The Linux kernel MPTCP subsystem uses the non‑sleepable lock_sock_fast() around sock_set_timestamp() and sock_set_timestamping(), operations that can sleep. This misuse can trigger a scheduling conflict and cause an atomic context panic, leading to a kernel crash that brings the entire system down. The flaw does not provide code execution or privilege escalation, but it disrupts availability and can result in data loss during an abrupt shutdown.

All Linux kernels that include the original MPTCP timestamp socket option code and have not yet incorporated the recent commit series fixing this issue are affected. The specific kernel versions are not listed, so any kernel build prior to the inclusion of the patch remains vulnerable. Distributions that ship unaffected kernels should upgrade to the latest stable kernel that contains the fixes referenced in the commit history.

The CVSS score is 5.5 and the EPSS score is < 1%, indicating a low exploitation probability. However, because the vulnerability leads directly to a kernel crash, its severity is high. The flaw can be exploited by triggering the MPTCP timestamp option, which normally requires at least local process privileges; consequently the likely attack vector is inferred to be local or requires privileged network access. The vulnerability is not listed in CISA’s KEV catalog, but its impact warrants immediate remediation.