CVE-2026-80685 in Linux
Summary
by MITRE • 08/28/2026
In the Linux kernel, the following vulnerability has been resolved:
mm/util: don't read __page_2 for order-1 folios in snapshot_page()
snapshot_page() currently reads __page_2 after checking nr_pages > 1, but it should only do so when nr_pages > 2.
If an order-1 folio is allocated at the end of a vmemmap section, __page_2 will not exist and reading it will cause a fault.
During DLPAR memory remove on a 22 TB ppc64le LPAR, snapshot_page() oopsed on the page isolation path while reading an order-1 folio's __page_2 from an adjacent absent section (unmapped vmemmap).
Fix this to avoid reading memmap that doesn't exist (e.g., a vmemmap hole).
If you want to get best quality of vulnerability data, you may have to visit VulDB.
Analysis
by VulDB Data Team • 08/28/2026
The Linux kernel vulnerability identified in the memory management subsystem involves a critical out-of-bounds read within the snapshot_page function, specifically affecting how page structures are accessed during memory operations. The core technical flaw lies in an incorrect boundary check for folio orders. The code currently verifies if the number of pages is greater than one before attempting to access the __page_2 field of the second page structure. However, this logic fails to account for order-1 folios that reside at the end of a vmemmap section where adjacent memory map structures may not exist. When such a condition occurs, the kernel attempts to read from an unmapped or absent virtual memory mapping area, leading to a segmentation fault or general protection fault within the kernel space. This represents a classic case of improper input validation regarding structural boundaries in low-level memory management routines.
The operational impact of this vulnerability is most severe during dynamic logical partitioning operations, particularly on PowerPC architectures like ppc64le where DLPAR memory removal triggers complex page isolation sequences. In scenarios involving large memory configurations such as a 22 TB LPAR, the likelihood of an order-1 folio being allocated at the boundary of a vmemmap section increases significantly. When snapshot_page is invoked during these operations, it may encounter a situation where __page_2 does not physically exist in the kernel's virtual memory map due to holes created by absent sections. The resulting crash disrupts system stability and can lead to denial of service conditions for running workloads relying on dynamic memory management features. This issue highlights the fragility of assumptions made about contiguous physical memory layouts when dealing with sparse or dynamically changing memory maps.
From a vulnerability classification perspective, this flaw aligns with CWE-125 Out-of-bounds Read and CWE-369 Divide By Zero which in broader contexts can manifest as improper boundary checks leading to invalid memory access. The attack vector is primarily local and requires specific hardware configuration states or timing conditions related to memory hotplug operations rather than direct user exploitation through standard application interfaces. In the context of MITRE ATT&CK, this vulnerability relates to techniques involving kernel-level instability that could potentially be leveraged for privilege escalation if an attacker can influence memory allocation patterns near vmemmap boundaries, although the primary impact remains system stability and availability. The lack of proper bounds checking on structural offsets within page structures is a common source of such vulnerabilities in operating systems with complex memory management architectures.
Mitigation strategies primarily involve applying the upstream kernel patch that corrects the conditional logic to ensure __page_2 is only accessed when nr_pages exceeds two, thereby guaranteeing the existence of the second page structure relative to the folio base. System administrators should prioritize updating their Linux kernels on affected PowerPC platforms to versions containing this fix. Additionally, monitoring logs for oops messages related to snapshot_page or vmemmap access violations can help identify systems still running vulnerable code paths. For environments utilizing dynamic memory removal features, ensuring that kernel updates are promptly applied is critical to maintaining system integrity and preventing crashes during routine maintenance operations involving memory reconfiguration.