[gst-devel] Question: How do I obtain a better gdb back traceof segmentation faults?

[Maciej Katafiasz]

Dnia 10-01-2005, pon o godzinie 19:48 -0500, David Parr napisał:
> RELATED QUESTION: ("Maciej Katafiasz") how did you know that
> "backtrace has definitely been stomped on, as can be seen from #..." ?
> Are there valid ranges for stack frames in linux?  See my #25 from the
> new trace.  It looks much like the frames you identified in my
> previous trace.

Yes, every system has characteristic address ranges in which you can
expect various things to be placed. After some time spent in debugger
you can catch clearly invalid references at a glance. Addresses as small
as 0x01 tend to be always invalid on Linux, hence my suspicions it was
heavily stomped on. But David has more clue anyway than I have, as he
knows stuff like stack frames inside out, unlike me :)

Cheers,
Maciej

-- 
Maciej Katafiasz <ml at mathrick.org>