reliable core dump - swarm1.3.1 linux 2.0, tcl/tk8.0

[Mark E. Madsen]

After poking around a bit in the swarm-support archives, the only core
dumps that I see reported appear to be related to tcl/tk paths and
versions.  I see something different - if I don't have an environment
variable called SWARMHOME defined I get a core dump with the following
backtrace:

Core was generated by `./market'.
Program terminated with signal 11, Segmentation fault.
Reading symbols from /usr/local/lib/libspace.so.2...done.
Reading symbols from /usr/local/lib/libanalysis.so.1...done.
Reading symbols from /usr/local/lib/libsimtools.so.1...done.
Reading symbols from /usr/local/lib/libsimtoolsgui.so.1...done.
Reading symbols from /usr/local/lib/libtkobjc.so.2...done.
Reading symbols from /usr/local/lib/libactivity.so.0...done.
Reading symbols from /usr/local/lib/librandom.so.0...done.
Reading symbols from /usr/local/lib/libobjectbase.so.1...done.
Reading symbols from /usr/local/lib/libcollections.so.1...done.
Reading symbols from /usr/local/lib/libdefobj.so.0...done.
Reading symbols from /usr/local/lib/libmisc.so.1...done.
Reading symbols from /usr/local/lib/libtclobjc.so.0...done.
Reading symbols from /usr/local/lib/libBLT8.0.so...done.
Reading symbols from /usr/lib/libtk8.0.so...done.
Reading symbols from /usr/lib/libtcl8.0.so...done.
Reading symbols from /usr/local/lib/libpng.so.2...done.
Reading symbols from /usr/local/lib/libffi.so.1...done.
Reading symbols from /usr/X11R6/lib/libXpm.so.4.9...done.
Reading symbols from /usr/X11R6/lib/libX11.so.6.1...done.
Reading symbols from /lib/libm.so.5.0.6...done.
Reading symbols from /lib/libdl.so.1.7.14...done.
Reading symbols from /lib/libc.so.5.3.12...done.
Reading symbols from /lib/ld-linux.so.1...done.
#0  0x400e0f86 in _i_Arguments___getSwarmHome__ (self=0x80936c0,
_cmd=0x400ecaa4, 
    ignoreEnvFlag=0) at Arguments.m:419
419                 char sigPathBuf[strlen (swarmHome) + strlen
(SIGNATURE_PATH) + 1];
(gdb) bt
#0  0x400e0f86 in _i_Arguments___getSwarmHome__ (self=0x80936c0,
_cmd=0x400ecaa4, 
    ignoreEnvFlag=0) at Arguments.m:419
#1  0x400e1028 in _i_Arguments__getSwarmHome (self=0x80936c0,
_cmd=0x4006e840)
    at Arguments.m:435
#2  0x4005e055 in setSecondaryPath (arguments=0x80936c0,
dirname=0x40067070 "tk8.0")
    at internal.m:105
#3  0x4005e2a6 in tkobjc_initTkInterp (arguments=0x80936c0) at
internal.m:136
#4  0x4006030e in initTkObjc (arguments=0x80936c0) at tkobjc.m:40
#5  0x4003be2b in initSimtoolsGUI () at simtoolsgui.m:19
#6  0x40026720 in initSwarmAppArguments (argc=1, argv=0xbffffaa0,
version=0x0, 
    bugAddress=0x0, argumentsClass=0x0) at simtools.m:71
#7  0x400265a3 in initSwarm (argc=1, argv=0xbffffaa0) at simtools.m:31
#8  0x804c819 in main (argc=1, argv=0xbffffaa0) at main.m:22
#9  0x804b88b in _start ()


However, if I have SWARMHOME defined, even as nothing, it works just fine.
It seems like Arguments.m is relying on the existence of the variable.  Is
this a bug, or is this documented in the changes for 1.3.1?  I don't
recall this working in quite the same fashion in 1.2.

___________________________________________________________________________
            e m e r g e n t  m e d i a  i n c o r p o r a t e d 
                      http://www.emergentmedia.com                        
Mark Madsen (MM1417)                                 address@hidden
Principal                                            206.292.3990
___________________________________________________________________________


                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.