METHOD AND APPARATUS FOR GENERATING MULTIPLE PROCESSOR- SPECIFIC CODE SEGMENTS IN A SINGLE EXECUTABLE

• US 20080034361A1
• Filed: 10/11/2007
• Published: 02/07/2008
• Est. Priority Date: 04/13/1998
• Status: Abandoned Application

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Smith, Kevin, Abraham, Seth, Ansari, Zia
• Application Number:
  US11/870,719
• Publication Number:
  US 20080034361A1
• Time in Patent Office: Days
• US Class Current: 717/154
• CPC Class Codes: G06F 8/447 Target code generation
• CPC Class Codes: G06F 9/44547 Fat binaries

Abstract

A computer-implemented method analyzes a source code segment which is to be compiled for execution by any one of several different processor types. The method determines whether a performance advantage would be achieved by generating a customized version of object code that can be executed by one of the processor types compared with generating a non-customized version. If a performance advantage would be achieved, the method generates at least one customized object code version and a non-customized version for the source code segment, and it generates a control section that causes one of the object code versions to be called during execution of the object code in accordance with an executing processor'"'"'s processor type. If no performance advantage would be achieved, the method generates a non-customized version of the object code that can be executed by any of the different processor types.

Claims

1. A method comprising:

analyzing a source code segment that is to be compiled for execution by any one of a plurality of types of processors;

determining whether generating a customized version of object code for the source code segment to execute on a particular one of the plurality of types of processors would provide a performance advantage over generating a non-customized version of object code; and

if so, generating the customized version of object code.

2. The method of claim 1 further comprising:

generating a non-customized version of object code that can be executed by more than one type of processor.

3. The method of claim 2 further comprising:

generating a control section to examine a processor variable and to cause a version of object code to be called during execution in accordance with an executing processor'"'"'s type, and wherein the control section examines the processor variable.

4. The method of claim 1 wherein, in determining, a determination is made whether generating two or more customized versions of object code to execute on two or more types of processors would provide performance advantages over a non-customized version of object code;

and

if so, generating the two or more customized versions of object code and a non-customized version of object code.

5. The method of claim 4 further comprising:

generating a control section that causes a first customized version of object code to be called during execution if the executing processor is a first type, the control section causing a second customized version of object code to be called during execution if the executing processor is a second type, and the control section causing the non-customized version of object code to be called during execution if the executing processor is neither the first or second type.

6. The method of claim 5 wherein the first customized version of the object code is more advanced than the second customized version of object code, and wherein the control section causes the first customized version of object code to be executed if the processor type variable corresponds to a processor type that can execute either the first or second customized versions of object code.

7. A system comprising:

a bus coupling elements of the system;

a memory coupled to the bus to store instructions; and

a processor coupled to the bus, the processor to execute instructions to perform operations comprising;

analyzing a source code segment that is to be compiled for execution by any one of a plurality of types of processors;

determining whether generating a customized version of object code for the source code segment to execute on a particular one of the plurality of types of processors would provide a performance advantage over generating a non-customized version of object code; and

if so, generating the customized version of object code.

8. The system of claim 7 wherein the operations further comprise the operation of generating a non-customized version of object code that can be executed by more than one type of processor.

9. The system of claim 7 wherein the operations further comprise the operation of generating a control section to examine a processor type variable, the control section to cause a version of object code to be called during execution in accordance with an executing processor'"'"'s type.

10. A processor to execute a compiler program to perform operations comprising:

analyzing a source code segment that is to be compiled for execution by any one of a plurality of types of processors;

determining whether generating a customized version of object code for the source code segment to execute on a particular one of the plurality of types of processors would provide a performance advantage over generating a non-customized version of object code; and

if so, generating the customized version of object code.

11. The compiler program of claim 10 further comprising the operation of generating a non-customized version of object code that can be executed by more than one type of processor.

12. The compiler program of claim 11 further comprising the operation of generating a control section that causes one of the versions of object code to be called during execution in accordance with an executing processor'"'"'s type.

13. The compiler program of claim 10 wherein, in determining, a determination is made whether generating two or more customized versions of object code to execute on two or more types of processors would provide performance advantages over a non-customized version of object code;

and

if so, the compiler program generates the two or more customized versions of object code.

14. The compiler program of claim 13 wherein the compiler program further comprises the operation of generating a non-customized version of object code that can be executed by more than one type of processor.

15. The compiler program of claim 14 wherein the compiler program further comprises the operation of generating a control section that causes a first customized version of object code to be called during execution if the executing processor is a first type, the control section causing a second customized version of object code to be called during execution if the executing processor is a second type, and the control section causing the non-customized version of object code to be called during execution if the executing processor is neither the first or second type.

16. The compiler program of claim 15 wherein the control section examines a processor type variable.

17. The compiler program of claim 16 wherein the first customized version of the object code is more advanced than the second customized version of object code, and wherein the control section causes the first customized version of object code to be executed if the processor type variable corresponds to a processor type that can execute either the first or second customized versions of object code.

18. An article comprising a computer-accessible medium having associated instructions, wherein the instructions, when accessed, result in a machine performing a method comprising:

analyzing a source code segment that is to be compiled for execution by any one of a plurality of types of processors;

determining whether generating a customized version of object code for the source code segment to execute on a particular one of the plurality of types of processors would provide a performance advantage over generating a non-customized version of object code; and

if so, generating the customized version of object code.

19. The article of claim 18 wherein the method further comprises generating a non-customized version of object code.

20. The article of claim 19 wherein the non-customized version of object code can be executed by more than one type of processor.

21. The article of claim 20 wherein the method further comprises generating a control section that causes one of the versions of object code to be called during execution in accordance with an executing processor'"'"'s type.

22. The article of claim 18 wherein the determining further comprises determining whether generating two or more customized versions of object code to execute on two or more types of processors would provide performance advantages over a non-customized version of object code;

and

if so, generating the two or more customized versions of object code.

23. The article of claim 22 wherein the method further comprises generating a non-customized version of object code.

24. The article of claim 23 wherein the method further comprises:

generating a control section to examine a processor type variable and to cause a first customized version of object code to be called during execution if the executing processor is a first type, the control section causing a second customized version of object code to be called during execution if the executing processor is a second type, and the control section causing the non-customized version of object code to be called during execution if the executing processor is neither the first or second type.

25. The article of claim 24 wherein the first customized version of the object code is more advanced than the second customized version of object code, and wherein the control section causes the first customized version of object code to be executed if the processor type variable corresponds to a processor type that can execute either the first or second customized versions of object code.