From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-97987][MODERATE REGULAR] virtio_input: reset device if input_register_device() fails [ Upstream Date: Fri, 25 Sep 2026 10:54:31 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-97987 X-AL-KERNEL-Priority: MODERATE REGULAR X-AL-KERNEL-Severity: MODERATE REGULAR X-AL-KERNEL-Base-Severity: MODERATE X-AL-KERNEL-KPANIC: NO X-AL-KERNEL-ActionableScore: 3 X-AL-KERNEL-ActionableScore-Lower: unknown X-AL-KERNEL-Commit: 9a7e107734137f098f9b81e5615aa5a02c5395c7 List-Id: CVE: CVE-2026-97987 Priority: MODERATE REGULAR AL-KERNEL base severity: MODERATE KPANIC flag: NO Patch: virtio_input: reset device if input_register_device() fails [ Upstream Commit: 9a7e107734137f098f9b81e5615aa5a02c5395c7 Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=9a7e107734137f098f9b81e5615aa5a02c5395c7 Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97987 Analysis date: Fri, 25 Sep 2026 10:54:31 -0400 ActionableScore: 3 ActionableScore lower bound: unknown Actionable bucket: Borderline / manual review recommended Manual review required: YES Summary: No one-sentence report phrase was found. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2026-97987 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97987 The original announcement does not normally provide a security severity estimate, CVSS assessment, or enough information to determine whether the reported kernel bug represents a practically relevant security issue. This report was generated by AL-KERNEL, an AI-assisted Linux kernel vulnerability analysis system developed by Alexander Larkin. It combines an autonomous classifier with LLM-assisted technical analysis and a separate ActionableScore mechanism. The purpose of this report is to prioritize Linux kernel CVEs before manual review, identify cases that require prompt investigation, and support automatic closure of issues that are unlikely to have meaningful security impact. Published priority for this report: MODERATE REGULAR Manual review required: YES A detailed explanation of the methodology and priority rules is included at the end of this message. ====================================================================== AL-KERNEL CLASSIFICATION RESULT ====================================================================== CVE-2026-97987 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-97987 CHECK WITH IMPACT FROM ORIG NN LOW Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:A/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H';*CWE-416;*CWE-664;*CWE-672;Other CVSS 'AV:A/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '7.5';DESCR 'A virtio input device can remain active after input_register_device() fails because the error path tears down virtqueues without first resetting a device that was already marked DRIVER_OK. The device may continue DMA operations through stale queue state after the queue resources have been freed or reused, creating a DMA based use after free condition in the guest kernel. For the CVSS the PR:N rating is used because a hostile virtio backend or adjacent virtio device does not require an account or local code execution in the affected guest. The attack vector is AV:A because the issue is reached through the virtio device interface and is not directly network reachable. Impact is at least guest kernel denial of service. In the worst case stale DMA to reused memory may provide confidentiality or integrity impact and potential privilege escalation. This issue should be reviewed manually.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline / manual review recommended (with actual score 3) YES VIRT SIMPLEFIX ERRORPATH DMAorINTERRUPT HARDWARE LINUS INCREASED_FROM_LOW_BASED_ON_REQUIREMANUALCHECK DECREASED_TO_MODERATEREG_BASED_ON_ACTIONABLESCORELESSTHAN4 NO NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=3 PrimitiveClass=CLASS_B RealUAF=YES HostAdminRequired=UNKNOWN NamespaceScopedPrivilege=UNKNOWN ImportantCandidate=NO ManualReview=RECOMMENDED AutoDowngradeAllowed=YES 1. ActionableScore * Raw conservative score: 3 * Conservative score: 3 * Raw paranoid score: 3 * Paranoid score: 3 * Final recommended bucket: **Borderline / manual review recommended**:: No floor was applied. This is a fixed async DMA stale-storage case, not a demonstrated active-object reuse, controlled reclaim, or repeated refcount mutation case. 2. Signal breakdown * +2 Real UAF or equivalent lifetime violation. The device remains DRIVER_OK and can continue DMA after `del_vqs()` tears down the virtqueues. This is a concrete post-lifetime access to queue-associated resources. * +1 Constrained kernel buffer overwrite. A live virtio input device may DMA device-generated input data through stale queue state after teardown. The affected DMA targets are previously supplied queue buffers or descriptors, not attacker-selected arbitrary kernel addresses. Important non-awarded signals: * +0 Controlled reclaim or controlled object replacement. No evidence shows that an attacker can place chosen kernel objects at a reused stale DMA target. * +0 Arbitrary or directed kernel write. The patch establishes stale DMA, but not arbitrary DMA address selection after queue teardown. * +0 Strong exploitation leverage. No callback, function pointer, credential, refcount takeover, active-object substitution, or attacker-controlled post-free structured-field consumption is shown. * +0 Reliable kernel crash. The commit identifies unsafe DMA after queue teardown, but does not provide an observed Oops, panic, or reproducible host-wide crash. * +0 Remote reachability. The relevant input is supplied through an adjacent virtio device or backend, not directly through a network protocol. * +0 Local unprivileged trigger. The supplied evidence does not establish that an ordinary guest user can control the virtio backend or reliably force `input_register_device()` to fail. * +0 Weak LPE concern. Generic theoretical reuse of DMA buffers is insufficient. * +0 Confidentiality or integrity impact. The stale DMA condition can potentially corrupt reused memory, but no read-back, protected-data disclosure, unauthorized modification target, or practical privilege escalation chain is established. * -0 Host-admin penalty. Host-admin-equivalent privilege is not positively established, so no negative privilege score is applied. * -0 Hard-race penalty. This is an ordered error-path lifetime failure, not a demonstrated narrow and uncontrollable timing race. * -0 niche-subsystem penalty. Virtio use alone does not prove that the issue is rare or unreachable. 3. Reachability analysis The relevant attacker model is a hostile or malfunctioning virtio input backend adjacent to the guest. The backend may continue processing previously visible virtqueue state after the guest driver destroys the queues without resetting the device. The complete trigger also requires `input_register_device()` to fail after `virtio_device_ready()` has marked the device DRIVER_OK. The supplied patch does not establish that a guest user, a guest administrator, or a hostile backend can reliably cause this registration failure. It may depend on resource allocation failure or another probe-time error. HostAdminRequired is UNKNOWN. The evidence does not prove that control of the relevant virtio backend requires host administrator privileges, but it also does not prove that a below-host-admin attacker can perform the entire sequence. NamespaceScopedPrivilege is UNKNOWN because no namespace-based capability path is relevant or established. Virtio requires a virtualized or paravirtualized deployment, but no rare physical hardware is required. The issue is not directly network reachable. A network attacker would need a separate path to compromise or control the virtio backend, which is outside the supplied evidence. 4. Primitive and severity interpretation The exact primitive is a DMA lifetime violation. The guest tears down virtqueues while the virtio input device remains live, allowing later device DMA through stale queue resources. This is fixed async stale storage rather than an active structured-object UAF. The patch does not show a stale kernel object being dereferenced through attacker-controlled fields after reuse. It also does not show a second `put`, `free`, refcount decrement, callback dispatch, or other repeated lifetime mutation through the same dangling object. The issue is therefore CLASS_B because a real post-lifetime DMA access exists. It is not CLASS_A because controlled reclaim, directed DMA, arbitrary read/write, controlled type confusion, or a demonstrated guest-to-host boundary compromise is not established. The demonstrated practical consequence is unsafe memory access and a plausible guest kernel crash or corruption condition if stale queue memory is reused. The speculative consequence is broader guest-kernel confidentiality, integrity, or privilege impact if a hostile backend can preserve descriptors and target useful reused memory. That additional mechanism is not established by the patch. ImportantCandidate is NO because the available evidence supports a constrained stale DMA path but not meaningful exploitation leverage beyond the lifetime violation itself. AutoDowngradeAllowed is YES because this is currently a constrained fixed-storage DMA case with no demonstrated controlled reuse or active-object manipulation. Likely operational severity is Borderline Moderate. It should receive manual validation when the deployment uses untrusted or externally managed virtio backends, but the patch alone does not justify Important classification. 5. One-sentence report phrase A virtio input device may continue DMA through stale virtqueue resources after probe failure and queue teardown, creating a constrained guest kernel use after free condition that requires manual review but has no demonstrated controlled corruption primitive. 6. Manual review recommendation MANUAL CHECK RECOMMENDED The patch establishes live-device DMA after virtqueue teardown, but the attacker ability to control the virtio backend, reliably induce `input_register_device()` failure, and exploit reused queue memory is unresolved. 7. Call-site confidence Call-site confidence: medium ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: virtio_input: reset device if input_register_device() fails [ Upstream Commit: 9a7e107734137f098f9b81e5615aa5a02c5395c7 Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=9a7e107734137f098f9b81e5615aa5a02c5395c7 Changed files: drivers/virtio/virtio_input.c Diff excerpt: Not included in this email. See the upstream URL for the full patch. Full patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=9a7e107734137f098f9b81e5615aa5a02c5395c7 ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-97987 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97987 The original CVE announcement normally does not include a security-level estimate. In particular, it may not contain a CVSS assessment, an impact level, or enough information to determine whether the reported bug is a practically relevant security issue. One purpose of this parallel CVE list is to provide that missing technical and prioritization information. The original goal of the AL-KERNEL project was to prioritize Linux kernel CVE analysis automatically before manual review. The system can also help identify non-security issues that may be suitable for automatic closure. This report was generated by AL-KERNEL, an AI-assisted Linux kernel vulnerability analysis system developed by Alexander Larkin. The first analysis stage combines an autonomous classifier with additional LLM-based analysis. The autonomous classifier runs locally on a CPU and is based on a backpropagation neural network. Together, these mechanisms produce a technical vulnerability description, identify likely weakness types, estimate CVSS severity, and provide input for ActionableScore. Two CVSS estimates are retained because incomplete kernel vulnerability information often permits more than one defensible interpretation: Conservative CVSS vector: AV:A/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:H The Best / paranoid CVSS vector: AV:A/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H The Best / paranoid CVSS score: 7.5 The conservative vector represents a lower-impact interpretation. The Best/paranoid vector intentionally represents a plausible upper-bound interpretation and should not automatically be treated as demonstrated real-world impact. CVSS may also need to be adjusted for a particular Linux deployment, because actual reachability, privileges, enabled kernel configuration, hardware, namespaces, exposed device nodes, and other environmental conditions can differ significantly between systems. A separate ActionableScore mechanism evaluates practical remediation urgency. Its analysis may include reachability, attack prerequisites, subsystem exposure, memory-corruption characteristics, denial-of-service reliability, and possible confidentiality, integrity, or privilege-escalation impact. Conservative ActionableScore: unknown Paranoid ActionableScore: 3 The final base severity is taken directly from the second tab-separated field of the AL-KERNEL classification result. ActionableScore does not replace or independently override that final AL-KERNEL decision, and if ActionableScore adjusted impact level of ALKERNEL, then you would see self-readable flags above like INCREASED_TO_HIGH_BASED_ON_ACTIONABLESCOREHIGHEREQTHAN7. For an AL-KERNEL result of MODERATE, this report uses the following additional presentation split: ActionableScore below 5 -> MODERATE REGULAR ActionableScore 5 or more -> MODERATE 7.0 The distinction between MODERATE REGULAR and MODERATE 7.0 makes it possible to identify Moderate issues that should receive manual analysis and fixes before lower-priority MODERATE REGULAR issues. In many cases, MODERATE REGULAR fixes may wait for a later rebase or routine update. There is one override in which MODERATE REGULAR becomes MODERATE 7.0 even when the ActionableScore is below 5. When the AL-KERNEL result contains the KPANIC flag, a MODERATE result is always presented as MODERATE 7.0. The KPANIC flag selected with few regexps without usage of AI at all, so it helps to detect cases when Kernel Crash happens and similar (to filter False-Negative results from the LLM usage). KPANIC indicates that a reliable kernel crash, kernel panic, or similarly serious kernel availability impact was identified by the classification workflow. AL-KERNEL base severity for this report: MODERATE KPANIC detected for this report: NO Published priority for this report (same as in Subject): MODERATE REGULAR These results are intended to support engineering triage. They are machine-generated estimates, and cases marked for manual review should be validated by a human security engineer before final disposition. For more info read docs linked from here: https://kernelcve.org/ (and you can submit you own patch there to generate such a report for non-existant CVE-id yet). Note that in many cases this AI tool selects higher severity, than real is (means you can expect Importants instead of Moderate 7.0 or Moderates 7.0 instead of regular Moderates). If you see such cases, please use reply email interface to add additional manual analyses info to this particular CVE. And please, please, let me know when you see Lows instead of Importants or Important instead of Low (because particular for such cases I need to tune this AI tool to make it better for this one and next similar). My contact email for such notifications is alexanjelausa@gmail.com (and both send reply to CVE record itself too and see "reply" button below for howto reply).