Method to Isolate Crash of an Embedded Multi-Threaded Application to a Shared Library Call Without Core Dump Files or Debugger

• US 20100107017A1
• Filed: 10/23/2008
• Published: 04/29/2010
• Est. Priority Date: 10/23/2008
• Status: Active Grant

• Current Assignee: Dell Products LP (Dell Technologies Inc.)
• Original Assignee: Dell Products LP (Dell Technologies Inc.)
• Inventors: Muthaiyan, Arun, Munjal, Ashish
• Granted Patent: US 7,870,443 B2
• Time in Patent Office: Days
• US Class Current: 714/49
• CPC Class Codes: G06F 11/0715 in a system implementing multitasking multitasking per se G06F9/46
• CPC Class Codes: G06F 11/0766 Error or fault reporting or storing
• CPC Class Codes: G06F 11/366 using diagnostics G06F11/0703 takes precedence

Abstract

The present disclosure relates in general to embedded information handling systems, and more particularly to systems and methods for isolating a crash of an embedded multi-threaded application to a shared library call in embedded information handling systems.

Claims

1. A method for isolating errors in an embedded information handling system including a multi-threaded application having a signal handler, a shared library having at least one function, and a library encapsulation module having a function clone associated with up to each function of the shared library, the method comprising:

(a) a first calling thread of the multi-threaded application calling a first function clone associated with a first function of the shared library;

(b) the called first function clone determining a process ID associated with the first calling thread and a function ID associated with the called first function;

(c) the called first function clone storing the process ID associated with the first calling thread and a function ID associated with the called first function to a library encapsulation table protected by a mutex;

(d) the called first function clone calling the called first function;

(e) the called first function generating an error;

(f) the signal handler catching the error;

(g) the signal handler determining the function ID and process ID associated with the error using the library encapsulation table; and

(h) the signal handler writing the process ID and function ID associated with the error to a log file.

2. A method for isolating errors in an embedded information handling system according to claim 1, wherein steps (a) through (h) are executed in sequence.

3. A method for isolating errors in an embedded information handling system according to claim 2 further comprising:

(a′

) a second calling thread of the multi-threaded application calling a second function clone associated with a second function of the shared library;

(b′

) the called second function clone determining a process ID associated with the second calling thread and a function ID associated with the called second function;

(c′

) the called second function clone storing the process ID associated with the second calling thread and a function ID associated with the called second function to a library encapsulation table protected by a mutex;

(d′

) the called second function clone calling the called second function;

(d′

)(1) executing the called second function without an error; and

(d′

)(2) removing the process ID associated with the second calling thread and a function ID associated with the called second function from the library encapsulation table, wherein the removing is performed by the called function clone,

wherein (a′

) through (d′

)(2) are performed before (a) through (h).

4. A method for isolating errors in an embedded information handling system according to claim 1 further comprising:

(i) closing the multi-threaded application upon or after an error.

5. A method for isolating errors in an embedded information handling system according to claim 1, wherein the multi-threaded application is a power management multi-threaded application.

6. A method for isolating errors in an embedded information handling system according to claim 1 further comprising performing (a) through (h) without creating or using core dump files or a debugger.

7. A method for isolating errors in an embedded information handling system according to claim 1, wherein the determining the function ID associated with the called first function further comprises deriving the string field of the function using a _FUNCTION_ preprocessor macro.

8. A method for isolating errors in an embedded information handling system including a multi-threaded application having a signal handler, a shared library having at least one function, and a library encapsulation module having a function clone associated with up to each function of the shared library, the method comprising:

(a) calling the function clone associated with one of the functions of the shared library, wherein the calling is performed by a calling thread of the multi-threaded application, the calling thread includes a calling thread ID, and the one of the functions of the shared library includes a function ID;

(b) determining the calling thread ID of the calling thread and the function ID of the function using the called function clone;

(c) storing the calling thread ID of the calling thread and the function ID of the function to a library encapsulation table protected by a mutex, wherein the storing is performed by the called function clone;

(d) calling the function associated with the called function clone, wherein the calling is performed by the called function clone; and

(e)(1) executing the called function without an error, whereupon the called function clone removes the calling thread ID and the called function ID from the library encapsulation table, or

(2) executing the called function with an error, whereupon the signal handler catches the error, determines the function ID and the process ID associated with the error using the library encapsulation table, and writes the process ID and function ID associated with the error to a log file.

9. A method for isolating errors in an embedded information handling system according to claim 8, wherein steps (a) through (e) are executed in sequence.

10. A method for isolating errors in an embedded information handling system according to claim 8 further comprising:

(f) closing the multi-threaded application upon or after an error.

11. A method for isolating errors in an embedded information handling system according to claim 8, wherein the multi-threaded application is a power management multi-threaded application.

12. A method for isolating errors in an embedded information handling system according to claim 8 further comprising performing (a) through (e)(2) without creating or using core dump files or a debugger.

13. A method for isolating errors in an embedded information handling system according to claim 8, wherein the determining the function ID associated with the called first function further comprises deriving the string field of the function using a _FUNCTION_ preprocessor macro.

14. An embedded information handling system comprising:

a processor;

a memory communicatively coupled to the processor;

a local storage resource communicatively coupled to the processor;

a multi-threaded application having a signal handler executable by the processor;

a shared library communicatively coupled to the processor, the shared library having at least one function; and

a library encapsulation module having a function clone associated with up to each function of the shared library,

wherein the embedded information handling system is configured to;

(a) call the function clone associated with one of the functions of the shared library, wherein the calling is performed by a calling thread of the multi-threaded application, the calling thread includes a calling thread ID, and the one of the functions of the shared library includes a function ID;

(b) determine the calling thread ID of the calling thread and the function ID of the function using the called function clone;

(c) store the calling thread ID of the calling thread and the function ID of the function to a library encapsulation table protected by a mutex, wherein the storing is performed by the called function clone;

(d) call the function associated with the called function clone, wherein the calling is performed by the called function clone; and

(e)(1) remove from the library encapsulation table the calling thread ID and the called function ID upon error-free execution of the called function, or

(2) catch any error that occurs during execution, determine the function ID and the process ID associated with the error using the library encapsulation table, and write the process ID and function ID associated with the error to a log file.

15. An embedded information handling system according to claim 14, wherein the embedded information handling system is configured to:

(f) close the multi-threaded application upon or after an error.

16. An embedded information handling system according to claim 14, wherein the multi-threaded application is a power management multi-threaded application.

17. An embedded information handling system according to claim 14, wherein the embedded information handling system lacks a core dump file.

18. An embedded information handling system according to claim 14, wherein the embedded information handling system lacks a debugger.

19. An embedded information handling system according to claim 14, wherein the embedded information handling system lacks a core dump file and a debugger.

20. An embedded information handling system according to claim 14 further comprising a _FUNCTION_ preprocessor macro operable to derive the string field of the function associated with the called function clone.