Re: Current (non-) state of gdbserver

[jtc@redback.com (J.T. Conklin)]

>>>>> "Daniel" == Daniel Jacobowitz <dmj+@andrew.cmu.edu> writes:
Daniel> Unless someone steps up with something already done (if you're
Daniel> out there, we're waiting...), I'm going to start working on
Daniel> this.

As far as I know, you're the first with both the time and inclination
to address this.  

Daniel> I'd also like to start building gdbserver by default on
Daniel> platforms which support it (and I have patches to extend the
Daniel> list a bit).  That way we can at least notice this sort of
Daniel> thing...

Ideally, you'd want to build a cross gdbserver when you build a cross
debugger, but that involves all sorts of mess figuring out what tools
to use to build gdbserver.  That could be pretty tricky to figure out.
In the mean time, just building gdbserver for native configs would
catch a lot of the bitrot that's occured from time to time over the
years.

But until you/we come up with a multi-arch strategy, we may run into
conflicts when an existing/working gdbserver port breaks when gdb is
multi-arched (for a new architecture).  Up till now, we've accepted
(for the most part implicitly, since usually we didn't realize that
gdbserver had broken until much later) that breakage as the price of
progress.

        --jtc

-- 
J.T. Conklin
RedBack Networks