VUzzer: Application-aware Evolutionary Fuzzing.

摘要: See, stats, and : https / www . researchgate net publication 311886374 VUzzer Application - aware Conference DOI 10 14722 ndss 2017 23404 CITATIONS 0 READS 17 6 , including Some Systems Sanjay Vrije Amsterdam Netherlands 38 SEE Ashish International 1 Cristiano VU 51 Herbert 163 836 All The text letting Abstract—Fuzzing is an effective software testing technique to find bugs Given the size complexity of real world applications modern fuzzers tend be either scalable but not in exploring that lie deeper execution or capable penetrating application In this paper we present evolutionary fuzzing strategy does require any prior knowledge input format order maximize coverage explore paths leverage control data flow features based on static dynamic analysis infer fundamental prop erties This enables much faster generation interesting inputs compared agnostic approach We implement our evaluate it three different datasets DARPA Grand Challenge binaries (CGC) a set (binary parsers) recently released LAVA dataset On all these yields significantly better results than state art by quickly finding several existing new