Debugging a high performance computing program

• US 8,813,037 B2
• Filed: 02/28/2013
• Issued: 08/19/2014
• Est. Priority Date: 02/23/2006
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Gooding, Thomas M.
• Primary Examiner(s): Puente, Emerson
• Assistant Examiner(s): Labud, Jonathan R
• Application Number:
  US13/780,215
• Publication Number:
  US 20130179866A1
• Time in Patent Office: 537 Days
• Field of Search: None
• US Class Current: 717/125
• CPC Class Codes: G06F 11/3636 by tracing the execution of the program
• CPC Class Codes: G06F 11/3664 Environments for testing or debugging software

Abstract

Methods, apparatus, and computer program products are disclosed for debugging a high performance computing program by gathering lists of addresses of calling instructions for a plurality of threads of execution of the program, assigning the threads to groups in dependence upon the addresses, and displaying the groups to identify defective threads.

Claims

1. A method for debugging a high performance computing program, the method comprising:

assigning a plurality of threads executing the program to groups in dependence upon addresses of calling instructions for the plurality of threads including iteratively assigning to a group threads having identical subsets of the lists of addresses of calling instructions; and

displaying the groups to identify defective threads including displaying the groups to identify defective threads further comprises displaying a group for an iteration and a count of the threads assigned to the group for the iteration.

2. The method of claim 1 wherein gathering lists of addresses of calling instructions for a plurality of threads of execution of the program further comprises:

gathering stack tracebacks for a plurality of threads of execution of the program, each stack traceback comprising a list of addresses of calling instructions; and

retrieving current values of program counters for a plurality of threads of execution of the program at the time of gathering the stack tracebacks.

3. The method of claim 2 further comprising inferring function names associated with the current values of the program counters.

4. The method of claim 2 further comprising inferring function names and line numbers in source code associated with the current values of the program counters.

5. The method of claim 1 wherein displaying the groups to identify defective threads further comprises inferring a function name associated with a calling instruction address located at a position in a list corresponding to an iteration.

6. An apparatus for debugging a high performance computing program, the apparatus comprising a computer processor, a computer memory operatively coupled to the computer processor, the computer memory having disposed within it computer program instructions capable of:

assigning a plurality of threads executing the program to groups in dependence upon addresses of calling instructions for the plurality of threads including iteratively assigning to a group threads having identical subsets of the lists of addresses of calling instructions; and

displaying the groups to identify defective threads including displaying the groups to identify defective threads further comprises displaying a group for an iteration and a count of the threads assigned to the group for the iteration.

7. The apparatus of claim 6 wherein gathering lists of addresses of calling instructions for a plurality of threads of execution of the program further comprises:

gathering stack tracebacks for a plurality of threads of execution of the program, each stack traceback comprising a list of addresses of calling instructions; and

retrieving current values of program counters for a plurality of threads of execution of the program at the time of gathering the stack tracebacks.

8. The apparatus of claim 7 further comprising computer program instructions capable of inferring function names associated with the current values of the program counters.

9. The apparatus of claim 6 wherein displaying the groups to identify defective threads further comprises displaying a group for an iteration and a count of the threads assigned to the group for the iteration.

10. A computer program product for debugging a high performance computing program, the computer program product disposed upon a computer readable recordable medium, wherein the computer readable recordable medium is not a signal, the computer program product comprising computer program instructions capable of:

assigning a plurality of threads executing the program to groups in dependence upon addresses of calling instructions for the plurality of threads including iteratively assigning to a group threads having identical subsets of the lists of addresses of calling instructions; and

displaying the groups to identify defective threads including displaying the groups to identify defective threads further comprises displaying a group for an iteration and a count of the threads assigned to the group for the iteration.

11. The computer program product of claim 10 wherein gathering lists of addresses of calling instructions for a plurality of threads of execution of the program further comprises:

gathering stack tracebacks for a plurality of threads of execution of the program, each stack traceback comprising a list of addresses of calling instructions; and

retrieving current values of program counters for a plurality of threads of execution of the program at the time of gathering the stack tracebacks.

12. The computer program product of claim 11 further comprising computer program instructions capable of inferring function names associated with the current values of the program counters.