The Linux kernel net liquidio driver contains a defect where, if the function to obtain the PF number fails, the code returns without unmapping the BAR resources that had been allocated. This leaks hardware resources and can leave memory mappings in an inconsistent state. The missing cleanup could lead to degraded performance, potential memory pressure, or driver instability, depending on how the leaked BAR resources are handled by the system. The available data does not specify whether an attacker can trigger the error path or what the exact consequences would be, but a resource leak in kernel space may be used to drive the system toward a denial‑of‑service condition.

This issue affects Linux kernel implementations that include the net liquidio driver prior to the patch. Because the driver is part of the core kernel, all distributions shipping a kernel that contains the v2.3:linux:linux_kernel component are potentially impacted until the code change is included. Specific kernel versions are not listed, so any kernel version that incorporates the affected code path should be examined.

No CVSS score or EPSS information is available, and the vulnerability is not listed in CISA KEV, so the objective risk level cannot be determined from the public data. However, kernel resource leaks can be leveraged by a local or privileged attacker to exhaust system resources or cause driver instability, which could in turn affect service availability. The likely attack vector would be exploitation of the network interface driver during initialization or operation of a network device that uses the liquidio backend. Because the code path is internal to the kernel, elevated privileges or direct access to the device would be necessary to trigger the error condition.