[rescue] iAPX-432, was Re: sparc10 cpu - what to do.

[Dave McGuire]

On 12/19/2016 01:28 PM, Jonathan Patschke wrote:
> I was just reading about '432 a few weeks ago, trying to decide if there's
> a smaller (more FPGA-friendly) machine inside of it to run the rest of the
> platform in microcode.  There's a lot of technical documentation out
> there, but so much of the design violates the zero-one-infinity[0] rule
> that it may really just belong in the dustbin of half-baked ideas.

  Eric Smith has been working pretty hard on reconstructing a lot of the
'432 stuff.  He says there are some important bits missing that probably
can't be recovered without de-capping a chip and reading out some
internal ROMs.  That could (and should) be done, of course, but there
are so few of the chips out there that it doesnt' seem likely to happen
unless someone with deep pockets finances the project.

> There are some really compelling ideas in there about multiple "views" (I
> forget the '432 term for them) into a given piece of address space, each
> "view" having a different set of access privileges.
> 
> My personal opinion is that most of what '432 really interesting belongs
> either in a virtual machine or operating system if only because they're
> radical enough that Intel could hardly have gotten the balance right the
> first time.

  That's what later AS/400s do.  They implement capability-based
addressing (IBM's official stance on it is that it had been removed
post-System/38, but most of it is still there) in a layer between OS/400
and the hardware.  The hardware is nearly-standard POWER, but it runs
the "LIC" (Licensed Internal Code) which it boots from a hard drive, and
on top of that sits OS/400.

  As I understand it, the iAPX-432 also implements capability-based
addressing.

  (Nobody seems to know this, but AS/400s are actually phenomenally
well-designed, well-architected, and generally fantastic machines.)

                -Dave

-- 
Dave McGuire, AK4HZ
New Kensington, PA