Re: strange segmentation fault error!

[Jan Kratochvil <rcpt-gtk-devel-list AT gnome org jankratochvil net>]

Hi,

Just some general GNU debugging notes:


On Tue, 07 Oct 2003 09:37:54 +0200, jscym singnet com sg wrote:
...
> I encountered a strange bug in my program and would like to seek the
> expertise of the group.
...
> the program terminates with a segmentation fault error. 

General first step: Run it as:
	LD_PRELOAD=/usr/lib/libefence.so ./testapp


> On debugging with gdb, the error that caused the segmentation fault was a
> file called eval.c could not be located.

Compile both your application, ! glib !, and gtk+ library with -ggdb3.
For rpm case (replace gtk with glib etc.):
	rpm -i gtk*.src.rpm
	modify /usr/src/*/SPECS/gtk*.spec
		-%configure --with-xinput=xfree --disable-gtk-doc
		+export CFLAGS="-ggdb3"
		+%configure --with-xinput=xfree --disable-gtk-doc --enable-debug=yes
	rpmbuild -bb /usr/src/*/SPECS/gtk*.spec
	rpm -U /usr/src/*/RPMS/i386/gtk*.i386.rpm


> Here is the error msg from gdb:
> program received signal SIGSEGV, segmentation faults.
> 0x00FA0000 in ??() at eval.c:41
> 41  eval.c:No such file or directory
>      in eval.c
> 
> and on using backtrace, the following msg was obtained:
> #0  0x00FA0000 in ??() at eval.c:41
      ^^^^^^^^^^
      This is just bogus address - myfunction() jumped to junk
      or some more serious stack frame corruption occured.

> #1  0x0804B9C3 in myfunction(data=0x0) at main.c:124
                                            ^^^^^^^^^^ what is on that line of yours?
> #2  0x4019B691 in g_timeout_dispatch() at eval.c:41
> #3  0x4019A773 in g_main_dispatch() at eval.c:41


...
> I am currently using GTK+ 1.2. Should I upgrade to GTK+ 2.0?

Before you upgrade I would not expect much help from the community.
1.x is deprecated now.


Bye,
Lace