System and method for compiling source code for multi-processor environments

• US 20050071828A1
• Filed: 09/25/2003
• Published: 03/31/2005
• Est. Priority Date: 09/25/2003
• Status: Abandoned Application

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Nutter, Mark Richard, Minor, Barry L., Brokenshire, Daniel Alan, To, VanDung Dang
• Application Number:
  US10/671,056
• Publication Number:
  US 20050071828A1
• Time in Patent Office: Days
• US Class Current: 717/147
• CPC Class Codes: G06F 8/447 Target code generation
• CPC Class Codes: G06F 9/44547 Fat binaries

Abstract

A system and method for compiling source code for multi-processor environments is presented. Source code is compiled which creates an object file whereby the object file includes multiple object code subtasks. Source code subtasks are compiled into object code subtasks using one of three approaches which are 1) a lowbrow approach, 2) a brute force approach, and 3) a program directive approach. Each object code subtask is formatted to run on a particular processor type with a particular architecture, such as a microprocessor-based architecture or a digital signal processor-based architecture. During runtime, each object code is loaded onto its corresponding processor type for execution.

Claims

1. A method for compiling source code for a plurality of heterogeneous processor types, said method comprising:

receiving source code;

selecting a processor type from the plurality of heterogeneous processor types; and

creating an object file that corresponds to the source code, wherein the object file is adapted to be processed by the selected processor type.

2. The method as described in claim 1 wherein the source code includes a plurality of source code subtasks and wherein the selecting is performed for each of the plurality of source code subtasks.

3. The method as described in claim 2 wherein the selecting is performed during compilation, the method further comprising:

retrieving one of the source code subtasks from the plurality of source code subtasks;

determining whether the source code subtask includes a program directive corresponding to one of the plurality of processors; and

performing the selecting in response to the determination.

4. The method as described in claim 2 further comprising:

retrieving one of the source code subtasks from the plurality of source code subtasks; and

compiling the retrieved source code subtask, the compiling resulting in byte code.

5. The method as described in claim 4 further comprising:

sending the byte code to a client over a computer network, wherein the byte code is adapted to be translated into client-specific object code by the client whereby the client-specific object code is formatted based upon a processor type that is located at the client.

6. The method as described in claim 2 further comprising:

retrieving one of the source code subtasks from the plurality of source code subtasks;

identifying one or more operations included in the retrieved source code subtask;

matching one or more of the operations with one of the processor types from the plurality of heterogeneous processor types; and

performing the selecting in response to the matching.

7. The method as described in claim 1 further comprising:

receiving a processor-specific command, the processor specific command identifying a processor type from the plurality of heterogeneous processor types; and

performing the selecting based upon the processor-specific command.

8. An information handling system comprising:

a plurality of heterogeneous processors;

a memory accessible by the heterogeneous processors;

one or more nonvolatile storage devices accessible by the heterogeneous processors; and

a source code compilation tool for compiling source code, the source code compilation tool comprising software code effective to;

receive source code from one of the nonvolatile storage devices;

select a processor type from a plurality of heterogeneous processor types, each of the plurality of heterogeneous processor types correspond to each of the plurality of heterogeneous processors; and

create an object file that corresponds to the source code, wherein the object file is adapted to be processed by the selected processor type.

9. The information handling system as described in claim 8 wherein the source code includes a plurality of source code subtasks and wherein the processor type selection is performed for each of the plurality of source code subtasks.

10. The information handling system as described in claim 9 wherein the processor type selection is performed during compilation, wherein the software code is further effective to:

retrieve one of the source code subtasks from the plurality of source code subtasks located in one of the nonvolatile storage devices;

determine whether the source code subtask includes a program directive corresponding to one of the plurality of processors; and

performing the selecting in response to the determination.

11. The information handling system as described in claim 9 wherein the software code is further effective to:

retrieve one of the source code subtasks from the plurality of source code subtasks; and

compile the retrieved source code subtask, the compiling resulting in byte code.

12. The information handling system as described in claim 11 wherein the software code is further effective to:

send the byte code to a client over a computer network, wherein the byte code is adapted to be translated into client-specific object code by the client whereby the client-specific object code is formatted based upon a processor type that is located at the client.

13. The information handling system as described in claim 9 wherein the software code is further effective to:

retrieve one of the source code subtasks from the plurality of source code subtasks located in one of the nonvolatile storage devices;

identify one or more operations included in the retrieved source code subtask;

match one or more of the operations with one of the processor types from the plurality of heterogeneous processor types; and

perform the selecting in response to the matching.

14. A computer program product stored on a computer operable media for compiling source code for a plurality of heterogeneous processor types, said computer program product comprising:

means for receiving source code;

means for selecting a processor type from the plurality of heterogeneous processor types; and

means for creating an object file that corresponds to the source code, wherein the object file is adapted to be processed by the selected processor type.

15. The computer program product as described in claim 14 wherein the source code includes a plurality of source code subtasks and wherein the selecting is performed for each of the plurality of source code subtasks.

16. The computer program product as described in claim 15 wherein the means for selecting is performed during compilation, the computer program product further comprising:

means for retrieving one of the source code subtasks from the plurality of source code subtasks;

means for determining whether the source code subtask includes a program directive corresponding to one of the plurality of processors; and

means for performing the selecting in response to the determination.

17. The computer program product as described in claim 15 further comprising:

means for retrieving one of the source code subtasks from the plurality of source code subtasks; and

means for compiling the retrieved source code subtask, the compiling resulting in byte code.

18. The computer program product as described in claim 17 further comprising:

means for sending the byte code to a client over a computer network, wherein the byte code is adapted to be translated into client-specific object code by the client whereby the client-specific object code is formatted based upon a processor type that is located at the client.

19. The computer program product as described in claim 15 further comprising:

means for retrieving one of the source code subtasks from the plurality of source code subtasks;

means for identifying one or more operations included in the retrieved source code subtask;

means for matching one or more of the operations with one of the processor types from the plurality of heterogeneous processor types; and

means for performing the selecting in response to the matching.

20. The computer program product as described in claim 14 further comprising:

means for receiving a processor-specific command, the processor specific command identifying a processor type from the plurality of heterogeneous processor types; and

means for performing the selecting based upon the processor-specific command.