The recent changes to emem.c are turning up previously undetected
errors. Unfortunately, the current code aborts abruptly without
providing any useful information. (Canary values in emem.c are checked
some time after they've been overwritten, so I'm not sure there _is_ a
way to provide any useful information other than "a canary got stomped.")
Bugs 665 & 666 were caused by the SRVLOC dissector. I pasted valgrind
output into bugs 667 and 669, but I haven't had a chance to look at them
yet.
To generate the valgrind output I uncommented the lines
/* #define EP_DEBUG_FREE 1 */
/* #define SE_DEBUG_FREE 1 */
in emem.c, recompiled, and ran tethereal under valgrind, e.g.
libtool --mode=execute valgrind ./tethereal \
-nVxr /tmp/editcap.43c51fa3.pcap > /dev/null
Jaap Keuter wrote:
> Hi,
>
> Is the FreeBSD buildbot going crazy? It's filling the same problem over
> and over again.
>
> Thanx,
> Jaap
>
> On Wed, 11 Jan 2006, Buildbot wrote:
>
>
>>Problems have been found with the following capture file(s):
>>
>>http://www.ethereal.com/distribution/buildbot-builds/randpkt/editcap.43c51fa3.pcap
>>
>>
>>Error information:
>>Core was generated by `lt-tethereal'.
>>Program terminated with signal 6, Aborted.
>>#0 0x2000000042984390 in kill () from /lib/libc.so.5
>>#0 0x2000000042984390 in kill () from /lib/libc.so.5
>>No symbol table info available.
>>#1 0x0000000000000000 in ?? ()
>>No symbol table info available.
>>../../helpers/gdb.cmds:2: Error in sourced command file:
>>Previous frame identical to this frame (corrupt stack?)
>>
>>
>>(no stderr output)
>>
>>
>>Bug 669 posted.
>>
>>_______________________________________________
>>Ethereal-dev mailing list
>>Ethereal-dev@xxxxxxxxxxxx
>>http://www.ethereal.com/mailman/listinfo/ethereal-dev
>>
>
>
> _______________________________________________
> Ethereal-dev mailing list
> Ethereal-dev@xxxxxxxxxxxx
> http://www.ethereal.com/mailman/listinfo/ethereal-dev
