Subject: Re: SMP/flogging a dead horse
To: Erik E. Fair <fair@clock.org>
From: Mason Loring Bliss <mason@acheron.middleboro.ma.us>
List: current-users
Date: 08/29/1998 11:58:23
On Sat, Aug 29, 1998 at 08:45:19AM -0700, Erik E. Fair wrote:

> I'm beginning to wonder if it wouldn't be quicker to do minimal changes to
> NetBSD to support an MP (not SMP) configuration, where the additional
> processor(s) are strictly user mode to get some support out there.

I don't know a whole lot about SMP, but isn't it mostly the case that MP
would give us essentially the same performance boost, if not potentially
more performance, given the lack of kernel code running on the secondary
processors?

Feel free to enlighten me if I'm missing something... It would seem that
the only difference would be that the kernel would have to know about
specific processors and the processes running on them, as opposed to seeing
a globbed together collection of processors that operate as a single source
of available computing power.

Thoughts?

PS: A pointer to somewhere that discusses this in a form that people who
are new to it can digest would be welcome.

-- 
Mason Loring Bliss..mason@acheron.middleboro.ma.us..acheron.dyn.ml.org/mason
"In the drowsy dark cave of the mind dreams build their nest with fragments
  dropped from day's caravan."--Rabindranath Tagore..awake ? sleep : dream;