Blog

Formidable colleagues, patchlevels and new features in Nitrogen
Pascal Cuoq on 5 December 2011

My colleagues are formidable. Over the last few days, three of them fixed four bugs in my code. And one of the fixes was a simple, obviously correct fix for the bug previously discussed here. In celebration of these terrific colleagues here is a new patchset for Nitrogen 20111001 patchlevel...

Read More

Explaining why Csmith matters even more than previously anticipated
Pascal Cuoq on 4 December 2011

Csmith as a static analyzer fuzzer A new version of Csmith, the generator of random defined C programs, was released a few days ago. This is the version that many functions in Frama-C Nitrogen were debugged against. Conversely a few bugs in the development versions of Csmith characterized by programs...

Read More

Static analysis tools comparisons
Pascal Cuoq on 25 November 2011

It is very easy to make snap judgments about other static analyzers when you are yourself working on one. You do not even need to act in bad faith: you just try the other tools on the programs that you worked hard to give interesting results on — difficult examples,...

Read More

Fixes in Nitrogen's value analysis
Pascal Cuoq on 23 November 2011

Speaking of bug fixes, here is what a proper post-release patch looks like. This patch fixes five issues identified since Nitrogen was released. I have not tested the patch itself (I have not even tried applying it to the original tarball) — this is where you come in dear reader....

Read More

Bug in Nitrogen's value analysis
Pascal Cuoq on 23 November 2011

In the course of restructuring the value analysis, I noticed a difference in some regression tests. The \after" version had some new warnings that weren't in the "before" version. After further investigation it turned out that displaying the warnings was correct and that the "before" version was unsound. This soundness...

Read More