Subject: Re: New spl level: splsched()
To: Erik E. Fair <fair@clock.org>
From: Jim Reid <jim@mpn.cp.philips.com>
List: tech-kern
Date: 03/03/1997 13:12:19
>>>>> "Erik" == Erik E Fair <(Time Keeper) <fair@clock.org>> writes:

    Erik> Isn't it time we started using a new mutual exclusion
    Erik> technique that is semaphore based, instead of interrupt
    Erik> based? We're going to need that if we ever have any SMP
    Erik> aspirations.

Indeed. But this is hard to do right, especially lock contention and
deadlock avoidance/detection.

IIRC Sequent's kernel has very fine grained locking - at the
individual buffer/resource level - so that they get near linear
throughput until they have ~40 processors.