Subject: Re: MP/Threading...
To: Greywolf <firstname.lastname@example.org>
From: Nathan J. Williams <nathanw@MIT.EDU>
Date: 03/20/2001 14:57:11
Greywolf <email@example.com> writes:
> I was seriously hoping that BigLock would never see an official release.
> It just seems the wrong thing to do. Yes, it's a stepping stone, and should
> probably be made available in a -current or -mp branch, but it just seemed
> to me that we could beat the game by being the first OS to skip the BigLock
> scheme on a release and just go *wham* "Here we are - granular-lock SMP".
Big-lock versus fine-grain SMP is, for most purposes, below the
application programming abstraction barrier. There's no technical
benefit to withholding big-lock SMP from a release just because it's
not the best SMP implementation.
The granularity of an SMP implementation can change over time and
through releases (probably getting finer grain, but not always...),
and users will see the benefit as it improves. It's not like exposing
a half-baked kernel interface that we'll then be forced to support in
the half-baked state.