Re: [PATCH] gdb/build-id: protect against weirdly short build-ids

[Andrew Burgess <aburgess@redhat.com>]

Kevin Buettner <kevinb@redhat.com> writes:

> On Tue, 19 Nov 2024 10:00:28 +0000
> Andrew Burgess <aburgess@redhat.com> wrote:
>
>> I propose that, in build_id_to_bfd_suffix we just return early if the
>> build-id is 1 byte (or less) with a return value that indicates no
>> separate file was found.
>
> I wonder if also printing some sort of warning might be appropriate in
> such cases?  If a general warning isn't suitable, I think it makes
> sense to at least use separate_debug_file_debug_printf() to output
> a suitable message when separate-debug-file debugging is enabled.

I posted a v2 which includes a debug log message that the lookup was
abandoned.  I was tempted with the idea of a warning, but wasn't sure if
that was justified or not... I'm still not really sure.  If you have a
strong preference for a warning then I can switch to that approach.

Anyway, let me know what you think of v2.

Thanks,
Andrew