RE: Segmentation Fault when trying to make Glib 1.06

[Ross McFarland <rwmcfa1 neces com>]

On Tue, 2004-09-28 at 13:00, John Ellis (Interlake Industries, Inc.)
wrote:
> Hi Ross,
>       Thanks for responding
>
> The output from perl -V follows.
>  perl -V
> Summary of my perl5 (revision 5.0 version 8 subversion 0) configuration:
>   Platform:
>     osname=linux, osvers=2.4.20, archname=i586-linux-thread-multi
>     uname='linux d20 2.4.20 #1 smp thu oct 10 18:10:26 utc 2002 i686 unknown
> unknown gnulinux '

nothing out of the ordinary or tell-tale there.

>     usethreads=define use5005threads=undef useithreads=define
> usemultiplicity=define

> Per the above I would say I am running threaded?

useithreads=define means yeah.

> I forgot to run "make test".
> I'll run it tonite and let you know result but if I remember correctly it
> fails.

you ever run make test?

it unfortunately doesn't seem to be anything obvious. and i don't have a
system that's having the same problem(s.) the next steps (if you're
still having issues) would be to figure out what code is causing the
core dump.

get core files on seg faults if you're not already
$ set ulimit unlimited 

then try to build
$ make

you should then get a core file when it seg-faults. you can then load
this core file with gdb and hopefully get some idea of where things are
going wrong.

$ gdb perl <corefile>
...
(gdb) bt

the output from the invocation of gdb on would be of interest. the 'bt'
will do a back track from the point of the core dump all the way back up
the stack. (it can be surprisingly deep.)

if you want to hop on to irc.gnome.org, #gtk-perl muppet, kaffee, and/or
myself may be able to help.

-- 
-rm
http://www.neces.com/