Re: [PATCH v2 00/31] Baby step for objfile splitting

[Andrew Burgess <aburgess@redhat.com>]

Tom Tromey <tom@tromey.com> writes:

>>>>>> "Andrew" == Andrew Burgess <aburgess@redhat.com> writes:
>
> Andrew> Tom Tromey <tom@tromey.com> writes:
>>> Every year or so, I take a stab at the objfile splitting work.  This
>>> always fails and I end up not sending anything.
>
> Andrew> For those of us who aren't familiar with the background here.  Do you
> Andrew> have a link to a previous description of, or could you give a
> Andrew> description of, what the "objfile splitting work" is about.  What's the
> Andrew> longer term goal, of which this series is a step towards.
>
> The basic idea is that debug info should be independent of the program
> space.  That way, symbol tables can be shared across inferiors.
>
> The project is somewhat documented here:
>
> https://sourceware.org/gdb/wiki/ObjfileSplitting
>
> It's been vaguely ongoing for a very long time.  I managed to implement
> it for minsyms, psyms, and line tables; but symbols and blocks have
> proven to be more difficult, with a trail of failed branches behind
> them.

Thanks.  I have heard this discussed before, I'd just not associated
with that name.

Thanks,
Andrew