Re: 80386 support

[David Laight <david%l8s.co.uk@localhost>]

On Mon, Sep 10, 2012 at 02:29:39PM -0700, Aaron J. Grier wrote:
> On Tue, Aug 28, 2012 at 09:30:39AM +0300, Jukka Ruohonen wrote:
> > It is just that i386 from the 1990s ("the infamous PC") is not
> > interesting from the perspective of ISA or design or historical
> > value...  Would you agree?
> 
> I disagree.  as the first 32-bit x86 implementation, I think the
> original 80386 was a watershed part.  tracing out how the architecture
> (both CPU and system) morphed through later designs to end up with the
> current implementations is incredibly interesting to me.  I'd like to
> think that solutions to past problems tend to be useful for the future,
> and that those who forget the past are bound to reimplement it poorly.

It is also worth rememering the bits that made life hard - and not
do them again.
IIRC the orginal plan was that the new 16bit 8086 would to be able to
run 8085 code - to avoid a lot of stuff needing to be rewritten.
This got watered down to 'source combatibility', and then to 'sort
of similar' source code.
However the damage was then done...

And if Motorola had got the 68k family working, IBM would likely have
used them for the 'PC' ....

        David

-- 
David Laight: david%l8s.co.uk@localhost