The flaw lies in the Linux kernel's memory‑management logic for x86 architectures. When a hugetlbfs mapping is requested, the core routine get_align_mask() supplies the page‑alignment mask for huge pages, but a subsequent call to get_align_bits() mistakenly adds general va_align bits even for a huge‑page mapping. This causes vm_unmapped_area() to return a virtual address that is not aligned to a huge page, and when the kernel later attempts to tear down the mapping with __unmap_hugepage_range(), a BUG_ON is triggered. The kernel therefore panics, resulting in an immediate denial of service. The weakness falls under incorrect alignment handling (CWE‑617).

It affects Linux kernel builds on AMD CPUs using the F15h microarchitecture, specifically Carrizo and FX‑8370E processors. All kernel versions released before the commit that fixes the alignment bug are insecure. The issue is confined to the x86/mm subsystem and will be corrected in any kernel that contains the listed patch references.

The CVSS score of 5.5 indicates a moderate severity. The EPSS score is less than 1%, showing a very low probability of exploitation in the wild, and the vulnerability is not listed in the CISA KEV catalog, implying no publicized attacks. Based on the description, it is inferred that a local attacker who can request hugetlbfs mappings requires kernel privileges or elevated rights to exploit the bug; non-privileged users would need an additional vector to gain access to huge‑page memory. Although the attack requires elevated rights, it can still be used to crash a system by creating a hugetlbfs mapping on the affected hardware.