In the Linux kernel, the function that flushes the device IOTLB performs a check using pci_dev_is_disconnected() to skip the operation when a device reports itself disconnected. That check does not detect link‑down states caused by faults, so when a VM releases a PCIe device that has become faulted but not yet reported as disconnected, the flush attempt continues and the kernel hard‑locks. The resulting denial of service can bring the entire host to a halt, affecting all workloads.

All Linux systems running a kernel version that includes this logic are affected, regardless of the distribution vendor. The advisory does not list specific kernel releases, but any kernel that contains the commit highlighted in the description (“iommu/vt‑d: Flush dev‑IOTLB only when PCIe device is accessible in scalable mode”) is at risk. The vulnerability is present in both the mainline Linux kernel and kernel variants used by major vendors.

The issue is not catalogued in CISA KEV and no public CVSS or EPSS score is available. Based on the description, it is inferred that an attacker who can trigger a link‑down fault on a PCIe device or force the destruction of a VM attached to such a device could exercise the vulnerability. While no exploit has been documented, the effect—a hard lockup that can stop the entire host—means the potential impact is high for environments that require continuous kernel operation. The likelihood of exploitation is uncertain, but in systems where virtual machines are actively managed, the risk is significant enough to warrant prompt mitigation.