On Mon, Feb 12, 2001 at 06:16:22PM +0900, [EMAIL PROTECTED] wrote:
> try "host 127.0.0.1 or host ::1".
Note that the above would go through a different code path, so it'd be a
test of whether there's something wrong with the code path in question -
which is what we want.
Speaking of that code, it appears that
memset(&mask128, 0xff, sizeof(mask128));
res0 = res = pcap_nametoaddrinfo(name);
if (res == NULL)
bpf_error("unknown host '%s'", name);
b = tmp = NULL;
tproto = tproto6 = proto;
if (off_linktype == -1 && tproto == Q_DEFAULT) {
tproto = Q_IP;
tproto6 = Q_IPV6;
}
for (res = res0; res; res = res->ai_next) {
switch (res->ai_family) {
case AF_INET:
if (tproto == Q_IPV6)
continue;
sin = (struct sockaddr_in *)
res->ai_addr;
tmp = gen_host(ntohl(sin->sin_addr.s_addr),
0xffffffff, tproto, dir);
break;
case AF_INET6:
if (tproto6 == Q_IP)
continue;
sin6 = (struct sockaddr_in6 *)
res->ai_addr;
tmp = gen_host6(&sin6->sin6_addr,
&mask128, tproto6, dir);
break;
}
if (b)
gen_or(b, tmp);
b = tmp;
}
could malfunction if "res->ai_family" is neither AF_INET nor AF_INET6
(yes, perhaps that "should not happen", but sometimes stuff that "should
not happen" happens anyway). However, I'd expect it to crash in
backpatch(b0, b1->head);
when it tries to dereference "b1", rather than blowing up in the bowels
of the optimizer.
It might be interesting to see what happens if
if (b)
gen_or(b, tmp);
is changed to
if (b && tmp)
gen_or(b, tmp);
-
This is the TCPDUMP workers list. It is archived at
http://www.tcpdump.org/lists/workers/index.html
To unsubscribe use mailto:[EMAIL PROTECTED]?body=unsubscribe
- [tcpdump-workers] Core dump inside pcap optimizer. Snor... Javier Kohen
- Re: [tcpdump-workers] Core dump inside pcap optimi... Guy Harris
- Re: [tcpdump-workers] Core dump inside pcap op... itojun
- Re: [tcpdump-workers] Core dump inside pca... Guy Harris
- Re: [tcpdump-workers] Core dump inside... Torsten Landschoff
- Re: [tcpdump-workers] Core dump i... itojun
- Re: [tcpdump-workers] Core du... Torsten Landschoff
- Re: [tcpdump-workers] Cor... Guy Harris
- Re: [tcpdump-workers] Cor... itojun
- Re: [tcpdump-workers] Cor... Guy Harris
- Re: [tcpdump-workers] Cor... itojun
- Re: [tcpdump-workers] Cor... Guy Harris
- Re: [tcpdump-workers] Cor... Torsten Landschoff
- Re: [tcpdump-workers] Cor... Guy Harris
- Re: [tcpdump-workers] Cor... Torsten Landschoff
