Biblio

Vendors who wish to provide software or services to large corporations and governments must often obtain numerous certificates of compliance. Each certificate asserts that the software satisfies a compliance regime, like SOC or the PCI DSS, to protect the privacy and security of sensitive data. The industry standard for obtaining a compliance certificate is an auditor manually auditing source code. This approach is expensive, error-prone, partial, and prone to regressions. We propose continuous compliance to guarantee that the codebase stays compliant on each code change using lightweight verification tools. Continuous compliance increases assurance and reduces costs. Continuous compliance is applicable to any source-code compliance requirement. To illustrate our approach, we built verification tools for five common audit controls related to data security: cryptographically unsafe algorithms must not be used, keys must be at least 256 bits long, credentials must not be hard-coded into program text, HTTPS must always be used instead of HTTP, and cloud data stores must not be world-readable. We evaluated our approach in three ways. (1) We applied our tools to over 5 million lines of open-source software. (2) We compared our tools to other publicly-available tools for detecting misuses of encryption on a previously-published benchmark, finding that only ours are suitable for continuous compliance. (3) We deployed a continuous compliance process at AWS, a large cloud-services company: we integrated verification tools into the compliance process (including auditors accepting their output as evidence) and ran them on over 68 million lines of code. Our tools and the data for the former two evaluations are publicly available.

Modern JavaScript applications extensively depend on third-party libraries. Especially for the Node.js platform, vulnerabilities can have severe consequences to the security of applications, resulting in, e.g., cross-site scripting and command injection attacks. Existing static analysis tools that have been developed to automatically detect such issues are either too coarse-grained, looking only at package dependency structure while ignoring dataflow, or rely on manually written taint specifications for the most popular libraries to ensure analysis scalability. In this work, we propose a technique for automatically extracting taint specifications for JavaScript libraries, based on a dynamic analysis that leverages the existing test suites of the libraries and their available clients in the npm repository. Due to the dynamic nature of JavaScript, mapping observations from dynamic analysis to taint specifications that fit into a static analysis is non-trivial. Our main insight is that this challenge can be addressed by a combination of an access path mechanism that identifies entry and exit points, and the use of membranes around the libraries of interest. We show that our approach is effective at inferring useful taint specifications at scale. Our prototype tool automatically extracts 146 additional taint sinks and 7 840 propagation summaries spanning 1 393 npm modules. By integrating the extracted specifications into a commercial, state-of-the-art static analysis, 136 new alerts are produced, many of which correspond to likely security vulnerabilities. Moreover, many important specifications that were originally manually written are among the ones that our tool can now extract automatically.

Language-based information-flow control (IFC) techniques often rely on special purpose, ad-hoc primitives to address different covert channels that originate in the runtime system, beyond the scope of language constructs. Since these piecemeal solutions may not compose securely, there is a need for a unified mechanism to control covert channels. As a first step towards this goal, we argue for the design of a general interface that allows programs to safely interact with the runtime system and the available computing resources. To coordinate the communication between programs and the runtime system, we propose the use of asynchronous exceptions (interrupts), which, to the best of our knowledge, have not been considered before in the context of IFC languages. Since asynchronous exceptions can be raised at any point during execution-often due to the occurrence of an external event-threads must temporarily mask them out when manipulating locks and shared data structures to avoid deadlocks and, therefore, breaking program invariants. Crucially, the naive combination of asynchronous exceptions with existing features of IFC languages (e.g., concurrency and synchronization variables) may open up new possibilities of information leakage. In this paper, we present MACasync, a concurrent, statically enforced IFC language that, as a novelty, features asynchronous exceptions. We show how asynchronous exceptions easily enable (out of the box) useful programming patterns like speculative execution and some degree of resource management. We prove that programs in MACasync satisfy progress-sensitive non-interference and mechanize our formal claims in the Agda proof assistant.

A wide range of tools exist to assist developers in creating secure software. Many of these tools, such as static analysis engines or security checkers included in compilers, use warnings to communicate security issues to developers. The effectiveness of these tools relies on developers heeding these warnings, and there are many ways in which these warnings could be displayed. Johnson et al. [46] conducted qualitative research and found that warning presentation and integration are main issues. We built on Johnson et al.'s work and examined what developers want from security warnings, including what form they should take and how they should integrate into their workflow and work context. To this end, we conducted a Grounded Theory study with 14 professional software developers and 12 computer science students as well as a focus group with 7 academic researchers to gather qualitative insights. To back up the theory developed from the qualitative research, we ran a quantitative survey with 50 professional software developers. Our results show that there is significant heterogeneity amongst developers and that no one warning type is preferred over all others. The context in which the warnings are shown is also highly relevant, indicating that it is likely to be beneficial if IDEs and other development tools become more flexible in their warning interactions with developers. Based on our findings, we provide concrete recommendations for both future research as well as how IDEs and other security tools can improve their interaction with developers.

This paper presents BackFlow, a compiler-based toolchain that enforces indirect backward edge control flow integrity for low-end ARM Cortex-M microprocessors. BackFlow is implemented within the Clang/LLVM compiler and supports the ARM instruction set and its subset Thumb. The control flow integrity generated by the compiler relies on a bitmap, where each set bit indicates a valid pointer destination. The efficiency of the framework is benchmarked using an STM32 NUCLEO F446RE microcontroller. The obtained results show that the control flow integrity solution incurs an execution time overhead ranging from 1.5 to 4.5%.

Ethereum, one of the most popular blockchain platforms, provides financial transactions like payments and auctions through smart contracts. Due to the immense interest in smart contracts in academia, the research community of smart contract security has made a significant improvement recently. Researchers have reported various security vulnerabilities in smart contracts, and developed static analysis tools and verification frameworks to detect them. However, it is unclear whether such great efforts from academia has indeed enhanced the security of smart contracts in reality. To understand the security level of smart contracts in the wild, we empirically studied 55,046 real-world Ethereum smart contracts written in Solidity, the most popular programming language used by Ethereum smart contract developers. We first examined how many well-known vulnerabilities the Solidity compiler has patched, and how frequently the Solidity team publishes compiler releases. Unfortunately, we observed that many known vulnerabilities are not yet patched, and some patches are not even sufficient to avoid their target vulnerabilities. Subsequently, we investigated whether smart contract developers use the most recent compiler with vulnerabilities patched. We reported that developers of more than 98% of real-world Solidity contracts still use older compilers without vulnerability patches, and more than 25% of the contracts are potentially vulnerable due to the missing security patches. To understand actual impacts of the missing patches, we manually investigated potentially vulnerable contracts that are detected by our static analyzer and identified common mistakes by Solidity developers, which may cause serious security issues such as financial loss. We detected hundreds of vulnerable contracts and about one fourth of the vulnerable contracts are used by thousands of people. We recommend the Solidity team to make patches that resolve known vulnerabilities correctly, and developers to use the latest Solidity compiler to avoid missing security patches.

Deep Learning methods have not only shown to improve software performance in compiler heuristics, but also e.g. to improve security in vulnerability prediction or to boost developer productivity in software engineering tools. A key to the success of such methods across these use cases is the expressiveness of the representation used to abstract from the program code. Recent work has shown that different such representations have unique advantages in terms of performance. However, determining the best-performing one for a given task is often not obvious and requires empirical evaluation. Therefore, we present ComPy-Learn, a toolbox for conveniently defining, extracting, and exploring representations of program code. With syntax-level language information from the Clang compiler frontend and low-level information from the LLVM compiler backend, the tool supports the construction of linear and graph representations and enables an efficient search for the best-performing representation and model for tasks on program code.

Android, being the most widespread mobile operating systems is increasingly becoming a target for malware. Malicious apps designed to turn mobile devices into bots that may form part of a larger botnet have become quite common, thus posing a serious threat. This calls for more effective methods to detect botnets on the Android platform. Hence, in this paper, we present a deep learning approach for Android botnet detection based on Convolutional Neural Networks (CNN). Our proposed botnet detection system is implemented as a CNN-based model that is trained on 342 static app features to distinguish between botnet apps and normal apps. The trained botnet detection model was evaluated on a set of 6,802 real applications containing 1,929 botnets from the publicly available ISCX botnet dataset. The results show that our CNN-based approach had the highest overall prediction accuracy compared to other popular machine learning classifiers. Furthermore, the performance results observed from our model were better than those reported in previous studies on machine learning based Android botnet detection.

Malware scanning of an app market is expected to be scalable and effective. However, existing approaches use either syntax-based features which can be evaded by transformation attacks or semantic-based features which are usually extracted by performing expensive program analysis. Therefor, in this paper, we propose a lightweight graph-based approach to perform Android malware detection. Instead of traditional heavyweight static analysis, we treat function call graphs of apps as social networks and perform social-network-based centrality analysis to represent the semantic features of the graphs. Our key insight is that centrality provides a succinct and fault-tolerant representation of graph semantics, especially for graphs with certain amount of inaccurate information (e.g., inaccurate call graphs). We implement a prototype system, MalScan, and evaluate it on datasets of 15,285 benign samples and 15,430 malicious samples. Experimental results show that MalScan is capable of detecting Android malware with up to 98% accuracy under one second which is more than 100 times faster than two state-of-the-art approaches, namely MaMaDroid and Drebin. We also demonstrate the feasibility of MalScan on market-wide malware scanning by performing a statistical study on over 3 million apps. Finally, in a corpus of dataset collected from Google-Play app market, MalScan is able to identify 18 zero-day malware including malware samples that can evade detection of existing tools.

Malicious scripts are an important computer infection threat vector for computer users. For internet-scale processing, static analysis offers substantial computing efficiencies. We propose the ScriptNet system for neural malicious JavaScript detection which is based on static analysis. We also propose a novel deep learning model, Pre-Informant Learning (PIL), which processes Javascript files as byte sequences. Lower layers capture the sequential nature of these byte sequences while higher layers classify the resulting embedding as malicious or benign. Unlike previously proposed solutions, our model variants are trained in an end-to-end fashion allowing discriminative training even for the sequential processing layers. Evaluating this model on a large corpus of 212,408 JavaScript files indicates that the best performing PIL model offers a 98.10% true positive rate (TPR) for the first 60K byte subsequences and 81.66% for the full-length files, at a false positive rate (FPR) of 0.50%. Both models significantly outperform several baseline models. The best performing PIL model can successfully detect 92.02% of unknown malware samples in a hindsight experiment where the true labels of the malicious JavaScript files were not known when the model was trained.

Android operating systems have become a prime target for attackers as most of the market is currently dominated by Android users. The situation gets worse when users unknowingly download or sideload cloning applications, especially gaming applications that look like benign games. In this paper, we present, a dynamic Android gaming malware detection system based on system call analysis to classify malicious and legitimate games. We performed the dynamic system call analysis on normal and malicious gaming applications while applications are in execution state. Our analysis reveals the similarities and differences between benign and malware game system calls and shows how dynamically analyzing the behavior of malicious activity through system calls during runtime makes it easier and is more effective to detect malicious applications. Experimental analysis and results shows the efficiency and effectiveness of our approach.

In transient distributed cloud computing environment, software is vulnerable to attack, which leads to software functional completeness, so it is necessary to carry out functional testing. In order to solve the problem of high overhead and high complexity of unsupervised test methods, an intelligent evaluation method for transient analysis software function testing based on active depth learning algorithm is proposed. Firstly, the active deep learning mathematical model of transient analysis software function test is constructed by using association rule mining method, and the correlation dimension characteristics of software function failure are analyzed. Then the reliability of the software is measured by the spectral density distribution method of software functional completeness. The intelligent evaluation model of transient analysis software function testing is established in the transient distributed cloud computing environment, and the function testing and reliability intelligent evaluation are realized. Finally, the performance of the transient analysis software is verified by the simulation experiment. The results show that the accuracy of the software functional integrity positioning is high and the intelligent evaluation of the transient analysis software function testing has a good self-adaptability by using this method to carry out the function test of the transient analysis software. It ensures the safe and reliable operation of the software.

The Rowhammer bug is a novel micro-architectural security threat, enabling powerful privilege-escalation attacks on various mainstream platforms. It works by actively flipping bits in Dynamic Random Access Memory (DRAM) cells with unprivileged instructions. In order to set up Rowhammer against binaries in the Linux page cache, the Waylaying algorithm has previously been proposed. The Waylaying method stealthily relocates binaries onto exploitable physical addresses without exhausting system memory. However, the proof-of-concept Waylaying algorithm can be easily detected during page cache eviction because of its high disk I/O overhead and long running time. This paper proposes the more advanced Memway algorithm, which improves on Waylaying in terms of both I/O overhead and speed. Running time and disk I/O overhead are reduced by 90% by utilizing Linux tmpfs and inmemory swapping to manage eviction files. Furthermore, by combining Memway with the unprivileged posix fadvise API, the binary relocation step is made 100 times faster. Equipped with our Memway+fadvise relocation scheme, we demonstrate practical Rowhammer attacks that take only 15-200 minutes to covertly relocate a victim binary, and less than 3 seconds to flip the target instruction bit.

Operating systems such as Linux break the power of the root user into separate privileges (which Linux calls capabilities) and give processes the ability to enable privileges only when needed and to discard them permanently when the program no longer needs them. However, there is no method of measuring how well the use of such facilities reduces the risk of privilege escalation attacks if the program has a vulnerability. This paper presents PrivAnalyzer, an automated tool that measures how effectively programs use Linux privileges. PrivAnalyzer consists of three components: 1) AutoPriv, an existing LLVM-based C/C++ compiler which uses static analysis to transform a program that uses Linux privileges into a program that safely removes them when no longer needed, 2) ChronoPriv, a new LLVM C/C++ compiler pass that performs dynamic analysis to determine for how long a program retains various privileges, and 3) ROSA, a new bounded model checker that can model the damage a program can do at each program point if an attacker can exploit the program and abuse its privileges. We use PrivAnalyzer to determine how long five privileged open source programs retain the ability to cause serious damage to a system and find that merely transforming a program to drop privileges does not significantly improve security. However, we find that simple refactoring can considerably increase the efficacy of Linux privileges. In two programs that we refactored, we reduced the percentage of execution in which a device file can be read and written from 97% and 88% to 4% and 1%, respectively.

Concurrency vulnerabilities are extremely harmful and can be frequently exploited to launch severe attacks. Due to the non-determinism of multithreaded executions, it is very difficult to detect them. Recently, data race detectors and techniques based on maximal casual model have been applied to detect concurrency vulnerabilities. However, the former are ineffective and the latter report many false negatives. In this paper, we present CONVUL, an effective tool for concurrency vulnerability detection. CONVUL is based on exchangeable events, and adopts novel algorithms to detect three major kinds of concurrency vulnerabilities. In our experiments, CONVUL detected 9 of 10 known vulnerabilities, while other tools only detected at most 2 out of these 10 vulnerabilities. The 10 vulnerabilities are available at https://github.com/mryancai/ConVul.

Localizing concurrency faults that occur in production is hard because, (1) detailed field data, such as user input, file content and interleaving schedule, may not be available to developers to reproduce the failure; (2) it is often impractical to assume the availability of multiple failing executions to localize the faults using existing techniques; (3) it is challenging to search for buggy locations in an application given limited runtime data; and, (4) concurrency failures at the system level often involve multiple processes or event handlers (e.g., software signals), which can not be handled by existing tools for diagnosing intra-process(thread-level) failures. To address these problems, we present SCMiner, a practical online bug diagnosis tool to help developers understand how a system-level concurrency fault happens based on the logs collected by the default system audit tools. SCMiner achieves online bug diagnosis to obviate the need for offline bug reproduction. SCMiner does not require code instrumentation on the production system or rely on the assumption of the availability of multiple failing executions. Specifically, after the system call traces are collected, SCMiner uses data mining and statistical anomaly detection techniques to identify the failure-inducing system call sequences. It then maps each abnormal sequence to specific application functions. We have conducted an empirical study on 19 real-world benchmarks. The results show that SCMiner is both effective and efficient at localizing system-level concurrency faults.

A key question for characterising a system's vulnerability against timing attacks is whether or not it allows an adversary to aggregate information about a secret over multiple timing measurements. Existing approaches for reasoning about this aggregate information rely on strong assumptions about the capabilities of the adversary in terms of measurement and computation, which is why they fall short in modelling, explaining, or synthesising real-world attacks against cryptosystems such as RSA or AES. In this paper we present a novel model for reasoning about information aggregation in timing attacks. The model is based on a novel abstraction of timing measurements that better captures the capabilities of real-world adversaries, and a notion of compositionality of programs that explains attacks by divide-and-conquer. Our model thus lifts important limiting assumptions made in prior work and enables us to give the first uniform explanation of high-profile timing attacks in the language of information-flow analysis.

Many web applications are vulnerable to Cross Site Scripting (XSS) attacks enabling attackers to steal sensitive information and commit frauds. Much research in this area have focused on detecting vulnerable web pages using static and dynamic program analysis. The best practice to prevent XSS vulnerabilities is to encode untrusted dynamic content. However, a common programming error is the use of a wrong type of encoder to sanitize untrusted data, leaving the application vulnerable. We propose a new approach that can automatically fix this common type of XSS vulnerability in many situations. This approach is integrated into the software maintenance life cycle through unit testing. Vulnerable codes are refactored to reflect the suggested encoder and then verified using an attack evaluating mechanism to find a proper repair. Evaluation of this approach has been conducted on an open source medical record application with over 200 web pages written in JSP.

Cyber-physical systems (CPS) consist of sensors, actuators, and controllers all communicating over a network; if any subset becomes compromised, an attacker could cause significant damage. With access to data logs and a model of the CPS, the physical effects of an attack could potentially be detected before any damage is done. Manually building a model that is accurate enough in practice, however, is extremely difficult. In this paper, we propose a novel approach for constructing models of CPS automatically, by applying supervised machine learning to data traces obtained after systematically seeding their software components with faults ("mutants"). We demonstrate the efficacy of this approach on the simulator of a real-world water purification plant, presenting a framework that automatically generates mutants, collects data traces, and learns an SVM-based model. Using cross-validation and statistical model checking, we show that the learnt model characterises an invariant physical property of the system. Furthermore, we demonstrate the usefulness of the invariant by subjecting the system to 55 network and code-modification attacks, and showing that it can detect 85% of them from the data logs generated at runtime.

To detect specific types of bugs and vulnerabilities, static analysis tools must be correctly configured with security-relevant methods (SRM), e.g., sources, sinks, sanitizers and authentication methods-usually a very labour-intensive and error-prone process. This work presents the semi-automated tool SWAN\_ASSIST, which aids the configuration with an IntelliJ plugin based on active machine learning. It integrates our novel automated machine-learning approach SWAN, which identifies and classifies Java SRM. SWAN\_ASSIST further integrates user feedback through iterative learning. SWAN\_ASSIST aids developers by asking them to classify at each point in time exactly those methods whose classification best impact the classification result. Our experiments show that SWAN\_ASSIST classifies SRM with a high precision, and requires a relatively low effort from the user. A video demo of SWAN\_ASSIST can be found at https://youtu.be/fSyD3V6EQOY. The source code is available at https://github.com/secure-software-engineering/swan.

Application Programming Interfaces (APIs) usually have usage constraints, such as call conditions or call orders. Incorrect usage of these constraints, called API misuse, will result in system crashes, bugs, and even security problems. It is crucial to detect such misuses early in the development process. Though many approaches have been proposed over the last years, recent studies show that API misuses are still prevalent, especially the ones specific to individual projects. In this paper, we strive to improve current API-misuse detection capability for large-scale C programs. First, We propose IMSpec, a lightweight domain-specific language enabling developers to specify API usage constraints in three different aspects (i.e., parameter validation, error handling, and causal calling), which are the majority of API-misuse bugs. Then, we have tailored a constraint guided static analysis engine to automatically parse IMSpec rules and detect API-misuse bugs with rich semantics. We evaluate our approach on widely used benchmarks and real-world projects. The results show that our easily extensible approach performs better than state-of-the-art tools. We also discover 19 previously unknown bugs in real-world open-source projects, all of which have been confirmed by the corresponding developers.

Today, large and complex software is developed with integrated components using application programming interfaces (APIs). Correct usage of APIs in practice presents a challenge due to implicit constraints, such as call conditions or call orders. API misuse, i.e., violation of these constraints, is a well-known source of bugs, some of which can cause serious security vulnerabilities. Although researchers have developed many API-misuse detectors over the last two decades, recent studies show that API misuses are still prevalent. In this paper, we provide a comprehensive empirical study on API-misuse bugs in open-source C programs. To understand the nature of API misuses in practice, we analyze 830 API-misuse bugs from six popular programs across different domains. For all the studied bugs, we summarize their root causes, fix patterns and usage statistics. Furthermore, to understand the capabilities and limitations of state-of-the-art static analysis detectors for API-misuse detection, we develop APIMU4C, a dataset of API-misuse bugs in C code based on our empirical study results, and evaluate three widely-used detectors on it qualitatively and quantitatively. We share all the findings and present possible directions towards more powerful API-misuse detectors.

With the popularity of smart devices such as mobile phones and tablets, the security problem of the widely used ARMv8-A processor has received more and more attention. Flush+Reload and Flush+Flush cache attacks have become two of the most important security threats due to their low noise and high resolution. In order to resist Flush+Reload and Flush+Flush attacks, researchers proposed many defense methods. However, these existing methods have various shortcomings. The runtime defense methods using hardware performance counters cannot detect attacks fast enough, effectively detect Flush+Flush or avoid a high false positive rate. Static code analysis schemes are powerless for obfuscation techniques. The approaches of permanently reducing the resolution can only be utilized on browser products and cannot be applied in the system. In this paper, we design two more secure collaborative APIs-flush operation API and high resolution time API-which can resist Flush+Reload and Flush+Flush attacks. When the flush operation API is called, the high resolution time API temporarily reduces its resolution and automatically restores. Moreover, the flush operation API also has the ability to detect and handle suspected Flush+Reload and Flush+Flush attacks. The attack and performance comparison experiments prove that the two APIs we designed are safer and the performance losses are acceptable.

Recently, damage caused by ransomware has been increasing in PC and Android environments. There are many studies into real-time ransomware detection because the most important time to prevent encryption is before ransomware is able to execute its malicious process. Traditional analyses determine an application is ransomware or not by static/dynamic methods. Those analyses can serve as components of a method to detect ransomware in real time. However, problems can occur such as the inability to detect new/variant/unknown ransomware. These types require signed patches from a trusted party that can only be created after attacks occur. In a previous study into realtime new/variant/unknown ransomware detection in a PC environment, important files are monitored and only programs that have been previously analyzed and evaluated as nonmalicious are allowed. As such, programs that have not been analyzed are restricted from accessing important files. In an Android environment, this method can be applied using Android applications to prevent emerging threats and verify consistency with user intent. Thus, this paper proposes a method of detecting new/variant/unknown ransomware in real time in an Android environment.

The similarity detection of the program has important significance in code reuse, plagiarism detection, intellectual property protection and information retrieval methods. Attribute counting methods cannot take into account program semantics. The method based on syntax tree or graph structure has a very high construction cost and low space efficiency. So it is difficult to solve problems in large-scale software systems. This paper uses different decision strategies for different levels, then puts forward a similarity detection method at the file level. This method can make full use of the features of the program and take into account the space-time efficiency. By using static analysis methods, we get function features and control flow features of files. And based on this, we establish the function call graph. The similar degree between two files can be measured with the two graphs. Experimental results show the method can effectively detect similar files. Finally, this paper discusses the direction of development of this method.