
Spectre Variant Returns: Branch Target Reuse Attack Extracts Root Password Hash from Linux Memory
Researchers from VUSec and Scuola Superiore Sant’Anna have disclosed Branch Target Reuse, a new Spectre v2 variant that exploits stale branch predictor entries in modern CPUs. The attack targets JIT compilers that generate and reuse executable code at runtime, allowing speculative execution of instructions from previously freed memory regions. On Intel systems with existing mitigations enabled, the researchers demonstrated extraction of the root password hash from the Linux kernel in minutes. Practical proof-of-concept exploits were developed against the Linux kernel, while PoCs were also prepared for Firefox and tested on GraalVM. The issue affects Intel, AMD, and Arm processors, although exploitation success depends on the specific JIT environment and predictor state. Defenses have already been merged into the Linux kernel and GraalVM, while Mozilla continues work on site isolation. The findings highlight that Spectre-class issues remain relevant as long as processors rely on aggressive speculative execution.
AntiMalwareSep 30, 2026Translated from Russian

