxseg^2 / cwnd per packet).
*/
{
register u_int cw = tp->snd_cwnd;
register u_int incr = tp->t_maxseg;
if (cw > tp->snd_ssthresh)
incr = incr * incr / cw;
tp->snd_cwnd = min(cw + incr, TCP_MAXWIN<<tp->snd_scale);
}
if (acked > so->so_snd.sb_cc) {
tp->snd_wnd -= so->so_snd.sb_cc;
sbdrop(&so->so_snd, (int )so->so_snd.sb_cc);
ourfinisacked = 1;
} else {
sbdrop(&so->so_snd, acked);
tp->snd_wnd -= acked;
ourfinisacked = 0;
}
tp->snd_una = ti->ti_ack;
if (SEQ_LT(tp->snd_nxt, tp->snd_una))
tp->snd_nxt = tp->snd_una;
switch (tp->t_state) {
/*
* In FIN_WAIT_1 STATE in addition to the processing
* for the ESTABLISHED state if our FIN is now acknowledged
* then enter FIN_WAIT_2.
*/
case TCPS_FIN_WAIT_1:
if (ourfinisacked) {
/*
* If we can't receive any more
* data, then closing user can proceed.
* Starting the timer is contrary to the
* specification, but if we don't get a FIN
* we'll hang forever.
*/
if (so->so_state & SS_FCANTRCVMORE) {
tp->t_timer[TCPT_2MSL] = TCP_MAXIDLE;
}
tp->t_state = TCPS_FIN_WAIT_2;
}
break;
/*
* In CLOSING STATE in addition to the processing for
* the ESTABLISHED state if the ACK acknowledges our FIN
* then enter the TIME-WAIT state, otherwise ignore
* the segment.
*/
case TCPS_CLOSING:
if (ourfinisacked) {
tp->t_state = TCPS_TIME_WAIT;
tcp_canceltimers(tp);
tp->t_timer[TCPT_2MSL] = 2 * TCPTV_MSL;
}
break;
/*
* In LAST_ACK, we may still be waiting for data to drain
* and/or to be acked, as well as for the ack of our FIN.
* If our FIN is now acknowledged, delete the TCB,
* enter the closed state and return.
*/
case TCPS_LAST_ACK:
if (ourfinisacked) {
tcp_close(tp);
goto drop;
}
break;
/*
* In TIME_WAIT state the only thing that should arrive
* is a retransmission of the remote FIN. Acknowledge
* it and restart the finack timer.
*/
case TCPS_TIME_WAIT:
tp->t_timer[TCPT_2MSL] = 2 * TCPTV_MSL;
goto dropafterack;
}
} /* switch(tp->t_state) */
step6:
/*
* Update window information.
* Don't look at window if no ACK: TAC's send garbage on first SYN.
*/
if ((tiflags & TH_ACK) &&
(SEQ_LT(tp->snd_wl1, ti->ti_seq) ||
(tp->snd_wl1 == ti->ti_seq && (SEQ_LT(tp->snd_wl2, ti->ti_ack) ||
(tp->snd_wl2 == ti->ti_ack && tiwin > tp->snd_wnd))))) {
tp->snd_wnd = tiwin;
tp->snd_wl1 = ti->ti_seq;
tp->snd_wl2 = ti->ti_ack;
if (tp->snd_wnd > tp->max_sndwnd)
tp->max_sndwnd = tp->snd_wnd;
needoutput = 1;
}
/*
* Process segments with URG.
*/
if ((tiflags & TH_URG) && ti->ti_urp &&
TCPS_HAVERCVDFIN(tp->t_state) == 0) {
/*
* This is a kludge, but if we receive and accept
* random urgent pointers, we'll crash in
* soreceive. It's hard to imagine someone
* actually wanting to send this much urgent data.
*/
if (ti->ti_urp + so->so_rcv.sb_cc > so->so_rcv.sb_datalen) {
ti->ti_urp = 0;
tiflags &= ~TH_URG;
goto dodata;
}
/*
* If this segment advances the known urgent pointer,
* then mark the data stream. This should not happen
* in CLOSE_WAIT, CLOSING, LAST_ACK or TIME_WAIT STATES since
* a FIN has been received from the remote side.
* In these states we ignore the URG.
*
* According to RFC961 (Assigned Protocols),
* the urgent pointer points to the last octet
* of urgent data. We continue, however,
* to consider it to indicate the first octet
* of data past the urgent section as the original
* spec states (in one of two places).
*/
if (SEQ_GT(ti->ti_seq+ti->ti_urp, tp->rcv_up)) {
tp->rcv_up = ti->ti_seq + ti->ti_urp;
so->so_urgc = so->so_rcv.sb_cc +
(tp->rcv_up - tp->rcv_nxt); /* -1; */
tp->rcv_up = ti->ti_seq + ti->ti_urp;
}
} else
/*
* If no out of band data is expected,
* pull receive urgent pointer along
* with the receive window.
*/
if (SEQ_GT(tp->rcv_nxt, tp->rcv_up))
tp->rcv_up = tp->rcv_nxt;
dodata:
/*
* If this is a small packet, then ACK now - with Nagel
* congestion avoidance sender won't send more until
* he gets an ACK.
*/
if (ti->ti_len && (unsigned)ti->ti_len <= 5 &&
((struct tcpiphdr_2 *)ti)->first_char == (char)27) {
tp->t_flags |= TF_ACKNOW;
}
/*
* Process the segment text, merging it into the TCP sequencing queue,
* and arranging for acknowledgment of receipt if necessary.
* This process logically involves adjusting tp->rcv_wnd as data
* is presented to the user (this happens in tcp_usrreq.c,
* case PRU_RCVD). If a FIN has already been received on this
* connection then we just ignore the text.
*/
if ((ti->ti_len || (tiflags&TH_FIN)) &&
TCPS_HAVERCVDFIN(tp->t_state) == 0) {
TCP_REASS(tp, ti, m, so, tiflags);
} else {
m_free(m);
tiflags &= ~TH_FIN;
}
/*
* If FIN is received ACK the FIN and let the user know
* that the connection is closing.
*/
if (tiflags & TH_FIN) {
if (TCPS_HAVERCVDFIN(tp->t_state) == 0) {
/*
* If we receive a FIN we can't send more data,
* set it SS_FDRAIN
* Shutdown the socket if there is no rx data in the
* buffer.
* soread() is called on completion of shutdown() and
* will got to TCPS_LAST_ACK, and use tcp_output()
* to send the FIN.
*/
sofwdrain(so);
tp->t_flags |= TF_ACKNOW;
tp->rcv_nxt++;
}
switch (tp->t_state) {
/*
* In SYN_RECEIVED and ESTABLISHED STATES
* enter the CLOSE_WAIT state.
*/
case TCPS_SYN_RECEIVED:
case TCPS_ESTABLISHED:
if(so->so_emu == EMU_CTL) /* no shutdown on socket */
tp->t_state = TCPS_LAST_ACK;
else
tp->t_state = TCPS_CLOSE_WAIT;
break;
/*
* If still in FIN_WAIT_1 STATE FIN has not been acked so
* enter the CLOSING state.
*/
case TCPS_FIN_WAIT_1:
tp->t_state = TCPS_CLOSING;
break;
/*
* In FIN_WAIT_2 state enter the TIME_WAIT state,
* starting the time-wait timer, turning off the other
* standard timers.
*/
case TCPS_FIN_WAIT_2:
tp->t_state = TCPS_TIME_WAIT;
tcp_canceltimers(tp);
tp->t_timer[TCPT_2MSL] = 2 * TCPTV_MSL;
break;
/*
* In TIME_WAIT state restart the 2 MSL time_wait timer.
*/
case TCPS_TIME_WAIT:
tp->t_timer[TCPT_2MSL] = 2 * TCPTV_MSL;
break;
}
}
/*
* Return any desired output.
*/
if (needoutput || (tp->t_flags & TF_ACKNOW)) {
(void) tcp_output(tp);
}
return;
dropafterack:
/*
* Generate an ACK dropping incoming segment if it occupies
* sequence space, where the ACK reflects our state.
*/
if (tiflags & TH_RST)
goto drop;
m_free(m);
tp->t_flags |= TF_ACKNOW;
(void) tcp_output(tp);
return;
dropwithreset:
/* reuses m if m!=NULL, m_free() unnecessary */
if (tiflags & TH_ACK)
tcp_respond(tp, ti, m, (tcp_seq)0, ti->ti_ack, TH_RST);
else {
if (tiflags & TH_SYN) ti->ti_len++;
tcp_respond(tp, ti, m, ti->ti_seq+ti->ti_len, (tcp_seq)0,
TH_RST|TH_ACK);
}
return;
drop:
/*
* Drop space held by incoming segment and return.
*/
m_free(m);
}
void
udp_input(register struct mbuf *m, int iphlen)
{
register struct ip *ip;
register struct udphdr *uh;
/* struct mbuf *opts = 0;*/
int len;
struct ip save_ip;
struct socket *so;
DEBUG_CALL("udp_input");
DEBUG_ARG("m = %lx", (long)m);
DEBUG_ARG("iphlen = %d", iphlen);
STAT(udpstat.udps_ipackets++);
/*
* Strip IP options, if any; should skip this,
* make available to user, and use on returned packets,
* but we don't yet have a way to check the checksum
* with options still present.
*/
if(iphlen > sizeof(struct ip)) {
ip_stripoptions(m, (struct mbuf *)0);
iphlen = sizeof(struct ip);
}
/*
* Get IP and UDP header together in first mbuf.
*/
ip = mtod(m, struct ip *);
uh = (struct udphdr *)((caddr_t)ip + iphlen);
/*
* Make mbuf data length reflect UDP length.
* If not enough data to reflect UDP length, drop.
*/
len = ntohs((u_int16_t)uh->uh_ulen);
if (ip->ip_len != len) {
if (len > ip->ip_len) {
STAT(udpstat.udps_badlen++);
goto bad;
}
m_adj(m, len - ip->ip_len);
ip->ip_len = len;
}
/*
* Save a copy of the IP header in case we want restore it
* for sending an ICMP error message in response.
*/
save_ip = *ip;
save_ip.ip_len+= iphlen; /* tcp_input subtracts this */
/*
* Checksum extended UDP header and data.
*/
if (UDPCKSUM && uh->uh_sum) {
memset(&((struct ipovly *)ip)->ih_mbuf, 0, sizeof(struct mbuf_ptr));
((struct ipovly *)ip)->ih_x1 = 0;
((struct ipovly *)ip)->ih_len = uh->uh_ulen;
/* keep uh_sum for ICMP reply
* uh->uh_sum = cksum(m, len + sizeof (struct ip));
* if (uh->uh_sum) {
*/
if(cksum(m, len + sizeof(struct ip))) {
STAT(udpstat.udps_badsum++);
goto bad;
}
}
/*
* handle DHCP/BOOTP
*/
if (ntohs(uh->uh_dport) == BOOTP_SERVER) {
bootp_input(m);
goto bad;
}
if (slirp_restrict)
goto bad;
/*
* handle TFTP
*/
if (ntohs(uh->uh_dport) == TFTP_SERVER) {
tftp_input(m);
goto bad;
}
/*
* Locate pcb for datagram.
*/
so = udp_last_so;
if (so->so_lport != uh->uh_sport ||
so->so_laddr.s_addr != ip->ip_src.s_addr) {
struct socket *tmp;
for (tmp = udb.so_next; tmp != &udb; tmp = tmp->so_next) {
if (tmp->so_lport == uh->uh_sport &&
tmp->so_laddr.s_addr == ip->ip_src.s_addr) {
tmp->so_faddr.s_addr = ip->ip_dst.s_addr;
tmp->so_fport = uh->uh_dport;
so = tmp;
break;
}
}
if (tmp == &udb) {
so = NULL;
} else {
STAT(udpstat.udpps_pcbcachemiss++);
udp_last_so = so;
}
}
if (so == NULL) {
/*
* If there's no socket for this packet,
* create one
*/
if ((so = socreate()) == NULL) goto bad;
if(udp_attach(so) == -1) {
DEBUG_MISC((dfd," udp_attach errno = %d-%s\n",
errno,strerror(errno)));
sofree(so);
goto bad;
}
/*
* Setup fields
*/
/* udp_last_so = so; */
so->so_laddr = ip->ip_src;
so->so_lport = uh->uh_sport;
if ((so->so_iptos = udp_tos(so)) == 0)
so->so_iptos = ip->ip_tos;
/*
* XXXXX Here, check if it's in udpexec_list,
* and if it is, do the fork_exec() etc.
*/
}
so->so_faddr = ip->ip_dst; /* XXX */
so->so_fport = uh->uh_dport; /* XXX */
iphlen += sizeof(struct udphdr);
m->m_len -= iphlen;
m->m_data += iphlen;
/*
* Now we sendto() the packet.
*/
if (so->so_emu)
udp_emu(so, m);
if(sosendto(so,m) == -1) {
m->m_len += iphlen;
m->m_data -= iphlen;
*ip=save_ip;
DEBUG_MISC((dfd,"udp tx errno = %d-%s\n",errno,strerror(errno)));
icmp_error(m, ICMP_UNREACH,ICMP_UNREACH_NET, 0,strerror(errno));
}
m_free(so->so_m); /* used for ICMP if error on sorecvfrom */
/* restore the orig mbuf packet */
m->m_len += iphlen;
m->m_data -= iphlen;
*ip=save_ip;
so->so_m=m; /* ICMP backup */
return;
bad:
m_freem(m);
/* if (opts) m_freem(opts); */
return;
}
/*
* sendto() a socket
*/
int
sosendto(struct socket *so, struct mbuf *m)
{
int ret;
SockAddress addr;
uint32_t addr_ip;
uint16_t addr_port;
DEBUG_CALL("sosendto");
DEBUG_ARG("so = %lx", (long)so);
DEBUG_ARG("m = %lx", (long)m);
if ((so->so_faddr_ip & 0xffffff00) == special_addr_ip) {
/* It's an alias */
int low = so->so_faddr_ip & 0xff;
if ( CTL_IS_DNS(low) )
addr_ip = dns_addr[low - CTL_DNS];
else
addr_ip = loopback_addr_ip;
} else
addr_ip = so->so_faddr_ip;
addr_port = so->so_faddr_port;
/*
* test for generic forwarding; this function replaces the arguments
* only on success
*/
unsigned long faddr = addr_ip;
int fport = addr_port;
if (slirp_should_net_forward(faddr, fport, &faddr, &fport)) {
time_t timestamp = time(NULL);
slirp_drop_log(
"Redirected UDP: src: 0x%08lx:0x%04x org dst: 0x%08lx:0x%04x "
"new dst: 0x%08lx:0x%04x %ld\n",
so->so_laddr_ip, so->so_laddr_port,
addr_ip, addr_port,
faddr, fport, timestamp
);
}
addr_ip = faddr;
addr_port = fport;
sock_address_init_inet(&addr, addr_ip, addr_port);
DEBUG_MISC((dfd, " sendto()ing, addr.sin_port=%d, addr.sin_addr.s_addr=%08x\n", addr_port, addr_ip));
/* Don't care what port we get */
ret = socket_sendto(so->s, m->m_data, m->m_len,&addr);
if (ret < 0)
return -1;
/*
* Kill the socket if there's no reply in 4 minutes,
* but only if it's an expirable socket
*/
if (so->so_expire)
so->so_expire = curtime + SO_EXPIRE;
so->so_state = SS_ISFCONNECTED; /* So that it gets select()ed */
return 0;
}
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;
/* int code; */
DEBUG_CALL("icmp_input");
DEBUG_ARG("m = %lx", (long )m);
DEBUG_ARG("m_len = %d", m->m_len);
STAT(icmpstat.icps_received++);
/*
* 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 */
STAT(icmpstat.icps_tooshort++);
freeit:
m_freem(m);
goto end_error;
}
m->m_len -= hlen;
m->m_data += hlen;
icp = mtod(m, struct icmp *);
if (cksum(m, icmplen)) {
STAT(icmpstat.icps_checksum++);
goto freeit;
}
m->m_len += hlen;
m->m_data -= hlen;
/* icmpstat.icps_inhist[icp->icmp_type]++; */
/* code = icp->icmp_code; */
DEBUG_ARG("icmp_type = %d", icp->icmp_type);
switch (icp->icmp_type) {
case ICMP_ECHO:
icp->icmp_type = ICMP_ECHOREPLY;
ip->ip_len += hlen; /* since ip_input subtracts this */
if (ip_geth(ip->ip_dst) == alias_addr_ip) {
icmp_reflect(m);
} else {
struct socket *so;
SockAddress addr;
uint32_t addr_ip;
uint16_t addr_port;
if ((so = socreate()) == NULL) goto freeit;
if(udp_attach(so) == -1) {
DEBUG_MISC((dfd,"icmp_input udp_attach errno = %d-%s\n",
errno,errno_str));
sofree(so);
m_free(m);
goto end_error;
}
so->so_m = m;
so->so_faddr_ip = ip_geth(ip->ip_dst);
so->so_faddr_port = 7;
so->so_laddr_ip = ip_geth(ip->ip_src);
so->so_laddr_port = 9;
so->so_iptos = ip->ip_tos;
so->so_type = IPPROTO_ICMP;
so->so_state = SS_ISFCONNECTED;
/* Send the packet */
if ((so->so_faddr_ip & 0xffffff00) == special_addr_ip) {
/* It's an alias */
int low = so->so_faddr_ip & 0xff;
if (low >= CTL_DNS && low < CTL_DNS + dns_addr_count)
addr_ip = dns_addr[low - CTL_DNS];
else
addr_ip = loopback_addr_ip;
} else {
addr_ip = so->so_faddr_ip;
}
addr_port = so->so_faddr_port;
sock_address_init_inet( &addr, addr_ip, addr_port );
if(socket_sendto(so->s, icmp_ping_msg, strlen(icmp_ping_msg), &addr) < 0) {
DEBUG_MISC((dfd,"icmp_input udp sendto tx errno = %d-%s\n",
errno,errno_str));
icmp_error(m, ICMP_UNREACH,ICMP_UNREACH_NET, 0,errno_str);
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:
STAT(icmpstat.icps_notsupp++);
m_freem(m);
break;
default:
STAT(icmpstat.icps_badtype++);
m_freem(m);
} /* swith */
end_error:
/* m is m_free()'d xor put in a socket xor or given to ip_send */
return;
}
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 adresses */
ip->ip_src = m->slirp->vhost_addr;
(void ) ip_output((struct socket *)NULL, m);
end_error:
return;
}
/*
* recvfrom() a UDP socket
*/
void
sorecvfrom(struct socket *so)
{
struct sockaddr_in addr;
socklen_t addrlen = sizeof(struct sockaddr_in);
DEBUG_CALL("sorecvfrom");
DEBUG_ARG("so = %lx", (long)so);
if (so->so_type == IPPROTO_ICMP) { /* This is a "ping" reply */
char buff[256];
int len;
len = recvfrom(so->s, buff, 256, 0,
(struct sockaddr *)&addr, &addrlen);
/* XXX Check if reply is "correct"? */
if(len == -1 || len == 0) {
u_char code=ICMP_UNREACH_PORT;
if(errno == EHOSTUNREACH) code=ICMP_UNREACH_HOST;
else if(errno == ENETUNREACH) code=ICMP_UNREACH_NET;
DEBUG_MISC((dfd," udp icmp rx errno = %d-%s\n",
errno,strerror(errno)));
icmp_error(so->so_m, ICMP_UNREACH,code, 0,strerror(errno));
} else {
icmp_reflect(so->so_m);
so->so_m = NULL; /* Don't m_free() it again! */
}
/* No need for this socket anymore, udp_detach it */
udp_detach(so);
} else { /* A "normal" UDP packet */
struct mbuf *m;
int len;
#ifdef _WIN32
unsigned long n;
#else
int n;
#endif
m = m_get(so->slirp);
if (!m) {
return;
}
m->m_data += IF_MAXLINKHDR;
/*
* XXX Shouldn't FIONREAD packets destined for port 53,
* but I don't know the max packet size for DNS lookups
*/
len = M_FREEROOM(m);
/* if (so->so_fport != htons(53)) { */
ioctlsocket(so->s, FIONREAD, &n);
if (n > len) {
n = (m->m_data - m->m_dat) + m->m_len + n + 1;
m_inc(m, n);
len = M_FREEROOM(m);
}
/* } */
m->m_len = recvfrom(so->s, m->m_data, len, 0,
(struct sockaddr *)&addr, &addrlen);
DEBUG_MISC((dfd, " did recvfrom %d, errno = %d-%s\n",
m->m_len, errno,strerror(errno)));
if(m->m_len<0) {
u_char code=ICMP_UNREACH_PORT;
if(errno == EHOSTUNREACH) code=ICMP_UNREACH_HOST;
else if(errno == ENETUNREACH) code=ICMP_UNREACH_NET;
DEBUG_MISC((dfd," rx error, tx icmp ICMP_UNREACH:%i\n", code));
icmp_error(so->so_m, ICMP_UNREACH,code, 0,strerror(errno));
m_free(m);
} else {
/*
* Hack: domain name lookup will be used the most for UDP,
* and since they'll only be used once there's no need
* for the 4 minute (or whatever) timeout... So we time them
* out much quicker (10 seconds for now...)
*/
if (so->so_expire) {
if (so->so_fport == htons(53))
so->so_expire = curtime + SO_EXPIREFAST;
else
so->so_expire = curtime + SO_EXPIRE;
}
/*
* If this packet was destined for CTL_ADDR,
* make it look like that's where it came from, done by udp_output
*/
udp_output(so, m, &addr);
} /* rx error */
} /* if ping packet */
}
/*
* 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;
}
/*
* recvfrom() a UDP socket
*/
void
sorecvfrom(struct socket *so)
{
struct sockaddr_storage addr;
struct sockaddr_storage saddr, daddr;
socklen_t addrlen = sizeof(struct sockaddr_storage);
DEBUG_CALL("sorecvfrom");
DEBUG_ARG("so = %p", so);
if (so->so_type == IPPROTO_ICMP) { /* This is a "ping" reply */
char buff[256];
int len;
len = recvfrom(so->s, buff, 256, 0,
(struct sockaddr *)&addr, &addrlen);
/* XXX Check if reply is "correct"? */
if(len == -1 || len == 0) {
u_char code=ICMP_UNREACH_PORT;
if(errno == EHOSTUNREACH) code=ICMP_UNREACH_HOST;
else if(errno == ENETUNREACH) code=ICMP_UNREACH_NET;
DEBUG_MISC((dfd," udp icmp rx errno = %d-%s\n",
errno,strerror(errno)));
icmp_send_error(so->so_m, ICMP_UNREACH, code, 0, strerror(errno));
} else {
icmp_reflect(so->so_m);
so->so_m = NULL; /* Don't m_free() it again! */
}
/* No need for this socket anymore, udp_detach it */
udp_detach(so);
} else { /* A "normal" UDP packet */
struct mbuf *m;
int len;
#ifdef _WIN32
unsigned long n;
#else
int n;
#endif
m = m_get(so->slirp);
if (!m) {
return;
}
switch (so->so_ffamily) {
case AF_INET:
m->m_data += IF_MAXLINKHDR + sizeof(struct udpiphdr);
break;
case AF_INET6:
m->m_data += IF_MAXLINKHDR + sizeof(struct ip6)
+ sizeof(struct udphdr);
break;
default:
g_assert_not_reached();
break;
}
/*
* XXX Shouldn't FIONREAD packets destined for port 53,
* but I don't know the max packet size for DNS lookups
*/
len = M_FREEROOM(m);
/* if (so->so_fport != htons(53)) { */
ioctlsocket(so->s, FIONREAD, &n);
if (n > len) {
n = (m->m_data - m->m_dat) + m->m_len + n + 1;
m_inc(m, n);
len = M_FREEROOM(m);
}
/* } */
m->m_len = recvfrom(so->s, m->m_data, len, 0,
(struct sockaddr *)&addr, &addrlen);
DEBUG_MISC((dfd, " did recvfrom %d, errno = %d-%s\n",
m->m_len, errno,strerror(errno)));
if(m->m_len<0) {
/* Report error as ICMP */
switch (so->so_lfamily) {
uint8_t code;
case AF_INET:
code = ICMP_UNREACH_PORT;
if (errno == EHOSTUNREACH) {
code = ICMP_UNREACH_HOST;
} else if (errno == ENETUNREACH) {
code = ICMP_UNREACH_NET;
}
DEBUG_MISC((dfd, " rx error, tx icmp ICMP_UNREACH:%i\n", code));
icmp_send_error(so->so_m, ICMP_UNREACH, code, 0, strerror(errno));
break;
case AF_INET6:
code = ICMP6_UNREACH_PORT;
if (errno == EHOSTUNREACH) {
code = ICMP6_UNREACH_ADDRESS;
} else if (errno == ENETUNREACH) {
code = ICMP6_UNREACH_NO_ROUTE;
}
DEBUG_MISC((dfd, " rx error, tx icmp6 ICMP_UNREACH:%i\n", code));
icmp6_send_error(so->so_m, ICMP6_UNREACH, code);
break;
default:
g_assert_not_reached();
break;
}
m_free(m);
} else {
/*
* Hack: domain name lookup will be used the most for UDP,
* and since they'll only be used once there's no need
* for the 4 minute (or whatever) timeout... So we time them
* out much quicker (10 seconds for now...)
*/
if (so->so_expire) {
if (so->so_fport == htons(53))
so->so_expire = curtime + SO_EXPIREFAST;
else
so->so_expire = curtime + SO_EXPIRE;
}
/*
* If this packet was destined for CTL_ADDR,
* make it look like that's where it came from
*/
saddr = addr;
sotranslate_in(so, &saddr);
daddr = so->lhost.ss;
switch (so->so_ffamily) {
case AF_INET:
udp_output(so, m, (struct sockaddr_in *) &saddr,
(struct sockaddr_in *) &daddr,
so->so_iptos);
break;
case AF_INET6:
udp6_output(so, m, (struct sockaddr_in6 *) &saddr,
(struct sockaddr_in6 *) &daddr);
break;
default:
g_assert_not_reached();
break;
}
} /* rx error */
} /* if ping packet */
}
