New type of attack can slip past the defenses in your computer’s processor
MIT CSAIL researchers found a way to exploit a split-second gap in chip security — and used it to acquire a Linux system’s password file.
Modern processors predict which way a program will branch, saving time when they're correct. However, traces of these guesses linger, allowing attackers to access sensitive memory information. Defenses aim to wipe or isolate the processor's prediction machinery, but a new MIT study reveals a gap between wiping and prediction use, enabling a class of attacks called TONTOU.
This involves injecting code during processor pauses to manipulate interrupts, leading to mispredictions. Tested on Intel and AMD processors, the attack bypasses protections on both, copying sensitive data like Linux root password hashes. While AMD's saferet defense leaves a vulnerable window, slowing down the processor can still breach it.
The researchers notified both companies and released their findings, suggesting additional cleaning steps for AMD and performance trade-offs for Intel.
Written by urgent.news from MIT News Research's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
