Coverity: A few billion lines of code later
Amongst my regular fiction, I'm also going over back issues of Communications of the ACM (I'm a member).
One very interesting article was A Few Billion Lines of Code Later: Using Static Analysis to Find Bugs in the Real World. It details the various ways Coverity had to adapt from being a research tool to a commercial tool.
I found particular interesting how they have to dumb down certain bug detection, so that users could understand them instead of thinking it was a false positive. Also, I liked the little table of various transforms they have to make between compilers to make sense of their code - QNX has lowest count, while Metrowerks is the worst.
Having a comment from Bjarne Stroustrup (designer of C++) is pretty cool, too!
One very interesting article was A Few Billion Lines of Code Later: Using Static Analysis to Find Bugs in the Real World. It details the various ways Coverity had to adapt from being a research tool to a commercial tool.
I found particular interesting how they have to dumb down certain bug detection, so that users could understand them instead of thinking it was a false positive. Also, I liked the little table of various transforms they have to make between compilers to make sense of their code - QNX has lowest count, while Metrowerks is the worst
Having a comment from Bjarne Stroustrup (designer of C++) is pretty cool, too!
Comments