The vulnerability involves the Linux kernel's KCM component, which temporarily replaces lower TCP socket callbacks with custom handlers and then restores them. The original code performed direct pointer assignments without enforcing proper memory visibility, so if another CPU reads an old callback pointer after it has been restored, the callback may execute with a mismatched target and stale state. This race condition can cause incorrect wake‑up handling, potentially leading to kernel panics, memory corruption, or denial of service. The weakness is a classic race condition on shared kernel data.

The affected product is the Linux kernel. No specific version ranges are listed in the CNA data, so any kernel including the vulnerable kcm_attach and kcm_unattach paths may be at risk. Systems should verify whether their running kernel contains the fix that replaces the patching changes.

The CVSS score of 7.8 indicates substantial severity and the EPSS score of 0.00136 (<1%) indicates very low exploitation likelihood, but the flaw is a local kernel race that can be triggered by interacting with TCP sockets. Because the bug involves internal kernel mechanisms, it is unlikely to be exploited remotely without local kernel code execution. The absence of KEV status suggests no known widespread exploitation. Nevertheless, the potential for a critical kernel crash warrants prompt remediation.