Technical impact
The Linux kernel contains a race condition in the KVM Secure Encrypted Virtualization subsystem. During SEV page population, the function sev_gmem_post_populate() marks a page as dirty using set_page_dirty() without holding the necessary page lock. Because the page is obtained via get_user_pages_fast(), its mapping may be altered concurrently, leading to memory corruption or inadvertent changes to kernel data structures. This flaw does not directly enable code execution but can trigger crashes or data loss when the race succeeds.
Affected Systems
All Linux kernel installations that run KVM with Secure Encrypted Virtualization enabled are affected. Any kernel version that lacks the commit introducing set_page_dirty_lock() for sev_gmem_post_populate() is considered vulnerable; the glitch exists before that patch across all releases.
Risk and Exploitability
The CVSS score is 7.8 and the EPSS score is < 1%. The vulnerability is not listed in CISA KEV, indicating no documented exploitation. Exploitation requires privileged, kernel-level access and the ability to trigger the race during SEV page population, which constrains the threat space. The risk is therefore low to moderate until the kernel is upgraded, as accidental corruption could occur during normal SEV operation but cannot be reliably targeted by an attacker without host privileges.