The vulnerability is a spinlock leak in the Linux kernel function that handles memory migration for huge pages. When a stability check fails after the PMD spinlock is acquired, the code jumps to an abort label that bypasses the unlock, leaving the spinlock permanently held. The locked spinlock blocks further progress of the migration path and can prevent other kernel operations from acquiring the resource, resulting in a deadlock that effectively halts memory management and can bring the system to a stand‑still.

All current stable Linux kernel releases that contain the vulnerable migrate_vma_insert_huge_pmd_page code path are affected. The fix is included in the kernel after commit 63451de1, so any system running a kernel older than that commit contains the flaw.

The CVSS score is not provided and the EPSS score is reported as less than 1 %, indicating a very low expected exploitation probability at this time. The vulnerability is not listed in CISA KEV. The description does not disclose an explicit attack vector, privilege requirement, or conditions beyond the failure of a stability check during a high‑memory‑usage migration operation. Consequently, the exact scope of exploitability cannot be determined from the available data, though the capacity to induce a system‑wide denial of service exists once the spinlock is held.