Re: Move gdbsupport to the top level

[Tom Tromey <tom@tromey.com>]

>>>>> "Pedro" == Pedro Alves <palves@redhat.com> writes:

Pedro> (I'm looking at the branch, so I'm not sure I'm commenting on the patch
Pedro> as it was posted.)

I think I made a few changes based on reviews, though I don't fully
recall.

Pedro> This all builds fine for me when I build normally, but,
Pedro> this breaks building gdbserver standalone, without configuring 
Pedro> from the top level.  I.e.:

I will take a look.

Meanwhile you should also look at this:

    https://sourceware.org/ml/gdb-patches/2019-08/msg00053.html

... because I think that's another blocker to this project.

That message neglected option 4, which is removing readline from the
tree.  (But making it continue to work if the sources are dropped in, as
we already do for libiconv etc.)

Pedro> What are the new rules here?  Add <config.h> on as-needed basis,
Pedro> or should we have some nat.h file that is included by
Pedro> all nat/ files, and same for arch/ ?  The former seems a bit
Pedro> error prone, given that you could move code around and not realize
Pedro> that an #ifdef is disabling something because you missed config.h.

Pedro> Alternatively, I guess we could move the required bits from
Pedro> gdb&gdbserver's configury to gdbsupport's, so that config.h
Pedro> wasn't ever necessary in shared code.  Not sure whether that
Pedro> would be a bit of an abstraction violation.

Yeah, I just did what was needed to make it work, but I didn't really
come up with a long-term plan.

Maybe moving nat and arch to gdbsupport is the another option?  That
would require some #include adjuments but otherwise maybe it's not such
a big deal.

Tom