A lean/small GDB ?

[Donny Kurniawan]

Hello,

I think I need to re-think my approach to my scalability problem (i.e.
debugging 1000 processes).... by solving it from a different angle.

I am thinking to reduce GDB (either through configure or removing some code,
e.g. readline) so that it can be as lean/small as possible while retaining
essential debug functionalities. Hopefully, the small size can help in
the scalability.

I am thinking to remove the UI and make GDB like "libGDB" by putting API-like
interface.

Is this approach feasible/easy-to-do?
Any comments/opinions are greatly appreciated.

Thanks,
Donny