[rescue] NeXT OS Disks?
Patrick Giagnocavo +1.717.201.3366
patrick at zill.net
Sat May 22 11:20:09 CDT 2004
On Sat, May 22, 2004 at 10:07:58AM -0400, Andrew Weiss wrote:
> You may technically be right, but as I recall, the 88K has nothing to
> do with the PPC other than in Apple machines the logic board designed
> for the 88K had all of its glue logic and such re-used to put a 601 in
> when the machine shipped. I don't think the 88K has anything in common
> with the CPU design or even later logic board designs.
No, you're wrong, again. May I suggest Google.com? This was one of
the first hits for "88110 powerpc".
Sorry if I seem harsh, but you have made points before based on
limited or no knowledge. I don't claim to know a lot either, but at
least I double-check things with other sources before commenting.
Apple was able to re-use the glue logic precisely BECAUSE the 88K's
bus design was used for the 601. Also the 88k had cache-coherency
logic which was intended for use with 2-way or 4-way SMP; this was
also used in PPC.
The bus interface unit (BIU) converts operations in the memory queue
into transactions on the 601 bus. The 601 bus protocols originate from
the interface defined by the Motorola 88110 microprocessor chip
. There are several reasons for choosing this interface. First, it
provides a general purpose framework for system-level design
alternatives. Second, the framework is easily extendable, so that
several 601-specific features could be added without disrupting the
overall footprint of the protocols. Finally, the base technology is
readily available, meets the objectives for the 601, and has been
proven in the 88110. This was critically important for achieving the
aggressive development cycle, and it also allowed an easy migration
path for some of the early users of the part.
More information about the rescue