This is the mail archive of the gdb@sourceware.org mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

infinite loop from backtrace in gdb (Powerpc64)


I am seeing an infinite loop from backtrace in gdb.6.3 on Powerpc64, looks
much like the problem Michael tried to fix for MIPS. In my case a library
that the app links to calls abort(), the remainder of backtrace from the
library seems cause gdb to go into an infinite loop.

Any clues?

 On Tue, 20 Apr 2004, Daniel Jacobowitz wrote:

> On Wed, Apr 21, 2004 at 12:20:53AM +0000, Michael Snyder wrote:
> > Hi Andrew,
> >
> > This change will prevent the caller(s) of mips_mdebug_frame_id from
> > infinite-looping when we get badly lost on the stack frame.
> >
>
> > 2004-04-21  Michael Snyder  <msnyder@redhat.com>
> >
> > 	* mips-tdep.c (mips_mdebug_frame_cache): Call error to prevent
> > 	infinite looping by caller.
> > 	(heuristic_proc_start): Warning() already prefixes "Warning: ".
>
> Since I have some patches to make this "I'm not sure how" case into a
> working part of the unwinder, I don't much like this.  They got hung up
> on the question of how to trust proc_desc's when we might be in the
> prologue.
>
> One time where this case occurs is in backtracing from a NULL pointer
> dereference, which happens in the GDB testsuite now.
>
>
> --
> Daniel Jacobowitz
> MontaVista Software                         Debian GNU/Linux Developer
>


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]