Hi list,
I'm struggling with a wicked bug. It looks like it is located in the WSP
dissector although I suspect some code in the WSP/MMSE dissector may corrupt
the stack (if I may believe gdb):
Program received SIGSEGV, Segmentation fault.
0x61093644 in strtosigno () from /usr/bin/cygwin1.dll
(gdb) bt
#0 0x61093644 in strtosigno () from /usr/bin/cygwin1.dll
#1 0x6103ec86 in malloc () from /usr/bin/cygwin1.dll
#2 0x0022d3d8 in ?? ()
#3 0x00000001 in ?? ()
#4 0x610882ce in cygwin1!__getreent () from /usr/bin/cygwin1.dll
Previous frame inner to this frame (corrupt stack?)
It does not appear with the 0.10.0 version of packet-mmse.c, but it
invariably appears with the post-0.10.0 version :(
If I add debug logging in WSP, then I see that add_content_type is called
from within the MMSE dissector, and I get up to the instant before the
"Start" parameter value will be appended to the Content-Type header field
value. It looks like it's the proto_item_append_string() call which never
returns... Anyone a clue?
BTW what do CLEANUP_PUSH(g_free, strval) and CLEANUP_CALL_AND_POP do?
Regards,
Olivier