Common architecture state presentation for processor having processing cores of different types

Patent metadata

Field	Value
Publication number	US-9367325-B2
Application number	US-201313931887-A
Country	US
Kind code	B2
Filing date	Jun 29, 2013
Priority date	Jun 29, 2013
Publication date	Jun 14, 2016
Grant date	Jun 14, 2016

Abstract

A method is described that includes deciding to migrate a thread from a first processing core to a second processing core. The method also includes automatically in hardware migrating first context of the thread of the first processing core whose register definition is also found on the second processing core to the second processing core. The method also includes automatically in hardware migrating second context of the thread of the first processing core whose register definition is not found on the second processing core to a first storage location external to the second processing core. The message also includes automatically in hardware migrating third context of the thread from a second storage location external to the second processing core to register definition found on the second processing core but not found on the first processing core.

First claim

What is claimed is:

1. A processor, comprising:
a first processing core having a first instruction execution pipeline and first and second register space coupled to said first instruction execution pipeline;
a second processing core having a second instruction execution pipeline and third and fourth register space coupled to said second instruction execution pipeline, wherein register definition of said first register space is the same as register definition of said third register space, register definition of said second register space is not found in register space coupled to said second instruction execution pipeline and register definition of said fourth register space is not found in register space coupled to said first instruction execution pipeline;
circuitry to support migration of a thread from said first instruction execution pipeline to said second instruction execution pipeline, said circuitry comprising:
first logic circuitry to automatically migrate, in response to a decision to perform said migration, context of said thread from said second register space to said third register space and automatically migrate context of said thread from said second register space to first storage space other than register space coupled to said second instruction execution pipeline;
second logic circuitry to automatically migrate, in response to said decision to perform said migration, context of said thread from second storage space other than register space coupled to said second instruction execution pipeline to said fourth register space.

2. The processor claim 1 wherein said first instruction execution pipeline has register renaming logic circuitry but said second instruction execution pipeline does not have register renaming logic circuitry.

3. The processor of claim 1 wherein said first instruction execution pipeline has associated accelerators for numerically intensive computations but said second instruction execution pipeline does not have associated accelerators for numerically intensive computations.

4. The processor of claim 1 wherein said first instruction execution pipeline has logic for performing soft error detection/recording or performance debug/monitoring but said second instruction execution pipeline does not have logic for performing soft error detection/recording or performance debug/monitoring.

5. The processor of claim 1 wherein said first and second instruction execution pipelines support the same instruction sets.

6. The processor of claim 1 wherein at least one of said first and second storage space other than register space coupled to said second instruction execution pipeline is found in one of the following:
directly on said first or second cores;
on said processor;
in system memory.

7. A method, comprising:
deciding to migrate a thread from a first processing core to a second processing core;
automatically in hardware migrating first context of said thread of said first processing core whose register definition is also found on said second processing core to said second processing core;
automatically in hardware migrating second context of said thread of said first processing core whose register definition is not found on said second processing core to a first storage location external to said second processing core;
automatically in hardware migrating third context of said thread from a second storage location external to said second processing core to register definition found on said second processing core but not found on said first processing core.

8. The method of claim 7 further comprising executing said thread on said second processing core.

9. The method of claim 8 further comprising, while said thread is executing on said second processing core, recognizing that said thread seeks to read from or write to said second context.

10. The method of claim 9 further comprising, in response to said recognizing, accessing from said second processing core said first storage location external to said second processing core.

11. The method of claim 7 wherein said first and second processing cores support the same instruction set.

12. The method of claim 7 where any of the following are true:
one of said processing cores implements out of order execution but the other of said processing cores does not implement out-of-order execution;
one of said processing cores implements hardware acceleration but the other of said processing cores does not implement hardware acceleration;
one of said processing cores implements branch prediction but the other of said processing cores does not implement branch prediction.

13. A computing system, comprising:
a system memory;
a processor coupled to said system memory, said processor comprising:
a first processing core having a first instruction execution pipeline and first and second register space coupled to said first instruction execution pipeline;
a second processing core having a second instruction execution pipeline and third and fourth register space coupled to said second instruction execution pipeline, wherein register definition of said first register space is the same as register definition of said third register space, register definition of said second register space is not found in register space coupled to said second instruction execution pipeline and register definition of said fourth register space is not found in register space coupled to said first instruction execution pipeline;
circuitry to support migration of a thread from said first instruction execution pipeline to said second instruction execution pipeline, said circuitry comprising:
first logic circuitry to automatically migrate, in response to a decision to perform said migration, context of said thread from said second register space to said third register space and automatically migrate context of said thread from said second register space to first storage space other than register space coupled to said second instruction execution pipeline;
second logic circuitry to automatically migrate, in response to said decision to perform said migration, context of said thread from second storage space other than register space coupled to said second instruction execution pipeline to said fourth register space.

14. The computing system claim 13 wherein said first instruction execution pipeline has logic for performing soft error detection/recording but said second instruction execution pipeline does not have logic for performing soft error detection/recording.

15. The computing system of claim 13 wherein said first instruction execution pipeline has associated accelerators for numerically intensive computations but said second instruction execution pipeline does not have associated accelerators for numerically intensive computations.

16. The computing system of claim 13 wherein said first instruction execution pipeline has logic for performing performance debug/monitoring but said second instruction execution pipeline does not have logic for performing performance debug/monitoring.

17. The computing system of claim 13 wherein said first and second instruction execution pipelines support the same instruction sets.

18. The computing system of claim 13 wherein at least one of said first and second storage space other than register space coupled to said second instruction execution pipeline is found in one of the following:
on said processor;
in said system memory.

Assignees

• Intel Corp

Inventors

• Toll Bret
• Brandt Jason W
• Holm John

Classifications

• G06F9/3012 Primary

  Organisation of register space, e.g. banked or distributed register file · CPC title

• G06F9/3867 Primary

  using instruction pipelines · CPC title

• G06F9/3851

  from multiple instruction streams, e.g. multistreaming · CPC title

• G06F9/3891

  organised in groups of units sharing resources, e.g. clusters · CPC title

• G06F9/3009 Primary

  Thread control instructions · CPC title