The vulnerability stems from a deadlock condition in the Linux mlx5 driver. When the eswitch workqueue processes a mode change or vport event, it may acquire the devlink lock, while the devlink operation also attempts to acquire the same lock. This ordering conflict causes the kernel scheduler to wait indefinitely, resulting in a kernel that becomes unresponsive or a halted network interface.

Every Linux kernel that includes the original mlx5 eswitch implementation before the commit that adds the generation counter and removes the workqueue flush is vulnerable. Administrators should verify whether the running kernel contains this patch by checking the kernel tree or the commit history.

The CVSS score of 5.5 reflects moderate severity. The EPSS score is below 1%, and the issue is not currently listed in the CISA KEV catalog, indicating that widespread exploitation has not been observed. Based on the description, it is inferred that an attacker with local privileges could trigger the deadlock by performing normal driver operations such as eswitch mode changes or vport configuration. The risk is limited to availability loss rather than confidentiality or integrity compromise.