CVE-2021-47112: x86/kvm: Teardown PV features on boot CPU as well
In the Linux kernel, the following vulnerability has been resolved:
x86/kvm: Teardown PV features on boot CPU as well
Various PV features (Async PF, PV EOI, steal time) work through memory
shared with hypervisor and when we restore from hibernation we must
properly teardown all these features to make sure hypervisor doesn't
write to stale locations after we jump to the previously hibernated kernel
(which can try to place anything there). For secondary CPUs the job is
already done by kvm_cpu_down_prepare(), register syscore ops to do
the same for boot CPU.
Security readout for executives and security teams
Plain-English summary
This Linux kernel issue affects KVM paravirtualization cleanup during hibernation restore. Some shared memory links with the hypervisor were not torn down for the boot CPU, creating a risk that the hypervisor writes to stale memory after resume. Business urgency depends on whether affected Linux kernels use KVM virtualization and hibernation.
Executive priority
Handle through normal kernel patch governance unless your environment uses KVM with hibernation workflows. There is no cited active exploitation or severity score, but kernel memory-state bugs can have high impact if exposure conditions exist.
Technical view
The fix registers syscore operations so x86 KVM paravirtual features, including async page fault, PV EOI, and steal time, are torn down for the boot CPU during hibernation restore. Secondary CPUs already used kvm_cpu_down_prepare(). The source record lists affected Linux kernel ranges and stable commit references, but provides no CVSS or CWE.
Likely exposure
Exposure appears limited to affected Linux kernels in KVM environments where hibernation restore is relevant. The provided data does not identify specific distributions, cloud images, or default configurations, so asset owners must map deployed kernel packages to vendor-fixed builds.
Exploitation context
The supplied sources do not report active exploitation, and the CVE is not listed as KEV. The issue is a kernel state-management flaw around stale hypervisor-shared memory after hibernation restore, not a documented remote exploit path in the provided evidence.
Researcher notes
Focus validation on x86 KVM PV feature teardown across hibernation restore. The key behavior is ensuring boot CPU cleanup matches secondary CPU cleanup for async PF, PV EOI, and steal time shared memory. Evidence is commit-level; product and exploit details are incomplete.
Mitigation direction
Update affected Linux kernels to versions containing the referenced stable fixes.
Use distribution advisories to map installed kernel packages to fixed builds.
Prioritize KVM virtualized systems where hibernation or resume workflows are enabled.
If patch status is unclear, request vendor guidance for CVE-2021-47112.
Validation and detection
Inventory Linux kernel versions across KVM guests and virtualization hosts.
Check whether deployed kernel builds include the listed stable commits.
Confirm whether hibernation or resume-from-hibernation is enabled in affected environments.
Track vendor security advisories for package-specific fixed versions.
Generated from the cited source records. This long-tail analysis has not been individually reviewed by a named human.
Potential ATT&CK relevance
Conservative CVE-to-ATT&CK context
These mappings and lookup hints may be relevant to the vulnerability behavior, CWE, affected product, or exposure path. Glexia-inferred context is not an official MITRE, ATT&CK, CWE, or CVE Program mapping.
ATT&CK lookup starting points
Use these exact CWE pages and searches to review the Glexia ATT&CK library from this CVE's weakness and description context.
cve · low confidence lookup
CVE-2021-47112 mapping review
Open the CVE-to-ATT&CK bridge for reviewed, inferred, or future official mappings tied to this CVE.
These fields come from the CVE record and ADP containers, not from Glexia's Take. They preserve time-varying source decisions such as CISA SSVC, KEV status, CVSS metrics, and provider references.