惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
H
Help Net Security
云风的 BLOG
云风的 BLOG
Apple Machine Learning Research
Apple Machine Learning Research
H
Hackread – Cybersecurity News, Data Breaches, AI and More
Hugging Face - Blog
Hugging Face - Blog
博客园_首页
D
Docker
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
Blog — PlanetScale
Blog — PlanetScale
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
GbyAI
GbyAI
博客园 - Franky
B
Blog RSS Feed
Stack Overflow Blog
Stack Overflow Blog
L
LangChain Blog
量子位
V
Visual Studio Blog
Y
Y Combinator Blog
小众软件
小众软件
N
Netflix TechBlog - Medium
博客园 - 三生石上(FineUI控件)
Microsoft Security Blog
Microsoft Security Blog
雷峰网
雷峰网

The Register - Security: Research

Novel Blue Moon kit targeting Chrome and Windows reflects new reality of AI-driven exploits Extortion crews have their eyes on high-value AI data, Google warns Researcher shows how Claude Code can be tricked simply by asking it to summarize a website Copilot tricked into telling reseachers how to hack itself Akira ransomware scum blocked victim How the famed USENIX Security conf is managing a flood of papers in the AI era www.theregister.com Self-destructing Mistic backdoor linked to access broker selling corporate footholds to ransomware gangs PRC-linked spies hid inside medical and military networks for more than a year, snooping through Gmail and stealing data Nobody needs Mythos or 0-days to build a chaos-causing computer worm – free open source models work just fine ChatGPT blindly trusts browser content, turning the page into a payload Russia-linked threat group put ChatGPT to work from lure to payload Kids can bypass some age checks with a drawn-on mustache What type of 'C2 on a sleep cycle' do they leave behind? Novel Chinese spy group found in critical networks in Poland, Asia ORNL builds more sensitive GPS interference detector Researchers find sabotage malware that may predate Stuxnet Vibe coding upstart Lovable denies data leak, cites 'intentional behavior,' then throws HackerOne under the bus Anthropic, Google, Microsoft paid AI bug bounties – quietly Security reserchers tricked Apple Intelligence into cursing Don't open that WhatsApp message, Microsoft warns Security boffins harvest bumper crop of API keys from web Lightning-fast exploits mean patch fast, says Cisco Talos AI agents are 'gullible' and easy to turn into your minions Smooth criminals talking their way into cloud environments, Google says Snoops plant info-stealing malware on iPhones, Google warns Cybercrime up 245% since the start of the Iran war Rogue AI agents can work together to hack systems Fake applicants are sending security-killing malware AI agent hacked McKinsey chatbot for read-write access Kaspersky: No signs Coruna iPhone exploit kit made by US
Ruh-roh. DDR5 memory vulnerable to new Rowhammer attack
Simon Sharwood · 2025-09-17 · via The Register - Security: Research

Research

Google and ETH Zurich found problems with AMD/SK Hynix combo, will probe other hardware

Researchers from Google and Swiss university ETH Zurich have found a new class of Rowhammer vulnerability that could allow attackers to access info stored in DDR5 memory.

As Google explains in its post on the discovery, DRAM stores data as electrical charges in small “cells” of memory. Those charges leak over time, corrupting data. Computer scientists have known this for ages, and the controllers that manage memory therefore periodically refresh cells to ensure memory remains reliable.

Infosec types with friendly and/or adversarial inclinations noticed that behavior and wondered what would happen if they repeatedly accessed specific rows of memory cells. They learned that by “hammering” those rows of cells with many access requests it’s possible to corrupt data in adjacent cells, degrade system performance, or even achieve privilege escalation.

REG AD

Rowhammer is a known problem and infosec researchers long ago developed defenses that system builders and memory-makers adopted. Last year, standards body the JEDEC Solid State Technology Association introduced a new DRAM data integrity measure called Per-Row Activation Counting (PRAC) that looks for the sort of activity involved in a Rowhammer attack and pauses traffic to stymie hostile action.

REG AD

Google’s researchers, however, assert that systems that include DDR5 have not employed PRAC. The web giant also created a pair of tools to test DDR5 modules for susceptibility to Rowhammer.

Researchers at ETH Zurich put those tools to work and found a new form of Rowhammer attack that works on DDR5 from SK Hynix, the world’s largest memory-maker.

The attack, called “Phoenix”, isn’t simple and is computationally expensive. But it works.

And that’s worrying because the paper [PDF] that describes the joint Google/ETH research opens by observing “DDR5 has shown an increased resistance to Rowhammer attacks in production settings. Surprisingly, DDR5 achieves this without additional refresh management commands.”

Google and ETH Zurich found their Rowhammer variant using a machine powered by an AMD Zen 4 processor and SK Hynix DDR5 and will attempt to replicate their work on memory and CPUs from other vendors.

If the researchers succeed it’s not a disgrace for impacted manufacturers because Rowhammer-style attacks are hard to defeat, with recent victims including Nvidia, DDR4 and everyone’s privacy thanks to a Rowhammer variant that makes it possible to fingerprint computing devices.

The attack discovered by Google and ETH Zurich is now known as CVE-2025-6202 and earned a 7.1 CVSS rating.

ETH Zurich says it conducted responsible disclosure of Phoenix that saw it inform SK Hynix, CPU vendors, and major cloud providers on June 6, 2025. AMD told the researchers it made a BIOS update to protect systems that use its processors. More information, including the source code for all the experiments and the exploit, can be found here. ®