Andrew Doran <ad@netbsd.org> wrote:
> With this change, spc_lock() will only lock LWPs that are on a run queue.
> Can anyone see a potential situation where this is likely to cause a problem
> for the scheduler?
But we can safely lock sq_mutex -> spc_mutex to lock both the sleeping and
running LWPs, can't we? In such case, I do not see the situation which could
cause a problem for me.

> I've attached lockstat output below. It's from a 10 second run on a 4-way
> system, with 10 user threads playing a game with a condition variable &
> mutex pair.
Looks great!

-- 
Best regards,
Mindaugas
www.NetBSD.org