Self-modifying code (SMC) allows programs to alter their own instructions, optimizing performance and functionality on x86 processors. Despite its benefits, SMC introduces unique microarchitectural behaviors that can be exploited for malicious purposes. In this paper, we explore the security implications of SMC by examining how specific x86 instructions affecting instruction cache lines lead to measurable timing discrepancies between cache hits and misses. These discrepancies facilitate refined cache attacks, making them less noisy and more effective. We introduce novel attack techniques that leverage these timing variations to enhance existing methods such as Prime+Probe and Flush+Reload. Our advanced techniques allow adversaries to more precisely attack cryptographic keys and create covert channels akin to Spectre across various x86 platforms. Finally, we propose a dynamic detection methodology utilizing hardware performance counters to mitigate these enhanced threats.

The cloud computing landscape has evolved sig-nificantly in recent years, embracing various sandboxes to meet the diverse demands of modern cloud applications. These sandboxes encompass container-based technologies like Docker and gVisor, microVM-based solutions like Fire-cracker, and security-centric sandboxes relying on Trusted Execution Environments (TEEs) such as Intel SGX and AMD SEV. However, the practice of placing multiple tenants on shared physical hardware raises security and privacy concerns, most notably side-channel attacks. In this paper, we investigate the possibility of fingerprinting containers through CPU frequency reporting sensors in Intel and AMD CPUs. One key enabler of our attack is that the current CPU frequency information can be accessed by user-space attackers. We demonstrate that Docker images exhibit a unique frequency signature, enabling the distinction of different containers with up to 84.5 % accuracy even when multiple containers are running simultaneously in different cores. Additionally, we assess the effectiveness of our attack when performed against several sandboxes deployed in cloud environments, including Google's gVisor, AWS’ Firecracker, and TEE-based platforms like Gramine (utilizing Intel SGX) and AMD SEV. Our empirical results show that these attacks can also be carried out successfully against all of these sandboxes in less than 40 seconds, with an accuracy of over 70 % in all cases. Finally, we propose a noise injection-based countermeasure to mitigate the proposed attack on cloud environments.

This paper investigates the performance of various Border Gateway Protocol (BGP) security policies against multiple attack scenarios using different deployment strategies. Through extensive simulations, we evaluate the effectiveness of defensive mechanisms such as Root Origin Validation (ROV), Autonomous System Provider Authorization (ASPA), and PeerROV across distinct AS deployment types. Our findings reveal critical insights into the strengths and limitations of current BGP security measures, providing guidance for future policy development and implementation.

Cache-based website fingerprinting (WF) attacks violate user privacy where the attacker leverages the shared last-level cache in CPUs and analyzes the fingerprints through machine learning and deep learning models. WF attacks are even applicable in Incognito and anonymized browser platforms, leading to a serious threat to the public. Several defense techniques inject random noise during website rendering to degrade the attack success rate, while these techniques either create large performance overhead or cannot obfuscate the WF dataset entirely when the attacker retrains a new learning model with noisy fingerprints.

With the rising number of devices connected to the internet, the number of cyber-attacks on these devices increases in parallel. There are several strategies that an attacker can pursue, like stealing intellectual property of a victim or encrypting data to demand ransom for the decryption. In this work, we are focusing on the detection of so called cryptojacking attacks, in which an attacker that gained access to a system, then introduces programs that use the processing power of the victim device to mine cryptocurrencies. The presence of such an attack is not obvious right away and the longer an attacker manages to remain undetected, the longer they can profit having the victim foot the power bill. In this study, we combine previous approaches to demonstrate that cryptojacking attacks can be detected with an accuracy of 96% by leveraging hardware performance counters on the Windows operating system. Further, we present a method to determine which performance events result in the best detection rates, thus allowing the selection of a few performance events that can be monitored simultaneously by modern consumer CPUs. In a next step, we show that the CPU counters-based detection mechanism fails when an attacker switches from using the CPU resources to GPUs for the mining tasks. Based on these findings we then improve the previous detection approaches by extending the CPU performance counters with GPU-specific metrics resulting in 99.86% accuracy for the GPU-based cryptojacking attack class. In addition to a high detection rate the presented approach only causes a negligible performance loss while monitoring the whole system, which allows for continuous monitoring of live systems.

Microarchitectural attacks have become increasingly threatening the society with diverse set of attacks such as Spectre and Meltdown. Vendor patches cannot keep up with the pace of the new threats, which makes the dynamic anomaly detection tools more evident than before. Unfortunately, hardware performance counters (HPCs) utilized in previous works can detect a few microarchitectural attacks due to the small number of counters that can be profiled concurrently while introducing high performance overhead. These challenges consequently yield to inefficient detection tools in real-world security-critical systems. In this study, we introduce MAD-EN dynamic detection tool that leverages system-wide energy consumption traces collected from a generic Intel RAPL tool to detect ongoing anomalies in two different microarchitectures, namely Intel Comet Lake and Intel Tiger Lake. In the first phase of MAD-EN, we can distinguish 16 variants from 11 different micro-architectural attacks from benign applications by utilizing a binary-class CNN-based model with an F1 score of 0.998, which makes our tool the most generic attack detection tool so far. In the second phase, MAD-EN can recognize the respective attack types with a 95% accuracy by utilizing a multi-class CNN-based classification technique after the anomaly is detected. We demonstrate that MAD-EN introduces 69.3% less performance overhead compared to performance counter-based detection mechanisms, allowing more feasible real-time detection tool for generic purpose systems.

The arm race between hardware security engineers and side-channel researchers has become more competitive with more sophisticated attacks and defenses in the last decade. While modern hardware features improve the system performance significantly, they may create new attack surfaces for malicious people to extract sensitive information about users without physical access to the victim device. Although many previously exploited hardware and OS features were patched by OS developers and chip vendors, any feature that is accessible from userspace applications can be exploited to perform software-based side-channel attacks.

Website Fingerprinting attacks aim to track the visited websites in browsers and infer confidential information about users. Several studies showed that recent advancements in Machine Learning (ML) and Deep Learning (DL) algorithms made it possible to implement website fingerprinting attacks even though various defense techniques are present in the network. Nevertheless, trained models for website detection are not analyzed deeply to identify the leakage sources which are not always visible to both attackers and Cyber Threat Intelligence engineers. This study focuses on explaining ML and DL models in the context of microarchitecture-based website fingerprinting attacks. In the attack model, performance counters and cache occupancy side-channels are implemented on Google Chrome and Tor browsers. After ML and DL models are trained, LIME and saliency map XAI methods are applied to examine the leakage points in the side-channel data. In order to match the leakage samples in the measurements to the network traces, a novel dataset is collected by utilizing Google Chrome and Firefox browser developer tools. Next, the efficiency of explainable methods are analyzed with XAI metrics. Finally, an XAI-based obfuscation defense technique is proposed as a countermeasure against microarchitecture-based website fingerprinting attacks

Several techniques have been proposed to detect vulnerable Spectre gadgets in widely deployed commercial software. Unfortunately, detection techniques proposed so far rely on hand-written rules which fall short in covering subtle variations of known Spectre gadgets as well as demand a huge amount of time to analyze each conditional branch in software. Moreover, detection tool evaluations are based only on a handful of these gadgets, as it requires arduous effort to craft new gadgets manually. In this work, we employ both fuzzing and deep learning techniques to automate the generation and detection of Spectre gadgets. We first create a diverse set of Spectre-V1 gadgets by introducing perturbations to the known gadgets. Using mutational fuzzing, we produce a data set with more than 1 million Spectre-V1 gadgets which is the largest Spectre gadget data set built to date. Next, we conduct the first empirical usability study of Generative Adversarial Networks (GANs) in the context of assembly code generation without any human interaction. We introduce SpectreGAN which leverages masking implementation of GANs for both learning the gadget structures and generating new gadgets. This provides the first scalable solution to extend the variety of Spectre gadgets. Finally, we propose FastSpec which builds a classifier with the generated Spectre gadgets based on a novel high dimensional Neural Embeddings technique (BERT). For the case studies, we demonstrate that FastSpec discovers potential gadgets with a high success rate in OpenSSL libraries and Phoronix benchmarks. Further, FastSpec offers much greater flexibility and time-related performance gain compared to the existing tools and therefore can be used for gadget detection in large-scale software.

Over the past years, literature has shown that attacks exploiting the microarchitecture of modern processors pose a serious threat to user privacy. This is because applications leave distinct footprints in the processor, which malware can use to infer user activities. In this work, we show that these inference attacks can greatly be enhanced with advanced AI techniques. In particular, we focus on profiling the activity in the last-level cache (LLC) of ARM processors. We employ a simple Prime+Probe based monitoring technique to obtain cache traces, which we classify with deep learning methods including convolutional neural networks. We demonstrate our approach on an off-the-shelf Android phone by launching a successful attack from an unprivileged, zero-permission app in well under a minute. The app detects running applications, opened websites, and streaming videos with up to 98% accuracy and a profiling phase of at most 6 seconds. This is possible, as deep learning compensates measurement disturbances stemming from the inherently noisy LLC monitoring and unfavorable cache characteristics. In summary, our results show that thanks to advanced AI techniques, inference attacks are becoming alarmingly easy to execute in practice. This once more calls for countermeasures that confine microarchitectural leakage and protect mobile phone applications, especially those valuing the privacy of their users.

Cross-VM attacks have emerged as a major threat on commercial clouds. These attacks commonly exploit hardware level leakages on shared physical servers. A co-located machine can readily feel the presence of a co-located instance with a heavy computational load through performance degradation due to contention on shared resources. Shared cache architectures such as the last level cache (LLC) have become a popular leakage source to mount cross-VM attack. By exploiting LLC leakages, researchers have already shown that it is possible to recover fine grain information such as cryptographic keys from popular software libraries. This makes it essential to verify implementations that handle sensitive data across the many versions and numerous target platforms, a task too complicated, error prone and costly to be handled by human beings. Here we propose a machine learning based technique to classify applications according to their cache access profiles. We show that with minimal and simple manual processing steps feature vectors can be used to train models using support vector machines to classify the applications with a high degree of success. The profiling and training steps are completely automated and do not require any inspection or study of the code to be classified. In native execution, we achieve a successful classification rate as high as 98% (L1 cache) and 78\% (LLC) over 40 benchmark applications in the Phoronix suite with mild training. In the cross-VM setting on the noisy Amazon EC2 the success rate drops to 60\% for a suite of 25 applications. With this initial study we demonstrate that it is possible to train meaningful models to successfully predict applications running in co-located instances.