On Thu, Mar 15, 2007 at 02:49:01AM +0200, Mindaugas R. wrote:
> 
> P.S. Hey, where is the 3'rd voice in such cases?

Although I'm not completely sure wether this would really
work, here's another suggestion. The problem is that you
need a distinction between "enqueue for yielding" and 
"normal" operation, right? Maybe it would just be enough
to check for the according schedstate_percpu flag
(SPCF_SEENRR or SPCF_SHOULDYIELD) within your implementation
of sched_enqueue(). If this is a feasable solution we could
move back to sched_nextlwp()/sched_switch() without an
implicit enqueue rule (which I don't like too much).

Hope this helps,
Daniel

-- 
Daniel Sieger
Faculty of Technology
Bielefeld University
wwwhomes.uni-bielefeld.de/dsieger