On Wed, Mar 09, 2005 at 12:27:52AM +0100, Daniel Sieger wrote:
> Just something like the O(1) scheduler Linux has

I've never seen more detailed description than the marketing fuzz about
this. We certainly need to put in some modern data structures to reduce
scheduling costs, but what else does Linux hide under this bogus name?
Is there any paper on it's design?

> a port from FreeBSD's ULE if it turns out to be stable

When I looked at the ULE paper (which didn't even tell me the game of the
name, so I consider it poor written ;-} ) it seemed to leave out important
details on their support for hyper threading intel CPUs. Did they realy
get that running with performance benefits?

> or something more
> sophisticated like a pluggable scheduler framework found in Solaris or
> Linux?

I'm with Bill that we probably want something like that in the longer
time frame, but that the first step is to minimize the MD parts.

Martin