The vulnerability corrects lock ordering for hugetlb file folios in the Linux kernel. A deadlock can arise when one task holds the folio lock and then attempts to acquire the i_mmap_rwsem read lock while another task holds the i_mmap_rwsem write lock and then attempts to acquire the folio lock. This race condition can cause the affected processes to hang and, in worst‑case scenarios, stall the system by blocking I/O or memory migration operations. The impact is a denial of service rather than a disclosure or privilege escalation.

All Linux kernel installations that include the affected code paths. Kernels 5.10 (release candidates rc4 through rc7) and 6.19 (release candidates rc1 through rc6) contain the pre‑fix code as shown in the CPE list. Any distribution using one of these kernel versions without the corresponding patch is exposed.

The CVSS score of 5.5 indicates moderate severity. EPSS is less than 1%, labeling the likelihood of exploitation extremely low. The bug does not provide remote code execution capabilities and typically requires concurrent local tasks to trigger the deadlock, making it unlikely to be exploited by remote attackers. The vulnerability is not currently listed in the CISA KEV catalog, further indicating limited real‑world exploitation. However, affected systems should still address the issue promptly to avoid potential service interruptions.