In the Linux kernel, the dm_cache policy segment management queue (smq) was missing synchronization protection when invalidating cache blocks in passthrough mode. The concurrent calls to invalidate_mapping from multiple workers create a race condition that can corrupt the allocated blocks counter or cause use‑after‑free of internal structures. Since the kernel operates with elevated privileges, such a flaw can lead to kernel panic or arbitrary code execution at the privilege level of the system.

All Linux kernel installations that use the dm_cache device with the smq policy are potentially affected. The flaw is in the dm_cache module itself, so any distribution that ships a kernel without the missing locks – there is no specific version given – falls under this risk.

Because no CVSS score is supplied, the severity cannot be quantified, and EPSS is unavailable. However, the race and use‑after‑free make this a serious vulnerability in the kernel, and the presence of a concurrent workload in passthrough mode is required to trigger it. No public exploit has been reported, nor is the flaw listed in the CISA KEV catalog, which suggests an exploitation window may still be open for active systems.