Date: Tue, 15 Dec 1998 04:05:06 -0800 From: Mike Smith <mike@smith.net.au> To: Richard Wackerbarth <rkw@nomad.dataplex.net> Cc: Mike Smith <mike@smith.net.au>, Kenjiro Cho <kjc@csl.sony.co.jp>, NAKAGAWA Yoshihisa <y-nakaga@nwsl.mesh.ad.jp>, current@FreeBSD.ORG Subject: Re: PAO Integration? Message-ID: <199812151205.EAA05066@dingo.cdrom.com> In-Reply-To: Your message of "Tue, 15 Dec 1998 05:34:32 CST." <Pine.BSF.4.05.9812150521010.7087-100000@nomad.dataplex.net>
next in thread | previous in thread | raw e-mail | index | archive | help
>
>
> On Sun, 13 Dec 1998, Mike Smith wrote:
>
> > It's worth pointing out that we are now quite a long way down the path
> > to this goal; by no means all the way there, but the new bus
> > architecture coupled with KLD modules does largely obviate the need for
> > config(8) in the first place.
> >
> > I'm of the strong opinion that our current direction, taking us away
> > from static configuration, is the right one to be taking in the context
> > of our current and projected future target architectures. It is no
> > longer adequate nor desirable to require the kernel to be rebuilt to
> > adapt to a new system configuration, and we need to reflect this in our
> > architecture.
>
> I support Mike's opinion on this matter. At the same time, I think that
> the PAO folks recognize that there is a need to support the rogue devices
> that cannot be handled in any fashion other than manual intervention
> ("Damn it! I know that this card uses IRQ .. and .... . Just do it that
> way.")
There's no way that we're going to obsolete "manual configuration" in
the forseeable future; ISA is not going to die that easily I fear.
> Can we have a scheme whereby the "ultra-new-config" utility generates
> glue modules for just those devices? By adding the resulting KLDs to the
> load mix, the legacy hardware would be accomodated in the same framework
> that the rest of the dynamic system uses.
The problem that we're seeing in this discussion is the confusion
between configuration information and driver code, or the module and
its metadata.
There will always be a mechanism for passing metadata to an instance of
a driver.
--
\\ Sometimes you're ahead, \\ Mike Smith
\\ sometimes you're behind. \\ mike@smith.net.au
\\ The race is long, and in the \\ msmith@freebsd.org
\\ end it's only with yourself. \\ msmith@cdrom.com
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199812151205.EAA05066>