Efficiency of static core turn-off in a system-on-a-chip with variation

• US 8,571,847 B2
• Filed: 03/19/2010
• Issued: 10/29/2013
• Est. Priority Date: 01/08/2010
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Cher, Chen-Yong, Coteus, Paul W., Gara, Alan, Kursun, Eren, Paulsen, David P., Schuelke, Brian A., Sheets, John E. II, Tian, Shurong
• Primary Examiner(s): Alhija, Saif
• Assistant Examiner(s): GEBRESILASSIE, KIBROM K
• Application Number:
  US12/727,984
• Publication Number:
  US 20110172984A1
• Time in Patent Office: 1,320 Days
• Field of Search: 703/13, 703/14, 718/1
• US Class Current: 703/17
• CPC Class Codes: G06F 1/206 comprising thermal management
• CPC Class Codes: G06F 1/3203 Power management, i.e. event-based initiation of a power-saving mode
• CPC Class Codes: G06F 1/3237 by disabling clock generation or distribution
• CPC Class Codes: G06F 11/24 Marginal checking or other specified testing methods not covered by G06F11/26, e.g. race tests
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A processor-implemented method for improving efficiency of a static core turn-off in a multi-core processor with variation, the method comprising: conducting via a simulation a turn-off analysis of the multi-core processor at the multi-core processor'"'"'s design stage, wherein the turn-off analysis of the multi-core processor at the multi-core processor'"'"'s design stage includes a first output corresponding to a first multi-core processor core to turn off; conducting a turn-off analysis of the multi-core processor at the multi-core processor'"'"'s testing stage, wherein the turn-off analysis of the multi-core processor at the multi-core processor'"'"'s testing stage includes a second output corresponding to a second multi-core processor core to turn off; comparing the first output and the second output to determine if the first output is referring to the same core to turn off as the second output; outputting a third output corresponding to the first multi-core processor core if the first output and the second output are both referring to the same core to turn off.

Claims

1. A processor-implemented method for improving efficiency of a core turn-off in a physical multi-core processor with variation, wherein the physical multi-core processor is based upon a multi-core processor design, the method comprising:

conducting via a simulation a turn-off analysis of the multi-core processor design at a design stage, wherein the turn-off analysis of the multi-core processor design includes a first output corresponding to a core of the multi-core processor design to turn off;

conducting a turn-off analysis of the physical multi-core processor at a testing stage, wherein the turn-off analysis of the physical multi-core processor includes a second output corresponding to a core of the physical multi-core processor to turn off;

comparing the first output and the second output to determine if the core referred to by the first output is the same core as the core referred to by the second output; and

outputting a third output corresponding to the core referred to by both the first output and the second output if the comparing indicates that the core referred to by the first output is the same core as the core referred to by the second output.

2. The method of claim 1, further comprising turning off the core of the physical multi-core processor corresponding to the third output.

3. The method of claim 1, wherein the turn-off analysis of the multi-core processor design at the design stage includes a static multi-core processor analysis conducted under an absence of variation and a first result of the static multi-core processor analysis is output and wherein an occurrence of a variation pattern is simulated on the first result and a second result is output and wherein the second result quantifies the occurrence of the variation pattern on the first result and the second result is output via a signal.

4. The method of claim 3, wherein the variation pattern is selected from a library of variation patterns and wherein the second result is output into a data structure.

5. The method of claim 4, wherein the data structure performs a function of a look-up table.

6. The method of claim 3, wherein the static multi-core processor analysis takes into consideration at least one of the following:

a thermal characteristic of the multi-core processor design, a power characteristic of the multi-core processor design and a performance characteristic of the multi-core processor design.

7. The method of claim 1, wherein the multi-core processor design has a plurality of power modes.

8. The method of claim 3, wherein the variation pattern includes data corresponding to at least one of the following:

process variation, packaging, cooling, power delivery and power distribution.

9. The method of claim 1, wherein the turn-off analysis of the physical multi-core processor at the testing stage is conducted during at least one of the following:

a wafer-level test and a module-level test; and

wherein the turn-off analysis of the physical multi-core processor includes assessing variation of each core of the physical multi-core processor by at least one of the following;

a variation tester, a ring oscillator and a temperature sensor.

10. A processor-implemented method for improving efficiency of a core turn-off in a physical multi-core processor with variation, wherein the physical multi-core processor is based upon a multi-core processor design, the method comprising:

determining via a simulation at a design stage a condition for core turn-off for a certain core of the multi-core processor design;

assessing whether the condition matches an actual variation in the certain core of the physical multi-core processor, the actual variation measured at a testing stage of the physical multi-core processor;

providing a core turn-off list based on the matching of the condition and the actual variation in the certain core; and

selecting a core of the physical multi-core processor to turn off based on the contents of the turn-off list.

11. The method of claim 10, wherein the determining step includes a static multi-core processor analysis of the multi-core processor design conducted under an absence of variation and a first result of the static multi-core processor analysis is output and wherein an occurrence of a variation pattern is simulated on the first result and a second result is output into a data structure and wherein the second result quantifies the occurrence of the variation pattern on the first result.

12. The method of claim 11, wherein the data structure performs a function of a look-up table.

13. The method of claim 12, wherein the multi-core processor design has a plurality of power modes and wherein the static multi-core processor analysis takes into consideration at least one of the following:

a thermal characteristic of the multi-core processor design, a power characteristic of the multi-core processor design and a performance characteristic of the multi-core processor design.

14. The method of claim 11, wherein the variation pattern includes data corresponding to at least one of the following:

process variation, packaging, cooling, power delivery and power distribution.

15. The method of claim 13, wherein the actual variation is measured during at least one of the following:

a wafer final test and module final test; and

wherein the actual variation is measured by at least one of the following;

a variation tester, a ring oscillator and a temperature sensor.

16. A computer system for improving efficiency of a core turn-off in a physical multi-core processor with variation, wherein the physical multi-core processor is based upon a multi-core processor design, the system comprising:

a computer memory;

a processor in communications with the computer memory, wherein the computer system is capable of performing a method comprising;

conducting via a simulation a turn-off analysis of the multi-core processor design at a design stage, wherein the turn-off analysis of the multi-core processor design includes a first output corresponding to a core of the multi-core processor design to turn off;

conducting a turn-off analysis of the physical multi-core processor at a testing stage, wherein the turn-off analysis of the physical multi-core processor includes a second output corresponding to a core of the physical multi-core processor to turn off;

comparing the first output and the second output to determine if the core referred to by the first output is the same core as the core referred to by the second output; and

outputting a third output corresponding to the core referred to by both the first output and the second output if the comparing indicates that the core referred to by the first output is the same core as the core referred to by the second output.

17. The computer system of claim 16, further comprising turning off the core of the physical multi-core processor corresponding to the third output.

18. The computer system of claim 16, wherein the turn-off analysis of the multi-core processor design at the design stage includes a static multi-core processor analysis conducted under an absence of variation and a first result of the static multi-core processor analysis is output and wherein an occurrence of a variation pattern is simulated on the first result and a second result is output and wherein the second result quantifies the occurrence of the variation pattern on the first result and the second result is output via a signal into a data structure and wherein the variation pattern is selected from a library of variation patterns.

19. The computer system of claim 18, wherein the data structure performs a function of a look-up table.

20. The computer system of claim 18, wherein the static multi-core processor analysis takes into consideration at least one of the following:

a thermal characteristic of the multi-core processor design, a power characteristic of the multi-core processor design and a performance characteristic of the multi-core processor design.

21. The computer system of claim 18, wherein the multi-core processor design has a plurality of power modes and wherein the turn-off analysis of the physical multi-core processor at the testing stage is conducted during at least one of the following:

a wafer final test and a module final test; and

wherein the turn-off analysis of the physical multi-core processor at the testing stage includes assessing variation of each core of the physical multi-core processor by at least one of the following;

a variation tester, a ring oscillator and a temperature sensor.

22. A computer program product for improving efficiency of a core turn-off in a physical multi-core processor with variation, wherein the physical multi-core processor is based upon a multi-core processor design, the computer program product comprising:

a storage device readable by a processing circuit and storing instructions for execution by the processing circuit, wherein said storage device is not propagating signal, for performing a method comprising;

conducting via a simulation a turn-off analysis of the multi-core processor design at a design stage, wherein the turn-off analysis of the multi-core processor design includes a first output corresponding to a core of the multi-core processor design to turn off;

conducting a turn-off analysis of the physical multi-core processor at a testing stage, wherein the turn-off analysis of the physical multi-core processor includes a second output corresponding to a core of the physical multi-core processor to turn off;

comparing the first output and the second output to determine if the core referred to by the first output is the same core as the core referred to by the second output; and

outputting a third output corresponding to the core referred to by both the first output and the second output if the comparing indicates that the core referred to by the first output is the same core as the core referred to by the second output.

23. The computer program product of claim 22, further comprising turning off the core of the physical multi-core processor corresponding to the third output and wherein the turn-off analysis of the multi-core processor design at the design stage includes a static multi-core processor analysis conducted under an absence of variation and a first result of the static multi-core processor analysis is output and wherein an occurrence of a variation pattern is simulated on the first result and a second result is output wherein the second result quantifies the occurrence of the variation pattern on the first result and the second result is output via a signal into a data structure and wherein the data structure performs a function of a look-up table and wherein the variation pattern is selected from a library of variation patterns.

24. The computer program product of claim 23, wherein the multi-core processor design has a plurality of power modes and wherein the static multi-core processor analysis takes into consideration at least one of the following:

a thermal characteristic of the multi-core processor design, a power characteristic of the multi-core processor design and a performance characteristic of the multi-core processor design; and

wherein the turn-off analysis of the physical multi-core processor at the testing stage is conducted during at least one of the following;

a wafer final test and a module final test; and

wherein the turn-off analysis of the physical multi-core processor at the testing stage includes assessing variation of each core of the physical multi-core processor by at least one of the following;

a variation tester, a ring oscillator and a temperature sensor.

25. A processor-implemented method for improving efficiency of a core turn-off in a physical multi-core processor with variation and a plurality of power modes, wherein the physical multi-core processor is based upon a multi-core processor design, the method comprising:

conducting via a simulation a turn-off analysis of the multi-core processor design at a design stage, wherein the turn-off analysis of the multi-core processor design includes a first output corresponding to a core of the multi-core processor design to turn off and wherein the first output is stored in a data structure performing a function of a look-up table;

conducting a turn-off analysis of the physical multi-core processor at a testing stage, wherein the turn-off analysis of the physical multi-core processor includes a second output corresponding to a core of the physical multi-core processor to turn off;

comparing the first output and the second output to determine if the core referred to by the first output is the same core as the core referred to by the second output; and

outputting a third output corresponding to the core referred to by both the first output and the second output if the comparing indicates that the core referred to by the first output is the same core as the core referred to by the second output.

26. The method of claim 25, further comprising turning off a core of the physical multi-core processor corresponding to the third output and wherein the turn-off analysis of the physical multi-core processor at the testing stage is conducted during at least one of the following:

a wafer-level test and a module-level test; and

wherein the turn-off analysis of the physical multi-core processor at the testing stage includes assessing variation of each core of the physical multi-core processor by at least one of the following;

a variation tester, a ring oscillator and a temperature sensor.

27. The method of claim 26, wherein the turn-off analysis of the multi-core processor design at the design stage includes a static multi-core processor analysis conducted under an absence of variation and a first result of the static multi-core processor analysis is output and wherein an occurrence of a variation pattern is simulated on the first result and a second result is output via a signal and wherein the second result quantifies the occurrence of the variation pattern on the first result.