From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-89823][MODERATE 7.0] drm: fix race between partial drm_dev_register() failure and ioctl Date: Wed, 16 Sep 2026 09:51:10 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-89823 X-AL-KERNEL-Priority: MODERATE 7.0 X-AL-KERNEL-Severity: MODERATE 7.0 X-AL-KERNEL-Base-Severity: MODERATE X-AL-KERNEL-KPANIC: YES X-AL-KERNEL-ActionableScore: 5 X-AL-KERNEL-ActionableScore-Lower: 4 X-AL-KERNEL-Commit: 26d9162213952927cd59c0d94b7e0e992b347bec List-Id: CVE: CVE-2026-89823 Priority: MODERATE 7.0 AL-KERNEL base severity: MODERATE KPANIC flag: YES Patch: drm: fix race between partial drm_dev_register() failure and ioctl Commit: 26d9162213952927cd59c0d94b7e0e992b347bec Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=26d9162213952927cd59c0d94b7e0e992b347bec Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89823 Analysis date: Wed, 16 Sep 2026 09:51:10 -0400 ActionableScore: 5 ActionableScore lower bound: 4 Actionable bucket: Actionable Moderate / Strong Important candidate requiring manual review Manual review required: YES Summary: No one-sentence report phrase was found. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2026-89823 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89823 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 7.0 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-2026-89823 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-89823 CHECK WITH IMPACT FROM ORIG NN LOW Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H';CWE-362;*CWE-416;*CWE-664;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '7';DESCR 'A race in drm_dev_register can occur when registration fails after at least one DRM minor has already been registered. A local userspace process may open that minor and enter a drm_dev_enter critical section while the error path starts tearing down device resources. Because the unplugged flag was not set on this failure path, an ioctl may continue to run against a device that is being cleaned up. This creates a plausible use after free style condition rather than a simple warning or pure resource leak. For the CVSS the PR:L value is used because access to DRM render nodes is commonly delegated to local users through render or video group membership or desktop session policy. The issue is not network reachable and requires local access to the DRM device node. Impact is at least local denial of service via kernel crash and in worst case may allow confidentiality or integrity impact due to memory corruption, so it should be reviewed manually.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 5) YES INIT RACE ERRORPATH HARDWARE LINUS KPANIC INCREASED_TO_MODERATE7_FROM_LOW_BASED_ON_GUESSCVSS_AND_ACTIONABLESCOREHIGHER4 NO NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=5 ActionableScoreLower=4 PrimitiveClass=CLASS_B RealUAF=YES HostAdminRequired=UNKNOWN NamespaceScopedPrivilege=NO ImportantCandidate=YES ManualReview=REQUIRED AutoDowngradeAllowed=NO 1. ActionableScore * Raw conservative score: 2 * Conservative score: 4 * Applied conservative floor/cap: Real UAF floor 4 applied because an ioctl may continue while DRM device resources are being torn down, and host-admin-equivalent reachability is not positively established. * Raw paranoid score: 2 * Paranoid score: 5 * Applied paranoid floor/cap: Real UAF floor 5 applied under the stronger still-defensible interpretation that stale ioctl execution against an actively tearing-down DRM device is meaningful active-object exploitation surface beyond an ordinary cleanup-only fault. * Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**:: 2. Signal breakdown Conservative signals: * +2 Real UAF / active lifetime violation. A DRM minor may be registered and opened by userspace, then an ioctl may enter a drm_dev_enter critical section while the drm_dev_register error path tears down device resources without first setting unplugged and waiting for SRCU readers. * +1 Reliable kernel crash / strong DoS. Continuing ioctl execution during device resource teardown realistically supports kernel fault or crash. * -1 Hard or unreliable race / special timing. The window requires partial drm_dev_register failure after at least one minor is published and concurrent userspace entry before cleanup synchronization. * Raw conservative score: 2. * Conservative floor 4 applied. Paranoid signals: * +2 Real UAF / active lifetime violation. Same as conservative. * +1 Reliable kernel crash / strong DoS. Same as conservative. * -1 Hard or unreliable race / special timing. Same as conservative. * Raw paranoid score: 2. * Paranoid floor 5 applied because the stale use is through active ioctl processing on a DRM device, not merely a passive diagnostic path. Important non-awarded signals: * Controlled reclaim not demonstrated. * Arbitrary or directed kernel write not demonstrated. * Concrete repeated release/refcount mutation not established. * Structured stale-object leverage was not separately awarded as a raw point because the patch does not show a specific stale field, callback, pointer, or replacement object being attacker-controlled. * Weak LPE was not awarded merely because this is a UAF-style race. * C/I impact was not awarded as an ActionableScore signal because confidentiality or integrity consequences are not concretely established beyond unresolved memory-corruption concern. * Extra Availability was not awarded because the same crash/DoS is already counted once. * Remote reachability is not established. Repeated-release/refcount check: * First release can end lifetime: not established as a concrete first put/release of the same referenced object. * Exact same reference survives to second release: not established. * Second operation mutates lifetime/refcount: not established. * +1 Concrete repeated release/refcount awarded: no. * Paranoid repeated-refcount leverage +1 awarded: no. * FLOOR 5 applied: yes, but due active ioctl stale-use surface, not repeated release/refcount. 3. Reachability analysis The vulnerable exposure begins when drm_dev_register partially succeeds and publishes at least one DRM minor, such as a render minor, before a later registration step fails. A local userspace process can potentially open that minor and issue an ioctl while registration failure cleanup is in progress. HostAdminRequired is UNKNOWN because the supplied evidence establishes userspace entry through an already published minor, but does not fully establish that a below-host-admin attacker can reliably cause the partial registration failure itself. It would be too strong to mark HostAdminRequired=YES because host-admin-equivalent privilege is not positively proven. It would also be too strong to mark HostAdminRequired=NO because complete below-host-admin control of the full trigger chain is not proven. NamespaceScopedPrivilege is NO because no namespace-scoped capability path is involved in the supplied evidence. DRM/KMS actor distinction: * Ordinary logged-in user: may have access to render nodes on many systems, but complete trigger control is not fully proven here. * DRM master / compositor / display server: may have stronger access to DRM ioctls depending on the minor and device policy. * Delegated graphics service: plausible practical actor if it can open the affected minor during probe failure. * Host administrator: can often influence driver load, probe, module lifecycle, or device binding, but requiring host admin for exploitation is not established by the supplied patch. * Network attacker: not applicable. Prior host setup may exist in practice because a DRM driver and device must be present, and the driver registration must fail after a minor is registered. This is a timing and state prerequisite, not proof of host-admin-only reachability. Special hardware exists only in the ordinary sense that DRM hardware or a DRM driver is involved. No rare or physical-only hardware penalty is applied because common GPU/display DRM hardware is not automatically rare. The race is likely narrow because it depends on partial registration failure and concurrent ioctl entry, so a timing penalty is applied. 4. Primitive and severity interpretation The exact primitive is an SRCU-protected lifetime/synchronization failure around drm_dev_enter and DRM device teardown. A userspace ioctl can proceed after observing the device as not unplugged while the error path cleans up resources following partial registration failure. This is an active structured object case, not fixed stale storage. The stale path is an ioctl path operating on a DRM device and driver resources, not merely a fixed DMA buffer or passive metadata. However, the supplied patch does not identify a specific controlled reclaimed object, callback, function pointer, arbitrary write, or disclosure path. This is not a concrete repeated release/refcount case. The patch fixes missing unplug synchronization before cleanup, not a visible second put/free on the exact same dangling reference. PrimitiveClass is CLASS_B because a real active lifetime violation is directly implied, but strong exploitation leverage is unresolved rather than demonstrated. Demonstrated consequence is at least local denial of service through crash or unsafe ioctl execution during teardown. Speculative consequence is broader memory corruption or privilege escalation if a driver-specific ioctl path can consume freed resources in an exploitable way. That stronger consequence is plausible enough for manual review, but not established enough for CLASS_A. ImportantCandidate is YES because an active-object UAF-style DRM ioctl race during resource teardown has meaningful unresolved exploitation leverage and should not be automatically downgraded from a small patch alone. AutoDowngradeAllowed is NO because automatic closure would discard an unresolved active lifetime/corruption primitive in a userspace-reachable DRM ioctl path. Likely operational severity is Actionable Moderate at minimum, with Strong Important candidate review warranted depending on which DRM driver ioctl paths can run in the stale window and what resources are freed before synchronization. 5. One-sentence report phrase A race in drm_dev_register can allow a local userspace ioctl to continue through drm_dev_enter while the partial registration failure path tears down DRM device resources, creating a UAF-style active-object lifetime hazard with likely DoS and unresolved memory-corruption impact. 6. Manual review recommendation MANUAL CHECK REQUIRED Manual review is required because the patch establishes an active DRM ioctl versus device teardown lifetime race, and driver-specific ioctl paths must be inspected to determine whether the stale execution is crash-only or can produce stronger memory corruption or privilege-escalation leverage. 7. Call-site confidence Call-site confidence: medium ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: drm: fix race between partial drm_dev_register() failure and ioctl Commit: 26d9162213952927cd59c0d94b7e0e992b347bec Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=26d9162213952927cd59c0d94b7e0e992b347bec Commit description: If drm_dev_register() fails after registering a minor (e.g. render minor registered, primary minor fails), userspace could have opened the first minor and entered a drm_dev_enter() critical section. Since the unplugged flag was never set, the ioctl proceeds while the error path tears down device resources. Fix this by introducing drm_dev_synchronize_unplug(), which sets the unplugged flag and waits for the SRCU barrier, ensuring all in-flight drm_dev_enter() critical sections complete before cleanup proceeds; call it on the error path of drm_dev_register(). Fixes: bee330f ("drm: Use srcu to protect drm_device.unplugged") Cc: stable@vger.kernel.org Reported-by: sashiko-bot@kernel.org Closes: https://lore.kernel.org/all/20260620190648.2E9F61F000E9@smtp.kernel.org/ Reviewed-by: Alexandre Courbot Reviewed-by: Lyude Paul Tested-by: Deborah Brouwer Link: https://patch.msgid.link/20260628145406.2107056-17-dakr@kernel.org Signed-off-by: Danilo Krummrich Signed-off-by: Greg Kroah-Hartman Changed files: drivers/gpu/drm/drm_drv.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=26d9162213952927cd59c0d94b7e0e992b347bec ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-89823 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89823 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:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H The Best / paranoid CVSS vector: AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H The Best / paranoid CVSS score: 7 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: 4 Paranoid ActionableScore: 5 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: YES Published priority for this report (same as in Subject): MODERATE 7.0 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).