Memory corruption bugs such as dangling pointers, double frees and uninitialized memory are some of the open issues in application security. Finding dangling pointers and similar vulnerabilities in large code bases it's arguably more difficult than overflows because of the complexity and heterogeneity of applications memory management. Fuzzing has been proved to be an effective method for finding such bugs in browsers and other similar COTS applications, nonetheless it's not uncommon to see bugs found by fuzzers burned after a short period of time because of multiple rediscovery of the same vulnerabilities. In this talk the challenges of finding such bugs with static analysis and the results we got will be discussed, specifically we will explore the algorithms and techniques borrowed from program analysis and graph theory that can be employed to achieve our goal. We will also discuss what improvements can be made in order to increase precision and reduce the number of false positives.
Secdocs is a project aimed to index high-quality IT security and hacking documents. These are fetched from multiple data sources: events, conferences and generally from interwebs.
Serving 8166 documents and 531.0 GB of hacking knowledge, indexed from 2419 authors from 163 security conferences.