In the Linux kernel, the function __do_ajdtimex, introduced to enable adjtimex on any timekeeper, mistakenly references the core timekeeper (tk_core) when operating on an auxiliary timekeeper. This causes the core timekeeper’s sequence lock to be written to without holding its spinlock, triggering lock‑debug diagnostics and leading to an inconsistent leap‑second state across the kernel’s timekeeping subsystem.

All Linux kernel builds that include the 6.19 series up to at least release candidate 6.19‑rc6 are vulnerable. This includes any distribution or custom kernel derived from that line, regardless of the distribution, because the flaw resides in the core kernel timekeeping source.

The CVSS score is 5.5, indicating moderate severity. The EPSS score is less than 1 %, showing a very low likelihood of exploitation. It is not listed in CISA’s KEV catalog. The vulnerability is triggered by the adjtimex system call when an auxiliary timekeeper is targeted; the call requires elevated privileges, likely root, but the precise privilege level is not explicitly stated. Applying the available patch mitigates the issue completely.