CVE-2026-46113

A flaw was found in the Linux kernel's KVM (Kernel-based Virtual Machine) x86 shadow paging mechanism. This use-after-free vulnerability arises from incorrect handling of Guest Frame Numbers (GFNs) when guest page tables are modified. A local attacker with control over a guest virtual machine could exploit this to dereference freed memory, potentially leading to information disclosure, privilege escalation, or a denial of service (DoS).

The attack vector is from the guest to the host if KVM being used (and Shadow paging being used). KVM x86 shadow paging can leave a stale rmap entry when a direct shadow page is reused with an unexpected GFN after guest page table changes between VM entries. Later memslot deletion, dirty logging, or MMU notifier invalidation can walk the stale rmap and dereference a freed kvm_mmu_page, resulting in a use-after-free in the host kernel. For the CVSS the PR:L is used because a local actor with access to KVM or control of a guest execution context can influence the guest mappings and related memory management operations. The issue is not a normal network reachable bug, but it can cross the guest to host boundary in virtualized deployments. Impact is at least host denial of service and in the paranoid score may include confidentiality and integrity impact due to a plausible UAF based memory corruption primitive. For the paranoid score, choose the highest still-defensible interpretation supported by the bug class and patch context, with preference for manual-review sensitivity over autoclosed false negat ives.

The CVSS score(s) listed for this vulnerability do not reflect the associated product's status, and are included for informational purposes to better understand the severity of this vulnerability.

Mitigation for this issue is either not available or the currently available options don't meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base or stability.