BugMiner: Mining the Hard-to-Reach Software Vulnerabilities through the Target-Oriented Hybrid Fuzzer

Greybox Fuzzing is the most reliable and essentially powerful technique for automated software testing. Notwithstanding, a majority of greybox fuzzers are not effective in directed fuzzing, for example, towards complicated patches, as well as towards suspicious and critical sites. To overcome these limitations of greybox fuzzers, Directed Greybox Fuzzing (DGF) approaches were recently proposed. Current DGFs are powerful and efficient approaches that can compete with Coverage-Based Fuzzers. Nevertheless, DGFs neglect to accomplish stability between usefulness and proficiency, and random mutations make it hard to handle complex paths. To alleviate this problem, we propose an innovative methodology, a target-oriented hybrid fuzzing tool that utilizes a fuzzer and dynamic symbolic execution (also referred to as a concolic execution) engine. Our proposed method aims to generate inputs that can quickly reach the target sites in each sequence and trigger potential hard-to-reach vulnerabilities in the program binary. Specifically, to dive deep into the target binary, we designed a proposed technique named BugMiner, and to demonstrate the capability of our implementation, we evaluated it comprehensively on bug hunting and crash reproduction. Evaluation results showed that our proposed implementation could not only trigger hard-to-reach bugs 3.1, 4.3, 2.9, 2.0, 1.8, and 1.9 times faster than Hawkeye, AFLGo, AFL, AFLFast, QSYM, and ParmeSan respectively but also scale to several real-world programs.

MaxAFL: Maximizing Code Coverage with a Gradient-Based Optimization Technique

Evolutionary fuzzers generally work well with typical software programs because of their simple algorithm. However, there is a limitation that some paths with complex constraints cannot be tested even after long execution. Fuzzers based on concolic execution have emerged to address this issue. The concolic execution fuzzers also have limitations in scalability. Recently, the gradient-based fuzzers that use a gradient to mutate inputs have been introduced. Gradient-based fuzzers can be applied to real-world programs and achieve high code coverage. However, there is a problem that the existing gradient-based fuzzers require heavyweight analysis or sufficient learning time. In this paper, we propose a new type of gradient-based fuzzer, MaxAFL, to overcome the limitations of existing gradient-based fuzzers. Our approach constructs an objective function through fine-grained static analysis. After constructing a well-made objective function, we can apply the gradient-based optimization algorithm. We use a modified gradient-descent algorithm to minimize our objective function and propose some probabilistic techniques to escape local optimum. We introduce an adaptive objective function which aims to explore various paths in the program. We implemented MaxAFL based on the original AFL. MaxAFL achieved increase of code coverage per time compared with three other fuzzers in six open-source Linux binaries. We also measured cumulative code coverage per total execution, and MaxAFL outperformed the other fuzzers in this metric. Finally, MaxAFL can also find more bugs than the other fuzzers.

SHFuzz: Selective Hybrid Fuzzing with Branch Scheduling Based on Binary Instrumentation

Hybrid fuzzing is a popular software testing technique that combines random fuzzing with concolic execution. It is widely used in the security domain known for its ability to find deeply hidden vulnerabilities and reach high code coverage. Hybrid fuzzing is based on negating branches in the execution path of a specific input to generate new test cases. However, due to numerous inputs and related branches, it does not show the best of its effectiveness without input and branch selection methods. In this paper, we systematically analyze the branch scheduling problem in the internal attributes of hybrid fuzzing, focusing on the synchronization mechanism. To solve the problems, we propose the Selective Hybrid Fuzzing (SHF) approach with branch scheduling based on binary instrumentation. There are two major parts to the SHF approach: (1) we propose a critical branch selection algorithm to select critical branches by three metrics: hit accuracy, solvability, and complexity; (2) we propose a priority score calculation algorithm to select inputs by the number of critical branches. With the SHF approach, we choose only the branches that can be negated to generate new coverage, instead of repeatedly executing the same branches and generating duplicates of inputs. We implement a hybrid fuzzer called SHFuzz with our SHF approach and compare it with the state-of-the-art hybrid fuzzer QSYM. In the evaluation, SHFuzz outperforms QSYM in 20 real-world applications from the Google Fuzzer Test Suite and other program suites in a 12 h test. On average, SHFuzz achieves 8.40% more code coverage and 100 more unique crashes in each application. Our work also finds existing vulnerabilities 7.85× faster than QSYM. We also find new bugs by SHFuzz, which QSYM fails to find. Our evaluation shows that the selective hybrid fuzzing approach can reduce the number of branches executed in concolic execution, enhancing hybrid fuzzing on code coverage and bug finding capabilities.

DeepDiver: Diving into Abysmal Depth of the Binary for Hunting Deeply Hidden Software Vulnerabilities

Fuzz testing is a simple automated software testing approach that discovers software vulnerabilities at a high level of performance by using randomly generated seeds. However, it is restrained by coverage and thus, there are chances of finding bugs entrenched in the deep execution paths of the program. To eliminate these limitations in mutational fuzzers, patching-based fuzzers and hybrid fuzzers have been proposed as groundbreaking advancements which combine two software testing approaches. Despite those methods having demonstrated high performance across different benchmarks such as DARPA CGC programs, they still present deficiencies in their ability to analyze deeper code branches and in bypassing the roadblocks checks (magic bytes, checksums) in real-world programs. In this research, we design DeepDiver, a novel transformational hybrid fuzzing tool that explores deeply hidden software vulnerabilities. Our approach tackles limitations exhibited by existing hybrid fuzzing frameworks, by negating roadblock checks (RC) in the program. By negating the RCs, the hybrid fuzzer can explore new execution paths to trigger bugs that are hidden in the abysmal depths of the binary. We combine AFL++ and concolic execution engine and leveraged the trace analyzer approach to construct the tree for each input to detect RCs. To demonstrate the efficiency of DeepDiver, we tested it with the LAVA-M dataset and eight large real-world programs. Overall, DeepDiver outperformed existing software testing tools, including the patching-based fuzzer and state-of-the-art hybrid fuzzing techniques. On average, DeepDiver discovered vulnerabilities 32.2% and 41.6% faster than QSYM and AFLFast respectively, and it accomplished in-depth code coverage.