STAR: Stack Trace based Automatic Crash Reproduction

PhD Thesis Proposal Defence


Title: "STAR: Stack Trace based Automatic Crash Reproduction"  

by

Mr. Ning CHEN


ABSTRACT:

Software crash reproduction is a necessary first step for debugging.
Unfortunately, it is often labor intensive. To automate crash
reproduction, many capture-replay approaches have been proposed. These
approaches capture software executions and reliably reproduce the captured
executions. However, these approaches usually incur substantial
performance overhead. Alternatively, post-failure-process approaches such
as Windows Error Reporting System and Google Breakpad collect memory dumps
or stack traces after crashes occurred. Since these approaches do not
incur any additional performance overhead, they are widely used in
practice. The information collected from post-failure-process approaches
is used to prioritize debugging effort and provide debugging hints for
developers. Eventually, developers need to manually reproduce crashes
using the information, which requires non-trivial effort. In this
proposal, I propose an automatic crash reproduction framework, STAR, which
can reproduce crashes using only the crash stack trace information. The
framework can generate crash reproducible test cases without incurring
performance overhead to real world executions.


Date:                   Thursday, 13 June 2013

Time:                   10:00am - 12:00noon

Venue:                  Room 3501
                        Lifts 25/26

Committee Members:      Dr. Sunghun Kim (Supervisor)
                        Dr. Charles Zhang (Chairperson)
 			Prof. Shing-Chi Cheung
 			Dr. Raymond Wong


**** ALL are Welcome ****