New Spectre v2 Branch Target Reuse attack can leak Linux root password hashes on Intel CPUs

Researchers disclosed a new Spectre v2 attack variant that can extract a Linux system’s root password hash from Intel-based machines in minutes. The attack, called Branch Target Reuse (BTR), abuses stale branch-predictor state after just-in-time code or classic BPF (Berkeley Packet Filter) code reuses memory, and was assigned CVE-2026-64507 and CVE-2026-64508. The researchers tested it against the Linux kernel’s cBPF path and examined exposure in Firefox SpiderMonkey and Oracle GraalVM; Linux kernel fixes have already been merged.
Why it matters: This is a serious local information-leak flaw because an unprivileged attacker on a vulnerable Linux Intel system may be able to recover the root password hash and try to crack it offline. Linux administrators should prioritize kernel updates and review any additional mitigations vendors publish for affected runtimes and browsers.

Sources

New Spectre v2 attack variant leaks Linux root password hash in minutes
Bill Toulas 2026.09.29 100% relevant
This article appears to be the first tracked report here tying the new Branch Target Reuse Spectre v2 variant to Linux root-hash theft, the assigned CVEs, and the merged Linux kernel fixes.
New Spectre-v2 BTR Attack Leaks Linux Memory Despite Existing Defenses
info@thehackernews.com (The Hacker News) 2026.09.29 97% relevant
This appears to be the same underlying disclosure of the Spectre-v2 Branch Target Reuse (BTR) attack, adding a report that the attack can leak Linux memory despite current defenses and reinforcing the impact on affected Intel-based Linux systems.
← Back to all stories