Olaf Seibert <rhialto@mbfys.kun.nl> writes:

>2. Perhaps you can try to determine differences between the working and
>the non-working version. (cmp -l, using nm to find out in which routine
>a difference occurs) The first sign of a difference may be enough hint
>to find out what's different and why.

Pretty difficult. The A2024 code seems to rely on some uninitialized
variables, it does work or does not work without any changes to the
A2024 code at all. Same for iteconfig, after some tries you often
end with a pure garbage display and a few more tries will panic the
kernel (probably a corrupted chip memory list).

Regards,
-- 
                                Michael van Elst

Internet: mlelstv@serpens.rhein.de
                                "A potential Snark may lurk in every tree."