--K8nIJk4ghYZn606h
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Wed, Mar 09, 2005 at 12:27:52AM +0100, Daniel Sieger wrote:
> sophisticated like a pluggable scheduler framework found in Solaris or

I think this is more along the lines of what we want, but I'm
undecided.  As Bill said, I think the first step is to make
the scheduler code more MI, which will give us more flexibility
right off.

I'd like to see some support for the current model (effectively),
soft real-time, and the ability to put something else in.  I'm
not sure what it would look like, though I would want it to be
a light-weight framework so that the impact of using the current
model would be no (or very little) compared to what we have in
place now.  But I care about underpowered (by today's standards)
systems.

I know some folks have given it more thought than I have, and
there are some folks who have done a lot of work with the
scheduler in the past.  I'll let them identify themselves,
though.

As for level of effort, I think that's going to be hard to gauge
without having some design for what, exactly, you want to do and
where you're coming from.

-allen

--=20
                  Use NetBSD!  http://www.netbsd.org/

--K8nIJk4ghYZn606h
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (NetBSD)

iD8DBQFCLleAtbG21IdtLQIRAnc0AKD412/1vBDi2D8ZVx3mQjBVSOUzfwCeIua7
15ysG1CSt1o9W/I7M+zr4i0=
=DhaM
-----END PGP SIGNATURE-----

--K8nIJk4ghYZn606h--