> Hi,
>
> I had loads of problems compiling ethereal 0.7.7 under
> AIX 4.3.2 with the IBM xlc 3.6.6 compiler.
>
> xlc did not like the C++ style '//' comments when compiling
> C code.
Fixed in the next release.
> I had other problems with things like guint8 not casting
> to the correct type.
>
> After fixing a few things, I got things to compile,
Could you send us a list of the changes you made, so we can incorporate
them into the next release?
> but when I tried to run it, I got a coredump. The stack
> trace is:
>
> Segmentation fault in glink.g_main_pending at 0xd1088884 ($t1)
> 0xd1088884 (g_main_pending+0x80) 800c0000 lwz r0,0x0(r12)
Oh, yuk. That routine consists of:
gboolean
g_main_pending (void)
{
return in_check_or_prepare ? FALSE : g_main_iterate (FALSE, FALSE);
}
and "in_check_or_prepare" is just a static integer variable and
"g_main_iterate()" is just a routine in "gmain.c", so there's something
odd going on here.
Could you send us a disassembly of "g_main_pending()", in the hopes that
we could at least try to guess what that instruction is *supposed* to be
doing? (I'm not an AIX expert, but I might be able to guess, and there
may be folks here who are AIX experts - so I'm CCing "ethereal-dev" as
well.)
> (dbx) where
> glink.g_main_pending() at 0xd1088884
> gtk_init_check(0x2ff22b18, 0x2ff22b1c), line 199 in "gtkmain.c"
> gtk_init(0x2ff22b18, 0x2ff22b1c), line 423 in "gtkmain.c"
> main(argc = 1, argv = 0x2ff22b48), line 620 in "main.c"
>
> I compiled gtk+-1.2.6 and glib-1.2.6 also.
> Any ideas?
>
> Thanks.
> --
> Craig Rodrigues
> http://www.gis.net/~craigr
> rodrigc@xxxxxxxxxxxx
