Re: RFC: target_create_inferior that does not call proceed

[Daniel Jacobowitz <drow@false.org>]

On Mon, Jan 16, 2006 at 09:58:39PM +0100, Mark Kettenis wrote:
> > I'll be needing to run some additional common code fragments between
> > creating an inferior and starting it, for an upcoming project.  Like most of
> > GDB they need the target to be stopped; that's why I separated the create
> > and proceed phases here.
> 
> That's the reason why you created post_create_inferior?  If yes, then,
> that answers the next question: Why did you introduce post_create_inferior?

That's right.  The followup patch I posted today calls solib functions
from there.  The followups I'll be working on later in the week will
hook the "target properties" and "available register sets" bits I
discussed on gdb@ (a long time ago now) into the same place.

> I vaguely remember there was a way to do that already, but when I last
> wanted to use it, I couldn't find a way to do it.  So yes, I'd welcome
> such a command.  Perhaps we could have something like:
> 
> (gdb) set stop-on-entry 1
> 
> after which "run" or "start" would stop at the entry point?

Hrm, I'm not a big fan of state variables, but at least that's clearer
than anything else I've thought of.  I suppose we could also call the
command create-inferior :-)

-- 
Daniel Jacobowitz
CodeSourcery