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

N
Netflix TechBlog - Medium
J
Java Code Geeks
爱范儿
爱范儿
雷峰网
雷峰网
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
博客园 - 三生石上(FineUI控件)
H
Hackread – Cybersecurity News, Data Breaches, AI and More
B
Blog RSS Feed
Google DeepMind News
Google DeepMind News
Jina AI
Jina AI
The GitHub Blog
The GitHub Blog
I
InfoQ
月光博客
月光博客
博客园 - 聂微东
博客园 - Franky
The Cloudflare Blog
阮一峰的网络日志
阮一峰的网络日志
博客园_首页
G
Google Developers Blog
Blog — PlanetScale
Blog — PlanetScale
L
LangChain Blog
罗磊的独立博客
Apple Machine Learning Research
Apple Machine Learning Research

Schneier on Security

More on the OpenAI Agent's Attack on Hugging Face - Schneier on Security The OpenAI Hack Shows the Genie Is Out of the Bottle - Schneier on Security Friday Squid Blogging: Squid Helps Discover New Marine Species - Schneier on Security Anthropic's Opus 5 Is Better at Resisting Prompt Injection - Schneier on Security Facial Recognition at Madison Square Garden - Schneier on Security American Being Prosecuted for Wiping His Phone Before Handing It Over to Border Officials - Schneier on Security Should You Use AI for a Task? Here’s a Simple Way to Decide - Schneier on Security Measuring the Tendency of AI Agents to Go Rogue - Schneier on Security Long-Lived Vulnerability in Microsoft Secure Boot - Schneier on Security Measuring LLMs' Ability to Perform Cryptanalysis - Schneier on Security Axon Is Another License Plate Surveillance Company - Schneier on Security Cognyte Sells a Mobile Cell Surveillance Van - Schneier on Security Friday Squid Blogging: Illex Squid Catch in the Falklands - Schneier on Security Why AI Needs a “Genie Coefficient” - Schneier on Security End-to-End Encryption and "Going Dark" - Schneier on Security First-Person Identity Theft Story - Schneier on Security MIT to Become Hotbed of AI Video Surveillance - Schneier on Security On Flock License Plate Tracking Cameras - Schneier on Security Friday Squid Blogging: Squid Washing Up on Cape Cod Beach - Schneier on Security Details of Alan Turing's Voice Encryption System - Schneier on Security Protecting Privacy in an AI Era - Schneier on Security A Video Screen That Is Also a Camera - Schneier on Security Upcoming Speaking Engagements - Schneier on Security Vulnerability in FIFA's Network - Schneier on Security AI Data Centers and the Concentration of Wealth - Schneier on Security Friday Squid Blogging: "Squidbleed" Vulnerability - Schneier on Security AI Surveillance and Social Progress - Schneier on Security The Language of AI Could Change How Humans Speak - Schneier on Security Cybersecurity and the Gap Between Skill and Ability - Schneier on Security Google Is Suing Chinese Scammers Who Are Using Gemini - Schneier on Security
Rowhammer Attack Against NVIDIA Chips
Bruce Schnei · 2026-05-06 · via Schneier on Security

A new rowhammer attack gives complete control of NVIDIA CPUs.

On Thursday, two research teams, working independently of each other, demonstrated attacks against two cards from Nvidia’s Ampere generation that take GPU rowhammering into new—­and potentially much more consequential—­territory: GDDR bitflips that give adversaries full control of CPU memory, resulting in full system compromise of the host machine. For the attack to work, IOMMU memory management must be disabled, as is the default in BIOS settings.

“Our work shows that Rowhammer, which is well-studied on CPUs, is a serious threat on GPUs as well,” said Andrew Kwong, co-author of one of the papers. “GDDRHammer: Greatly Disturbing DRAM Rows­Cross-Component Rowhammer Attacks from Modern GPUs.” “With our work, we… show how an attacker can induce bit flips on the GPU to gain arbitrary read/write access to all of the CPU’s memory, resulting in complete compromise of the machine.”

Update Friday, April 3: On Friday, researchers unveiled a third Rowhammer attack that also demonstrates Rowhammer attacks on the RTX A6000 that achieves privilege escalation to a root shell. Unlike the previous two, the researchers said, it works even when IOMMU is enabled.

The second paper is GeForge: Hammering GDDR Memory to Forge GPU Page Tables for Fun and Profit:

…does largely the same thing, except that instead of exploiting the last-level page table, as GDDRHammer does, it manipulates the last-level page directory. It was able to induce 1,171 bitflips against the RTX 3060 and 202 bitflips against the RTX 6000.

GeForge, too, uses novel hammering patterns and memory massaging to corrupt GPU page table mappings in GDDR6 memory to acquire read and write access to the GPU memory space. From there, it acquires the same privileges over host CPU memory. The GeForge proof-of-concept exploit against the RTX 3060 concludes by opening a root shell window that allows the attacker to issue commands that run unfettered privileges on the host machine. The researchers said that both GDDRHammer and GeForge could do the same thing against the RTC 6000.

Tags: , , ,

Posted on May 6, 2026 at 6:36 AM0 Comments

Sidebar photo of Bruce Schneier by Joe MacInnis.