Re: [RFA] Async mode fixes.

[Daniel Jacobowitz <drow@false.org>]

On Sun, Mar 09, 2008 at 12:08:31PM +1300, Nick Roberts wrote:
>  > > My changes aren't just for Linux, but an exec target.  I mean native
>  > > debugging (I think) with just one implementation - linux.  It sounds like
>  > > I'm inflating what I've done but other native targets can presumably be
>  > > adapted to make use of the changes in inf-ptrace.c, exec.c etc.
>  > 
>  > I think you mean "child" target?  target exec just reads an exec file
>  > on disk.
> 
> Well, my changes are in exec.c to exec_ops methods, not to target methods in
> inf-child.c.  Then ISTR these get inherited by target methods in inf-ptrace.c
> 
> Are you saying that these changes should really be in inf-child.c or just that
> I've misunderstood the concept of an exec target?

An exec target is the bit that owns the executable file.  Then other
targets live on top of it (on the "target stack") - target remote,
target sim, target child (which is just "run").  Putting things in the
exec target and not overriding them in any other target means that
they will work whenever there is an executable file, regardless of
target, and not work if we are attached without an executable file.

So if they go in exec.c, perhaps they shouldn't be in the target
vector at all.  Or perhaps they should be the default versions in
target.c if no target overrides them (like
default_region_ok_for_hw_watchpoint).

-- 
Daniel Jacobowitz
CodeSourcery