Back

CVE-2026-99312: Linux KVM Flaw Lets Guest VMs Access Host Memory on AMD Hosts
SiTech AI Team4 min read

CVE-2026-99312: Linux KVM Flaw Lets Guest VMs Access Host Memory on AMD Hosts

CVE-2026-99312 affects KVM on AMD hosts with nested virtualization enabled. Researchers read about 127 GB of a DigitalOcean server's memory from a guest virtual machine.

A vulnerability in KVM, tracked as CVE-2026-99312, allows a guest virtual machine on an AMD host with nested virtualization enabled to access host memory. Oren Yomtov of Terra Security found the flaw and reported it through DigitalOcean's bug bounty program; DigitalOcean confirmed the exploit worked on its own hosts and treated it as critical. Fixed kernels shipped upstream on Tuesday, and DigitalOcean patched its fleet with a live patch.

What the flaw allowed

On a DigitalOcean Droplet, the first data returned was the host's kernel version banner, followed by about 127 GB of host RAM mapped read-only. A separate write test on the researchers' own bare-metal AMD server wrote 32 chosen bytes into a live page of the host kernel's memory, confirmed by a witness module loaded into that kernel. Both tests required root access inside a VM controlled by the researchers; an attacker would need the same position, not just network access. Intel-based KVM hosts are not affected, and many cloud providers surveyed were not vulnerable because they had not enabled nested virtualization on AMD.

How the bug worked

Terra's technical analysis describes a pointer left pointing at freed memory. KVM uses a nested page table to translate a VM's memory addresses into host memory, and the processor finds that table through a pointer kept for each virtual CPU. When an INIT reset stops a nested VM, KVM skips restoring the outer VM's pointer, leaving it pointing to a page the kernel has already freed. Once the kernel reused that page, the guest could fill it with entries naming any physical page on the host. The flaw was created by a 2022 patch for CVE-2022-3344, which switched a reset path back to the outer VM's control block but missed the page-table pointer; the error then sat in the kernel for just under four years.

Severity and scope

The Linux kernel's CVE team scores the flaw 8.8 out of 10, with high impact on confidentiality, integrity, and availability. Red Hat's preliminary assessment rates it Moderate at 5.6 and treats the impact as denial of service only. DigitalOcean treated it as critical for its shared cloud. The Linux KVM maintainers were told on October 21 and merged a fix on November 2. VirtFusion says there are no reports of the flaw being used outside research. No exploit code has been published, and no guidance exists for detecting a past attempt.

What to do

The fix lives in the host kernel; updating operating systems inside VMs changes nothing, and a host keeps running its old kernel until rebooted. Upstream, the fix shipped in stable kernel 7.3.5 and long-term kernels 6.18.64 and 6.12.118. The affected code entered the 6.1 kernel in 2022 and was carried into the 5.15.81 and 6.0.11 releases. Distribution kernels mostly do not have the fix yet: Debian's security tracker lists all releases as vulnerable with a fix only in unstable; Ubuntu's CVE tracker rates the flaw High and lists kernels as vulnerable with no fixed package; Red Hat lists Enterprise Linux 9 and 10 kernels as affected with the fix deferred, while Enterprise Linux 8 is not affected.

Until a fixed host kernel is installed and the host rebooted, the only protection is to turn nested virtualization off. That closes this path completely, but VMs can no longer run VMs of their own, so warn tenants who depend on it before making the change. Nested virtualization is on by default in the kernel's kvm_amd module, according to VirtFusion's operator guide, so an unpatched host is exposed unless the setting was deliberately changed. Check the setting with cat /sys/module/kvm_amd/parameters/nested; a value of 1 or Y means nested virtualization is on. If the host cannot be updated yet, add options kvm_amd nested=0 to the host's module configuration. Reboot the host. The setting cannot change while VMs are running, and a newly installed kernel does not protect the host until the reboot either. Confirm with uname -r and by reading the nested setting again.

Sources: The Hacker News

SiTech

SiTech — AI-powered web development

We build fast, modern websites and bring AI into real business workflows. Have a project or a question? We'd love to help.