A Differential Approach to Undefined Behavior Detection, TOCS 2015.
Discussion lead: Eddie Yan & Josue Rios
How is “unstable” code defined in the paper? Is it possible to identify unstable code using other methods e.g. instrumentation, exhaustive testing, and so on?
Compare the benefits and drawbacks of the undefined behavior approach vs. the deviant behavior approach. Are there any examples of bugs that would be caught by one and missed by the other?
Provide a list of questions you would like to discuss in class. Feel free to provide any comments on the paper and related topics (e.g., which parts you like and which parts you find confusing).