Researchers at the Massachusetts Institute of Technology (MIT) have made a significant discovery in the realm of cybersecurity, unveiling a new attack vector known as TONTOU. This attack exploits timer interrupts to reopen the branch predictor poisoning window, effectively bypassing Spectre defenses implemented on Intel and AMD CPUs.
Key Insights
The TONTOU attack is particularly noteworthy as it has been proven to work on Zen 2 CPUs, a modern processor architecture. The key points of this discovery include the TONTOU attack's ability to bypass Spectre defenses on both Intel and AMD CPUs, its exploitation of timer interrupts to reopen the branch predictor poisoning window, and the successful demonstration of a working exploit on Zen 2 CPUs.
