KVM flaw lets AMD guests escape to the host (CVE-2026-99312)
Researchers read about 127 GB of host RAM from a DigitalOcean droplet; a host write was shown only on their own server.
Stale KVM MMU context on nested SVM INIT
Linux KVM on x86 can keep a stale MMU context when it emulates INIT while a nested SVM guest (L2) runs with paging disabled (CR0.PG=0), then continues L1 with the wrong page role. A guest that can cause that INIT path can trigger the bug; Intel VMX is not affected because INIT is always intercepted there. A related report describes this as a guest-to-host escape. Affected kernel builds are not listed.
What to do: Apply the Linux kernel update that includes the fix for CVE-2026-99312; no fixed version or workaround is stated.
| Product | Affected |
|---|---|
| Linux kernel | — |
Order-of-magnitude estimate by the model from install counts, market share and public scan data it knows; verify before quoting.
In the Linux kernel, the following vulnerability has been resolved: KVM: x86: Unconditionally reset MMU context when emulating INIT Force an MMU reset when emulating an INIT even if paging was already disabled, as the MMU page role includes guest_mode (L2), i.e. an INIT while L2 is running with CR0.PG=0 will result in KVM running L1 with a stale MMU context. Alternatively, the reset could be scoped strictly to paging being enabled or L2 being active, but SMM is also encoded in the page role. Even though it _should_ be impossible for the vCPU to be INIT'd while SMM is active, there's zero value in not handling SMM. At that point, the only scenario where KVM _doesn't_ need to reset the MMU context is on an INIT with the vCPU running in Real Mode or Unpaged Protect Mode, and while INIT in those modes is somewhat common, it's far from a hot path and re-initializing the MMU is not expensive, especially when the role isn't changing (KVM elides most of the re-configuration in that case). Note, like the Fixes commit, this bug only affects nested SVM as INIT is an unconditional intercept on VMX, i.e. it's impossible to emulate INIT while L2 is active. Temporarily keep the guest TLB flush when CR0.PG is being cleared even though it is fully redundant with the flush that is emitted on all INIT events (KVM itself only does RESET when a vCPU is being created, i.e. when CR0.PG must be '0').
Researchers read about 127 GB of host RAM from a DigitalOcean droplet; a host write was shown only on their own server.
kernel.org says the bug affects nested SVM only, because INIT is an unconditional intercept on VMX.
Today 2.7K readers · 3K views · 145 reads · all time 207K views