The FreeBSD Project uses the Synopsys Coverity Scan service for static analysis of FreeBSD source code to identity security vulnerabilities and other code quality issues.
Note 1: At this time, we only provide access to view scan results to FreeBSD developers.
Note 2: If you are a FreeBSD developer, please register using your FreeBSD.org email address
PR: <issue-id> in commit log messagesCoverity ID: <CIDXXXXX> in commit log messageStatic code analysis has some limitations in its ability to understand certain dynamic operations. This limitation may result in falsely detecting defects. Since most false-positive defects are caused by few functions in your code base, Coverity allows you to tell the analysis engine to treat these functions differently. This is called a Modeling File. By providing a modeling file, most projects reduce their false-positive rate to the ballpark of 10%.
https://scan.coverity.com/Coverity Modeling Reference (This link is broken)It seems that the TAILQ macros confuse Coverity. For example TAILQ_FIRST() seems to trigger a use after free error in <list>.tqh_first when entries are removed in a loop (loop by hand or with FOREACH). In particular, it does not recognize that TAILQ_REMOVE puts a new pointer in place and the pointer which previously aliases the tqh_first is not aliasing anymore, but is the only valid pointer.
Such cases should be marked as XXX (at the moment marked as IGNORE until we determine if it should be marked as a FALSE or if this should stay as IGNORE until a custom model is developed for this).
TODO: Model for TAILQ macros added to Modelling File