CVE-2026-89485

CriticalCVSS 9.8

Published: 11/09/2026(UTC)Updated: 13/09/2026(UTC)Translated: 12/09/2026(UTC)NVD NIST

Exploitation Probability (EPSS)

Low risk

0.20%

10th percentile - higher than 10% of all known CVEs

Summary

In the Linux kernel, lockd's nlm_traverse_files() pins the current file with f_count++ across a mutex unlock but does not pin the saved next pointer. A concurrent nlm_release_file() can free next during the unlock window, and the iterator dereferences freed memory on the next loop step. The fix pins both files and releases them after the iteration.

Risk Assessment

Can cause a kernel use-after-free, potentially leading to system instability or privilege escalation. Affects systems using NFS and lockd.

Recommendation

Update the Linux kernel to a version with the fix that pins both current and next files before the lock-drop. Until patched, consider limiting NFS file operations.

Other vulnerabilities in Linux kernel (lockd)

CVE-2026-89484Low risk· EPSS 10%
In the Linux kernel, lockd's nlmclnt_locks_init_private() installs NLM file lock operations even when nlmclnt_find_lockowner() fails to allocate a lockowner. The VFS still tears down the partially initialized file_lock and calls locks_release_private(), which dereferences fl->fl_u.nfs_fl.owner and crashes (NULL dereference). The fix clears fl_ops before initialization and installs NLM operations only after a lockowner is allocated.

Original NVD description (English source)

In the Linux kernel, the following vulnerability has been resolved: lockd: pin next file across nlm_inspect_file lock-drop nlm_traverse_files() pins the current file with f_count++ across a mutex_unlock for nlm_inspect_file(), but nothing pins the saved next pointer. A concurrent nlm_release_file() can kfree the next file during the unlock window, and the iterator dereferences freed memory on the next loop step. Pin both current and next before the lock-drop. Advance by swapping the pinned cursors at the end of each iteration so next is always held alive across the unlock. Always call nlm_file_release() after dropping the iteration pin, regardless of whether the file matched the predicate. Use nlm_file_inuse(), which does a live walk of the inode lock list, rather than the cached f_locks field, so skipped files that never ran nlm_inspect_file() are evaluated correctly. Because every file in a hash bucket is now pinned and released, files skipped by the is_failover_file predicate that have no locks, blocks, shares, or external references are deleted during traversal. The old code never evaluated skipped files for cleanup. The new behavior is intentional: such files are stale and should not persist in the table.