GCC correctness fuzzing
In 2008 I wrote about generating random JavaScript to find differences between optimization modes and differences between JavaScript engines (rough list of bugs).
How do you do this kind of testing on a language like C where the behavior of many programs is undefined per spec? John Regehr explains how in his talk Exposing Difficult Compilers Bugs With Random Testing at GCC Summit 2010.
November 11th, 2010 at 8:12 am
Hi Jesse,
You should really update your bookmarklets, as they’re almost 7 years old. Many things changed in that time. I recommend you to check your bookmarklets with JSLint (http://jslint.com) and make sure there is no error. I think it’s enough to use the default options, but if you check “the good partsâ€, it will be even better.