Hi,
I've also been trying to run dia under NetBSD
and got the same:
> "Couldn't find standard object when looking for object lib..."
message (the so files are named .so.0.0). So I made links to them
with just .so and also tried renaming the file.
When I start dia I get a seg fault. Here's the stack trace...
Any suggestions about where to go from here?
Thanks.
- Scott Presnell ([EMAIL PROTECTED])
#0 0x0 in ?? ()
#1 0x401ee680 in g_module_open (
file_name=0xefbfd47c
"/usr/local/lib/dia/libstandard_objects.so",
flags=G_MODULE_BIND_LAZY) at gmodule.c:266
#2 0x6378 in register_objects_in (directory=0x673a
"/usr/local/lib/dia")
at app_procs.c:425
#3 0x680a in register_all_objects () at app_procs.c:503
#4 0x596d in app_init (argc=1, argv=0xefbfd75c) at
app_procs.c:195
#5 0x12967 in main (argc=1, argv=0xefbfd75c) at main.c:28
...so in digging a little deeper, gmodule seems broken under NetBSD
1.4.1 . I've
checked 1.2.[67] Why? I have not yet figured it out, but the offending
line
is listed above: 266 in gmodule.c is a check_init() call to see if
you've loaded
the .so library. If you remove that check (and therefore assume the
library is
loaded correctly) the program doesn't dump, and apparently things work
out
alright (dia runs correctly), but I'd rather see this fixed.
Where is the best place to report a bug like this? I don't see a bug
tracking
mechanism on www.gtk.org.
Thanks.
- Scott
If I'm way off because someone else with NetBSD can make dia (and
therefore
gmodule) work, please speak up. I can confirm the problem with the
simple
testgmodule program supplied with the glib distribution.
