Subject: Re: smparc - success with mixed modules?!
To: Herb Peyerl <hpeyerl@beer.org>
From: Greg A. Woods <woods@weird.com>
List: port-sparc
Date: 01/18/2003 15:47:58
[ On Saturday, January 18, 2003 at 07:39:29 (-0700), Herb Peyerl wrote: ]
> Subject: Re: smparc - success with mixed modules?!
>
> Hubert Feyrer <hubert.feyrer@informatik.fh-regensburg.de> wrote:
> > On Sun, 19 Jan 2003, matthew green wrote:
> > > you're about the Nth person to discover this recently :-) there appears
> > > to be nothing inherent that would cause this to fail and netbsd seems to
> > > work just fine.. 40 vs 75 is the largest difference i've seen so far.
> >
> > Leet. :)
>
> I have a 90Mhz 0-cache cpu and a 150Mhz 512k cache cpu. I mentioned this
> to 'pk' and he said it wouldn't work due to different cache sizes. I
> assume that still holds as a basic truth. Barring the fact that these
> are hypersparc cpu's. It would seem I don't have any 2 cpu's with the
> same amount of cache except a dual 40 module. yuck.
There's a load of information about mixing MBus modules here:
http://mbus.sunhelp.org/misc/genconf.htm
I find all of mbus.sunhelp.org is very useful and it should probably be
linked to directly from the "other sources" list at the bottom of:
http://www.netbsd.org/Ports/sparc/faq.html
--
Greg A. Woods
+1 416 218-0098; <g.a.woods@ieee.org>; <woods@robohack.ca>
Planix, Inc. <woods@planix.com>; VE3TCP; Secrets of the Weird <woods@weird.com>