CVE-2026-93827

A flaw was found in the Linux kernel's virtio-fs file system driver. When queue initialization fails—such as when a device advertises more request queues than supported—the driver fails to clear memory pointers after releasing them, resulting in a double-free condition during error cleanup. An attacker capable of presenting a malformed virtio-fs device can exploit this flaw to cause memory corruption, leading to a system crash and Denial of Service (DoS).

This vulnerability is rated as Important because an attacker capable of emulating virtual devices or a compromised host presenting malformed virtio devices can induce kernel memory corruption within guest systems, potentially causing denial of service. In Red Hat environments, virtio-fs is exclusively used for host-to-guest shared filesystems, meaning bare-metal deployments and virtual machines without virtio-fs devices are unaffected. The flaw is constrained to initial device discovery when the guest kernel probes and configures request queues.

The CVSS score(s) listed for this vulnerability do not reflect the associated product's status, and are included for informational purposes to better understand the severity of this vulnerability.

To mitigate this vulnerability, prevent the virtiofs kernel module from loading on the system if virtio-fs shared filesystems are not required.

1. Add a configuration rule to disable autoloading of the module:

echo "install virtiofs /bin/true" > /etc/modprobe.d/disable-virtiofs.conf

2. If the module is currently loaded, unload it:

modprobe -r virtiofs

Warning: If virtiofs is in active use by mounted filesystems, unloading the module will fail. A system reboot may be required to completely unload the driver and ensure configuration changes take effect.

Caveats: Disabling the virtiofs module prevents the guest system from mounting filesystems shared via virtio-fs from the hypervisor.