IBM Patent: x86-->PowerPC; Architecture A--> B?

[j^aws]

Okay this is an old patent filed by IBM, but it's assigned to a John Keaty, a senior member of the CELL design team and part of the ISSCC paper on CELL,

10.2 The Design and Implementation of a First-Generation CELL Processor
9:00 AM
D. Pham(1), S.Asano(2),M. Bolliger(1), M. Day(1) , H. Hofstee(1), C. Johns(1) , J. Kahle(1) , A. Kameyama(3) , J. Keaty(1),Y. Masubuchi(2), M. Riley(1), D. Shippy(1), D. Stasiak(1) , M.Wang(1) , J.Warnock(1), S.Weitzel(1), D.Wendel(1) , T.Yamazaki(1) , K.Yazawa(2)

1 - IBM, Austin, TX
2 - Sony, Tokyo, Japan
3 - Toshiba, Austin, TX

A CELL Processor is a multi-core chip consisting of a 64b Power architecture processor, multiple streaming processors, a flexible IO interface, and a memory interface controller. This SoC is implemented in 90nm SOI technology. The chip is designed with a high degree of modularity and reuse to maximize the custom circuit content and achieve a high-frequency clock-rate.

http://www.beyond3d.com/forum/viewtopic.php?p=421487#421487

This very old patent below,

Computer system supporting control transfers between two architectures

is the ONLY patent assigned to John Keaty.

Computer system supporting control transfers between two architectures

Abstract

A computer system supports control transfers between two architectures with different address ranges and/or different methods of calling programs which involve passing parameters, stack pointers and return addresses. Control is transferred between the architectures through a control transfer mode bit.
...

SUMMARY OF THE INVENTION

It is thus an advantage of the present invention to provide a computer system where program control transfers take place between processor architectures with different address ranges and linkage methods (i.e. the methods of passing parameters, stack pointers and return addresses).

It is another advantage of the invention to provide a computer system which does not require use of the process stack in memory for transferring parameters and process return information.

It is yet another advantage of the invention to provide backward compatibility with the x86 legacy applications or any other similar applications used in transfer controls.

The foregoing and other objects of the inventions are realized by a microprocessor which transfers program control between multiple architectures, for example RISC and x86 CISC, by using a bit in a descriptor table entry to describe whether or not the target return segment contains software from the first architecture or the second architecture. Registers used for the linkage methods int he first architecture, but not previously defined by the second architecture, are used by the second architecture to hold excess work bits, allowing for architectures of different word lengths. Furthermore, the stack pointer and return address of the calling program is left in registers rather than in memory, improving the performance of the control transfer.
...

I know this patent is very old but as the assignee is a senior member of the Cell team and even though it may not be connected to Cell, there are interesting possibiities at least! How about a x86 compatible PowerPC chip!

 

[hey69]

ps3 backwards compatible with ps2 ps1 AND xbox..

just imagine they already have nvidia in the pocket!

 

[j^aws]

ps3 backwards compatible with ps2 ps1 AND xbox..

just imagine they already have nvidia in the pocket!

First thing that crossed my mind!

 

[AlNom]

ps3 backwards compatible with ps2 ps1 AND xbox..

just imagine they already have nvidia in the pocket!

Except they'd have to pay eleventy billion dollars to Microsoft in licensing that ability/claim. Maybe xbox will be profitable after all.

 

[Acert93]

If Sony's PS3 is backwards compatible with the Xbox I think we will see Sony and nVidia go out of business

Now x86 comatibility would be useful for PC and Xenon ports though...

 

[AzBat]

We've brought this up before...

http://www.beyond3d.com/forum/viewtopic.php?p=388626

Maybe the latter patent has to do with the PowerPC 615?

Tommy McClain

 

[j^aws]

We've brought this up before...

http://www.beyond3d.com/forum/viewtopic.php?p=388626

Maybe the latter patent has to do with the PowerPC 615?

Tommy McClain

You could be right there. This patent does smell of the phantom PPC 615 with hardware x86 compatability.

Here's an old article with an IBM engineer claiming MS killed the PPC 615, prolly seeing it as a threat to its OS...

http://www.theregister.co.uk/1998/10/01/microsoft_killed_the_powerpc/

and an older Byte article on the PPC 615,

http://byte.com/art/9406/sec6/art6.htm

It will be interesting if this tech makes it's way into CELL and the rumoured PPC 615 successor (AMD involvement here?)...If this is realised then surely all IBM guns (STI) will be pointing at Wintel but more specifically Intel. Especially considering it's recent PC division sell-off, it's POWER everywhere initiative and CELL involvement. But this is nothing new as IBM always seems to get twitchy trigger fingers every 10 years!