Skip to main content
View

Attack ChainAIDerived

Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain

  1. Recon
    Gain privileged access inside Q35/KVM guest
  2. Delivery
    Issue VAPIC setup hypercall with out-of-window alias
  3. Exploit
    Map writable RAM alias over locked SMRAM
  4. Install
    Write past option ROM window bypassing D_LCK
  5. C2
    Inject code into SMM memory
  6. Execute
    Execute in System Management Mode
  7. Impact
    Compromise host firmware / escape guest isolation

Vulnerability Assessment

Weakness
CWE
ASVS 5.0
V15 - Secure Coding and Architecture Bounds checking, write limits
MITRE ATT&CK

2 techniques across 2 tactics: Execution, Privilege Escalation

The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up.

Exploitation Conditions

Requires an attacker with privileged (root-equivalent) access already inside a QEMU guest running specifically on a Q35 machine type with the KVM accelerator, where the VAPIC device path is in use. Exploitation is local to the guest (no network vector) and needs no user interaction on the host. Limiting factors: the attacker must first hold privileged guest access (PR:H), the target must be the Q35/KVM configuration rather than another machine type, and success depends on the guest being able to invoke the VAPIC setup hypercall - so this is a multi-tenant/hostile-tenant escape scenario, not a remote or unauthenticated exposure.

Impact

This is a genuinely high-severity issue whose real-world risk is bounded by a demanding precondition. The CVSS 3.1 vector (AV:L/AC:L/PR:H/UI:N/S:C/C:H/I:H/A:H, 8.2) is internally consistent with the description: exploitation is local to the guest (AV:L), requires a privileged guest user (PR:H), and crosses a security boundary (S:C) by escaping guest isolation into System Management Mode - the most privileged x86 execution context - yielding total confidentiality, integrity, and availability impact on the host/firmware. SSVC corroborates this: Technical Impact 'total', but Automatable 'no' (the attacker needs pre-existing privileged guest access and a specific chipset/emulation path, not a mass-scannable network entry point), and Exploitation only 'poc'. EPSS is low at 0.20% (10th percentile), and there is NO CISA KEV listing, so this is publicly-known-fragile rather than widely-exploited - priority should be driven by the total technical impact and multi-tenant blast radius, not by imminent mass exploitation. The one caveat to verify: SSVC marks exploitation as 'poc' but no public exploit URL is present in the provided references (only Red Hat/QEMU tracker links), so treat 'proof-of-concept exists' as vendor-asserted rather than independently confirmed here.

Exploit ScenarioDerived

A malicious or compromised root/privileged user inside a Q35/KVM guest issues the VAPIC setup hypercall with parameters that place the writable RAM alias outside the option ROM window, overlapping locked SMRAM. Because QEMU does not bounds-check the alias against the ROM window, the guest writes past the intended region into SMM memory despite chipset D_LCK protection, injecting code that later executes in System Management Mode. SSVC indicates proof-of-concept-level exploitation exists, though no public exploit link is in the provided references.

This Week Schedule remediation this week. Why: HIGH severity (CVSS 7.0-8.9).
Patch available 0.2% exploitation probability (EPSS)

Get an e-mail when this CVE is exploited, patched or re-scored. Free account.

Alert me

AnalysisAIv2 – Sep 21, 2026

Privileged users inside a QEMU guest running on the Q35 machine type with the KVM accelerator can abuse the VAPIC setup hypercall (hw/i386/vapic.c) to break SMM isolation, because the code fails to verify that a writable RAM alias stays inside the option ROM window. By repositioning that alias over chipset-locked SMRAM, an attacker with root-equivalent guest access defeats D_LCK protection and injects code into System Management Mode memory, which QEMU scores as a scope-changing, total-impact compromise (CVSS 8.2). Exploitation is local to the guest with high privileges required (AV:L, PR:H, S:C), publicly available exploit code exists according to the CISA SSVC assessment, but there is no confirmed active exploitation and EPSS is low at 0.20% (10th percentile).

Earlier versions of this analysis (1)
v1 Sep 18, 2026 - 11:20 CVSS 6.7

Privileged users inside a guest running on a Q35/KVM QEMU virtual machine can abuse the VAPIC setup hypercall in hw/i386/vapic.c to position a writable RAM alias over locked SMRAM, bypassing the Q35 chipset's D_LCK protection and injecting code into System Management Mode memory - the highest-privilege execution context on the platform. The flaw is an out-of-bounds write (CWE-787) scored 6.7 under CVSS:3.1/AV:L/AC:L/PR:H/UI:N/S:U/C:H/I:H/A:H: it is local to the guest, requires an already root-equivalent account there, and does not affect non-Q35 machine types or non-KVM configurations, which substantially narrows realistic exposure. No public exploit code was identified at time of analysis and there is no confirmed active exploitation, but the SMM-level payoff makes it attractive as the final step in a guest-to-firmware escalation chain.

RemediationAI

No vendor-released patch version is specified in the available data, so treat affected QEMU/KVM packages in the Red Hat Enterprise Linux 6/7/8/9/10, RHEL for NVIDIA 26, OpenShift Container Platform 4 and OpenStack 13 lines as unpatched until an erratum lands; monitor https://access.redhat.com/security/cve/CVE-2026-81627, https://bugzilla.redhat.com/show_bug.cgi?id=2536950 and the upstream fix at https://gitlab.com/qemu-project/qemu/-/work_items/4206, and apply the vendor update as soon as an exact fixed build is published. Until then, the most effective compensating control is to remove the attack surface: stop using the Q35 machine type with KVM for untrusted guests (substitute an i440fx/pc machine type) and avoid any VAPIC/option-ROM configuration that creates the writable alias, accepting that guests lose VAPIC interrupt-acceleration performance. Disabling SMM for guests that do not require it (-machine smm=off) eliminates the SMRAM target entirely, at the cost of breaking firmware features that depend on SMM, such as OVMF/UEFI secure-boot paths and certain TPM/ACPI flows. Because the vector is AV:L/PR:H, the strongest practical mitigation is restricting which tenants can obtain root inside guests on the same host - do not co-locate mutually untrusted tenants, or tenants needing guest root, on hypervisors whose firmware relies on SMM/D_LCK for secrets; note that the exploit is not automatable per SSVC and no active exploitation has been confirmed, so these trade-offs can be prioritized against the real precondition of privileged guest access.

Affects your stack

Find out whether the stacks and servers you run are on an affected version.

Checking this CVE against your own stacks and servers is part of the Basic plan and up.

Technical ContextAI · show details

QEMU's hw/i386/vapic.c implements the paravirtualized VAPIC option ROM used to accelerate interrupt delivery for guests, and it relies on the core memory API to alias a writable RAM region over the ROM window at a guest-controlled address. The vulnerability is an out-of-bounds write (CWE-787): no bounds check confirms that the requested alias lies within the option ROM area, so the memory-region aliasing can be pointed at SMRAM, the System Management RAM reserved for System Management Mode. On Intel-derived Q35 chipsets SMRAM is protected by the D_LCK bit, which normally locks the SMM base/control registers so that even firmware cannot remap SMM memory; VAPIC's SMM bypass defeats that lock, allowing code to be written into SMM memory. The flaw is reachable only through a path that requires the Q35 machine type, the KVM accelerator, and the VAPIC device path being exercised by a guest that already has privileged (root-equivalent) access inside the VM, making it a hostile-tenant/VM-escape class issue rather than a remote or unauthenticated exposure.

Know when Qemu gets its next vulnerability

  • Watch products and get an email when something new is published
  • Check this CVE against your own stack and servers
  • Exploit and PoC references, threat intelligence from 22 sources

No credit card · 7-day full trial

Severity by source

Vendor (redhat) PRIMARY 8.2 MEDIUM
AV:L/AC:L/PR:H/UI:N/S:C/C:H/I:H/A:H
vuln.today AI 8.2 HIGH

Local guest vector (AV:L) needing privileged guest access (PR:H); escaping guest into host SMM is a scope change (S:C) with total host firmware compromise (C:H/I:H/A:H).

3.1 AV:L/AC:L/PR:H/UI:N/S:C/C:H/I:H/A:H
4.0 AV:L/AC:L/AT:N/PR:H/UI:N/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H
SUSE HIGH
Red Hat 6.7 MEDIUM

Primary rating from Vendor (redhat).

CVSS VectorVendor: redhat

Attack Vector
Local
Attack Complexity
Low
Privileges Required
High
User Interaction
None
Scope
Changed
Confidentiality
High
Integrity
High
Availability
High

Lifecycle Timeline

4
Analysis Updated
Sep 21, 2026 - 12:28 vuln.today
v2 - CVSS changed
Severity Changed
Sep 21, 2026 - 12:22 NVD
MEDIUM HIGH
CVSS changed
Sep 21, 2026 - 12:22 NVD
6.7 (MEDIUM) 8.2 (HIGH)
CVE Published
Sep 18, 2026 - 10:54 cve.org
MEDIUM 6.7

DescriptionCVE.org

A flaw was found in QEMU. The VAPIC setup hypercall in hw/i386/vapic.c does not validate that the writable RAM alias remains within the option ROM window. A privileged guest user on a Q35/KVM machine can position this alias over locked SMRAM, bypassing chipset D_LCK protection and injecting code into System Management Mode memory.

Executive ViewAI

Why It Matters

CVE-2026-81627 affects QEMU deployments using the Q35 machine type with KVM acceleration, where an authenticated, high-privileged guest user can break SMM isolation and undermine platform integrity assumptions. Organizations running multi-tenant or security-sensitive virtualization should treat this as a high-priority isolation risk because publicly available exploit code exists and no vendor patch is currently available, even though active exploitation has not been confirmed.

Business Impact: High
Compensating Controls

No vendor-released patch identified at time of analysis. Interim mitigations include restricting root-equivalent guest access to trusted users only, disabling SMM for workloads that do not require it, avoiding the Q35 machine type with KVM where feasible, isolating or segmenting high-risk guests, and monitoring guest and hypervisor logs for anomalous VAPIC setup hypercalls or SMM memory modifications.

Threat Intelligence Overview

EPSS Score
0.2%
Top 90% exploitable
Prevalence
High
QEMU/KVM is a foundational hypervisor across cloud and enterprise virtualization
Data Quality
7 sources
Confidence: High
EUVD NVD CVE.org EPSS Vulnrichment MSRC Vendor VulDB

How this decision was reached

The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.

SignalValue
Severity HIGH
CISA KEV No
Public POC No
EPSS 0.2%
Patch available Yes
Active threat No
Deciding rule
HIGH severity (CVSS 7.0-8.9)
What every action level means
Emergency Immediate action required. Criteria: CRITICAL + in CISA KEV (actively exploited) without patch, OR CRITICAL + EPSS > 30% without patch, OR CRITICAL + active threat signals without patch.
Act Now Prioritize within 24-48h. Criteria: in KEV with patch available, OR CRITICAL + public POC, OR HIGH + EPSS > 10%, OR CRITICAL + active threat, OR any CRITICAL severity.
This Week Schedule remediation this week. Criteria: HIGH severity (CVSS 7.0-8.9). Patch within normal change window.
This Month Plan within 30 days. Criteria: MEDIUM severity (CVSS 4.0-6.9), lower exploitation probability. Include in regular patch cycle.
Monitor Track and reassess. Criteria: LOW severity (CVSS 0.1-3.9), minimal exploitation risk. No immediate action needed.

SSVC AssessmentENISA

SSVC Decision
Scheduled
Exploitation
Poc
System Exposure· derived
Small
Automatable
No
Human Impact
High
Technical Impact
Total

Priority Score

41
Low Medium High Critical
KEV: 0
EPSS: 0
CVSS: +41
PoC: 0
Weaponized: 0
Threat intel: 0
Gov advisory: 0

Attack Technique

A buffer overflow occurs when a program writes more data to a memory buffer than it was allocated to hold, causing the excess data to spill into adjacent memory regions.

Memory corruption occurs when a program writes data beyond the boundaries of allocated memory regions or accesses memory in unintended ways, violating the integrity of the process's address space.

Affected Products (CPE)CVE.org

cpe:2.3:a:red_hat:red_hat_enterprise_linux_10:*:*:*:*:*:*:*:*
cpe:2.3:a:red_hat:red_hat_enterprise_linux_6:*:*:*:*:*:*:*:*
cpe:2.3:a:red_hat:red_hat_enterprise_linux_7:*:*:*:*:*:*:*:*
cpe:2.3:a:red_hat:red_hat_enterprise_linux_7:*:*:*:*:*:*:*:*
cpe:2.3:a:red_hat:red_hat_enterprise_linux_8:*:*:*:*:*:*:*:*
cpe:2.3:a:red_hat:red_hat_enterprise_linux_8:*:*:*:*:*:*:*:*
cpe:2.3:a:red_hat:red_hat_enterprise_linux_9:*:*:*:*:*:*:*:*
cpe:2.3:a:red_hat:red_hat_enterprise_linux_for_nvidia_26:*:*:*:*:*:*:*:*
cpe:2.3:a:red_hat:red_hat_openshift_container_platform_4:*:*:*:*:*:*:*:*
cpe:2.3:a:red_hat:red_hat_openshift_container_platform_4:*:*:*:*:*:*:*:*
cpe:2.3:a:red_hat:red_hat_openstack_platform_13_(queens):*:*:*:*:*:*:*:*

More in Qemu

View all
CVE-2015-3456 HIGH POC
7.7 May 13

The Floppy Disk Controller (FDC) in QEMU, as used in Xen 4.5.x and earlier and KVM, allows local guest users to cause a

CVE-2015-8556 CRITICAL POC
10.0 Mar 24

Local privilege escalation vulnerability in the Gentoo QEMU package before 2.5.0-r1. Rated critical severity (CVSS 10.0)

CVE-2019-12928 CRITICAL POC
9.8 Jun 24

The QMP migrate command in QEMU version 4.0.0 and earlier is vulnerable to OS command injection, which allows the remote

CVE-2019-12929 CRITICAL POC
9.8 Jun 24

The QMP guest_exec command in QEMU 4.0.0 and earlier is prone to OS command injection, which allows the attacker to achi

CVE-2015-7512 CRITICAL
9.0 Jan 08

Buffer overflow in the pcnet_receive function in hw/net/pcnet.c in QEMU, when a guest NIC has a larger MTU, allows remot

CVE-2016-7161 CRITICAL
9.8 Oct 05

Heap-based buffer overflow in the .receive callback of xlnx.xps-ethernetlite in QEMU (aka Quick Emulator) allows attacke

CVE-2024-24474 HIGH POC
8.8 Feb 20

QEMU before 8.2.0 has an integer underflow, and resultant buffer overflow, via a TI command when an expected non-DMA tra

CVE-2022-35414 HIGH POC
8.8 Jul 11

softmmu/physmem.c in QEMU through 7.0.0 can perform an uninitialized read on the translate_fail path, leading to an io_r

CVE-2016-4002 CRITICAL
9.8 Apr 26

Buffer overflow in the mipsnet_receive function in hw/net/mipsnet.c in QEMU, when the guest NIC is configured to accept

CVE-2024-3447 MEDIUM POC
6.0 Nov 14

A heap-based buffer overflow was found in the SDHCI device emulation of QEMU. Rated medium severity (CVSS 6.0), this vul

CVE-2017-15268 HIGH POC
7.5 Oct 12

Qemu through 2.10.0 allows remote attackers to cause a memory leak by triggering slow data-channel read operations, rela

CVE-2022-2962 HIGH POC
7.8 Sep 13

A DMA reentrancy issue was found in the Tulip device emulation in QEMU. Rated high severity (CVSS 7.8), this vulnerabili

Vendor StatusVendor

SUSE

Severity: Important
Product Status
openSUSE Tumbleweed Fixed
SUSE Linux Enterprise Server 16.0 Affected
SUSE Linux Enterprise Server for SAP applications 16.0 Affected
SUSE Linux Micro 6.2 Affected
SUSE Linux Enterprise Desktop 15 SP7 Affected

This site uses cookies essential for authentication and security. No tracking or analytics cookies are used. Privacy Policy