[15947] in Kerberos_V5_Development
Re: design choices for a loadable module interface
daemon@ATHENA.MIT.EDU (Nicolas Williams)
Wed Jun 30 12:46:08 2010
Date: Wed, 30 Jun 2010 11:47:44 -0500
From: Nicolas Williams <Nicolas.Williams@oracle.com>
To: Jeffrey Hutzelman <jhutz@cmu.edu>
Message-ID: <20100630164744.GH11785@oracle.com>
MIME-Version: 1.0
Content-Disposition: inline
In-Reply-To: <541735687952A0BABBC402F0@atlantis.pc.cs.cmu.edu>
Cc: krbdev@mit.edu
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Errors-To: krbdev-bounces@mit.edu
On Tue, Jun 29, 2010 at 11:50:11PM -0400, Jeffrey Hutzelman wrote:
> --On Tuesday, June 29, 2010 04:15:11 PM -0500 Nicolas Williams
> <Nicolas.Williams@oracle.com> wrote:
>
> >If you're using dynamic linking at all then there's no difference in how
> >hard it is to set such breakpoints in either scheme. If you don't agree
> >then please explain in detail.
>
> In gdb I can set a breakpoint on a symbol in a dynamic object that
> has not yet been loaded, such as a plugin. The debugger will defer
> the symbol lookup until an object is loaded containing that symbol,
> which is extremely convenient.
>
> It is easy to break on foo_do_something() even though the plugin foo
> has not been loaded. It's rather harder to break on the plugin
> foo's do_something() if every plugin has a function by that name.
Is it really harder? I use dbx, and in dbx we have a way to name
modules. Surely gdb has the same feature! It's just a matter of typing
the right sequence of characters, and it's going to be very similar in
both cases. So I see nothing here distinguishing one solution from the
other.
Nico
--
_______________________________________________
krbdev mailing list krbdev@mit.edu
https://mailman.mit.edu/mailman/listinfo/krbdev
