March 9, 2017 By Larry Loeb 2 min read

Software Guard Extensions (SGX) are essentially an isolation mechanism. Like an RSA key, it tries to protect code and data from modification or disclosure, even if all privileged software is considered malicious.

Now, Intel added the SGX to its hardware starting with Skylake models in an effort to get a protective fence nailed down around data as it floats through the cloud.

The SGX Concept

The idea was that if you stored the keys inside this special enclave memory area, it would be encrypted and kept isolated from the rest of the chip through Intel’s hardware. Even if attackers got into the enclave somehow, they would find data encrypted and useless — or so Intel thought.

But five people at Graz University of Technology in Austria came up with a way to hose SGX. They devised an attack method that utilizes the first known SGX malware, which they also developed.

These researchers wrote a paper, titled “Malware Guard Extension: Using SGX to Conceal Cache Attacks,” that gets into the details of how they operate the malware. To make their point, they showed it working both in a native environment and across multiple Docker cloud containers.

The malware was a key-stealing proof of concept (PoC) that was directed at another co-located, secured enclave. The malware used SGX against itself, keeping its existence secret from other software processes by hiding inside of the SGX isolation.

Recovering the RSA Key

The researchers reported that the PoC malware was able to recover RSA keys by monitoring the cache access patterns of an RSA signature process in a semisynchronous attack. They found that a “Prime+Probe cache side-channel” attack could get 96 percent of an RSA private key from just one single trace, according to the paper.

SecurityWeek noted that the attack could extract the full RSA private key from 11 traces within five minutes of operation. Once the attackers have the private key, the enclave can be decrypted.

Researchers were able to come up with “highly accurate” timings within an enclave without access to the native time stamp counter. In fact, they reported that the method they used to replace the timer is even more accurate than if they had access to the native counter.

There is no easy mitigation for this problem. A comprehensive solution may require changes to the enclave, operating system and hardware itself. Until a resolution is found, this is a serious and ubiquitous vulnerability, and SGX is making it undetectable.

More from

ChatGPT 4 can exploit 87% of one-day vulnerabilities: Is it really that impressive?

2 min read - After reading about the recent cybersecurity research by Richard Fang, Rohan Bindu, Akul Gupta and Daniel Kang, I had questions. While initially impressed that ChatGPT 4 can exploit the vast majority of one-day vulnerabilities, I started thinking about what the results really mean in the grand scheme of cybersecurity. Most importantly, I wondered how a human cybersecurity professional’s results for the same tasks would compare.To get some answers, I talked with Shanchieh Yang, Director of Research at the Rochester Institute…

ONCD releases request for information: Open-source software security

3 min read - Open-source software is a collective partnership across the development community that requires both private and public buy-in. However, securing open-source software can be tricky. With so many different people working on the coding, security measures are often overlooked, increasing the chances that a vulnerability will fall through the cracks and be exploited. The Open-Source Software Security Initiative (OS31) aims to provide governance over open-source security processes. After the Log4Shell vulnerability, securing open-source software became a top priority for the federal…

How cyber criminals are compromising AI software supply chains

3 min read - With the adoption of artificial intelligence (AI) soaring across industries and use cases, preventing AI-driven software supply chain attacks has never been more important.Recent research by SentinelOne exposed a new ransomware actor, dubbed NullBulge, which targets software supply chains by weaponizing code in open-source repositories like Hugging Face and GitHub. The group, claiming to be a hacktivist organization motivated by an anti-AI cause, specifically targets these resources to poison data sets used in AI model training.No matter whether you use…

Topic updates

Get email updates and stay ahead of the latest threats to the security landscape, thought leadership and research.
Subscribe today