Back Feeds.4Sysops

TONTOU attack bypasses Spectre v2 fixes to leak Linux password hashes

Feeds.4Sysops •IT News • August 7, 2026

A new TONTOU “interrupt injection” attack can re-poison branch predictors after Spectre v2 defenses run, allowing unprivileged code to extract sensitive data from Linux kernel memory. An AMD kernel fix is already available, but shared hosts remain at risk until administrators deploy it and review workload isolation. Source

The full story

This article is one source in a clustered incident — the cluster page carries the summary, timeline and every other outlet covering it.

New TONTOU Attack Exploits Spectre v2 Defenses on Intel and AMD CPUs MIT researchers have unveiled a new speculative execution attack named TONTOU, which bypasses mitigations against Spectre v2 on Intel and AMD processors. The attack exploits a timing vulnerability in the post-neutralization window of branch predictors, allowing unprivileged code to re-poison these predictors and leak sensitive data, such as Linux password hashes. The researchers demonstrated the attack on AMD Zen 2 and Intel Cascade Lake Refresh processors. The method involves scheduling high-frequency timer interrupts to redirect control flow during critical execution phases. An AMD kernel fix is available, but many systems remain vulnerable until updates are applied. The attack highlights weaknesses in existing Spectre mitigations and poses a significant risk to Linux systems. The researchers will present their findings at DEF CON 34. 4 sources 71

Extracted Entities

TONTOU attack bypasses Spectre v2 fixes to leak Linux password hashes

The full story

Extracted Entities

Attack Types (1)

CWE Weaknesses (1)

MITRE ATT&CK (1)

Platforms (1)

Vulnerabilities (1)