[rescue] SMP on intel wasteful?

[Stephen D. B. Wolthusen]

Hi,

On 24-Jun-2002 Dave McGuire wrote:
> But I'll throw you a bone, since I'm such a nice guy...first, take a
> look at IDE/ATA disks.  Don't make the argument that "but I can put
> SCSI on a PeeCee!" because then your "but it's cheaper!" argument gets
> blown even further out of the water than it already is.  

Apart from a number of inexcusable backward compatibility hacks, the I/O
architecture on x86 CPUs could've been made to work. It's just a pretty moot
argument since the relevant decisions were made by the Q and M$ in '85 or
thereabouts. And somehow I'm reasonably confident that some of the kludges will
be perpetuated in the Itanic X architectures. Oh well. 

Some way of accessing the real back end (register windowing, pipelines) and
relegating the x86 instruction set to an emulation status would also go a long
way to alleviate *that* particular ugliness and free the designers from some of
the more harebrained deadlocks between pipes, but given the way Intel has been
burned time and again on backward-compatibility benchmarks they're acting
rationally -- marketing-wise. 

> Then look at
> ROM monitor ("BIOS" for you PeeCee guys) support.  Four partitions?
> No diag capabilities?  No analog of "environment variable" support to
> pass defaults to the booting OS?  Come ON, the VAX had that in 1978.

No. The 11/780 (which I assume is what you mean) most definitely did not have a
ROM monitor. It had an LSI-11. And depending on what functions you need, it
wasn't ROM, it was an RX01/RX02. 

ROM came only with the 11/750, which was after '78 as far as I know.

-- 

        later,
        Stephen

Stephen Wolthusen (stephen at wolthusen.com)