On 8/14/24 16:47, Tom de Vries wrote:
> On 3/14/24 18:45, Aaron Merey wrote:
>> We should be able to reproduce this in a testcase by overwriting part of
>> a small binary with zeros.  I'll resubmit the patch with a testcase.
> 
> I ran into this problem as well via PR32081 (where correct debuginfo is 
> read incorrectly).

After writing the test-case for this patch ( 
https://sourceware.org/pipermail/gdb-patches/2024-August/211114.html ) I 
realized it would be easy to modify it to trigger (a form of) this 
problem.  Result attached.

Feel free to ignore this, incorporate this into a v2, or suggest to me 
to submit a v2 in case you don't have the bandwidth to work on this any 
time soon.

Thanks,
- Tom