Subject: Re: About SMP
To: None <port-i386@NetBSD.ORG>
From: Greg A. Woods <woods@most.weird.com>
List: port-i386
Date: 06/30/1998 00:12:05
[ On , June 24, 1998 at 11:54:05 (GMT), Heiko W.Rupp wrote: ]
> Subject: Re: About SMP
>
> In a recent german c't magazine, they explained it with the fact, that you
> need some specific interrupt register/controller on the cpu. Intel uses
> its proprietary controller for this, while the others adhere to some
> open standard. Dunno if this is right though.

In reading through the data sheets for the Intel 440LX chipset today I
discovered what appear to be some rudiments of support for redundant CPU
support -- i.e. a master/slave dual-CPU scenario where one CPU checks
the results of the other and reports any discrepancies through the SMB.

That's as far as I got though -- just a hint of a possibility....

-- 
							Greg A. Woods

+1 416 443-1734      VE3TCP      <gwoods@acm.org>      <robohack!woods>
Planix, Inc. <woods@planix.com>; Secrets of the Weird <woods@weird.com>