From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-90106][MODERATE REGULAR] net: bridge: arp/nd proxy: fix reading neigh ha [ Upstream Date: Fri, 18 Sep 2026 01:33:57 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-90106 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: 4 X-AL-KERNEL-ActionableScore-Lower: 3 X-AL-KERNEL-Commit: 291c9e137b7c74e8a512cded4845352da4c4adda List-Id: CVE: CVE-2026-90106 Priority: MODERATE REGULAR AL-KERNEL base severity: MODERATE KPANIC flag: NO Patch: net: bridge: arp/nd proxy: fix reading neigh ha [ Upstream Commit: 291c9e137b7c74e8a512cded4845352da4c4adda Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=291c9e137b7c74e8a512cded4845352da4c4adda Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90106 Analysis date: Fri, 18 Sep 2026 01:33:57 -0400 ActionableScore: 4 ActionableScore lower bound: 3 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-2026-90106 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90106 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-2026-90106 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-90106 CHECK WITH IMPACT FROM ORIG NN MODERATE Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:A/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:L';CWE-362;CWE-662;CWE-667;Other CVSS 'AV:A/AC:H/PR:N/UI:N/S:U/C:N/I:L/A:L';BEST CVSS score: '5';DESCR 'The bridge ARP and ND proxy suppression path reads the neighbour hardware address directly from n ha while the neighbour entry can be updated concurrently. This can produce a torn or mixed MAC address and can make the bridge use an inconsistent address for FDB lookup and for ARP or ND proxy replies. For the CVSS the PR:N is used in the paranoid score because a host on the same L2 or adjacent bridge domain may be able to race neighbour updates and proxy requests without local code execution on the victim. The issue is adjacent network reachable rather than Internet remote, and it depends on bridge neighbour suppression being enabled. Impact is mainly incorrect proxy replies, traffic misdelivery, or local network denial of service. Limited confidentiality or integrity impact is possible only through misdirected traffic within the affected L2 domain, with no evidence of kernel memory corruption or privilege escalation.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline / manual review recommended (with actual score 4) YES OOB DANGER SIMPLEFIX SKB IPV6 DECREASED_TO_MODERATEREG_BASED_ON_FALSEPOSCHECKOFKP NO NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=4 ActionableScoreLower=3 PrimitiveClass=CLASS_C RealUAF=NO HostAdminRequired=NO NamespaceScopedPrivilege=NO ImportantCandidate=NO ManualReview=RECOMMENDED AutoDowngradeAllowed=YES 1. ActionableScore * Raw conservative score: 3 * Conservative score: 3 * Raw paranoid score: 4 * Paranoid score: 4 * Final recommended bucket: **Borderline / manual review recommended**:: No floor or cap changed the score. 2. Signal breakdown Conservative positive signals: * +2 Adjacent network reachable packet path. ARP or ND packets can reach the vulnerable bridge proxy suppression path when BR_NEIGH_SUPPRESS is enabled. * +1 Availability or traffic correctness impact. A torn neighbour hardware address can cause wrong FDB lookup, wrong ARP or ND proxy replies, traffic misdelivery, or local L2 connectivity disruption. Paranoid additional signal: * +1 Limited confidentiality or integrity concern. If the torn MAC happens to correspond to another reachable FDB entry, traffic may be misdirected within the affected L2 domain. This is limited and topology dependent. Negative signals: * No privilege penalty. Host-admin privilege is required to configure the bridge feature, but not positively required for an adjacent attacker to send ARP or ND traffic once the feature is deployed. * No race penalty. The issue is a race, but the supplied evidence does not prove a narrow one-shot or externally uncontrollable timing window. * No rare-subsystem penalty. Bridge ARP or ND suppression is configuration-dependent, but not enough evidence is supplied to treat it as rare and unreachable. Important non-awarded signals: * No RealUAF. No freed or stale post-lifetime object use is established. * No memory corruption. The bug is a torn or partial read of a MAC address, not an OOB, UAF, double free, or overwrite. * No arbitrary write or controlled reclaim is demonstrated. * No kernel crash is demonstrated. * No privilege escalation is demonstrated. * No authorization boundary bypass is established. * Extra availability was not awarded because only traffic disruption or misdelivery is supported, not host-wide crash or persistent failure. 3. Reachability analysis The evaluated attacker is an adjacent L2 or bridge-domain host that can send ARP or IPv6 ND traffic reaching a bridge port with neighbour suppression enabled. The vulnerable path is data-plane packet handling, not a local ioctl or netlink-only control path. Host administrator action is normally needed beforehand to configure bridge neighbour suppression and the relevant topology, but the attacker does not need host-admin rights on the victim after that deployment exists. Therefore HostAdminRequired is NO for the deployed vulnerable path. Namespace-scoped privilege is not the relevant model here. The issue is triggered by network packets in an L2 domain, not by acquiring CAP_NET_ADMIN inside an attacker-created namespace. The race depends on concurrent neighbour hardware address update and proxy handling. Repeated packets and neighbour churn may make attempts practical, but the patch does not establish a reliable exploit beyond inconsistent MAC use. 4. Primitive and severity interpretation The exact primitive is an unstable concurrent read of neighbour->ha. The bridge can observe a mixed MAC address and then use it for FDB lookup and ARP or ND reply construction. This is not active structured UAF, fixed stale storage UAF, double release, refcount mutation, or memory corruption. It is best classified as CLASS_C invalid-state behavior with possible network correctness and limited L2 traffic impact. Demonstrated consequence is incorrect proxy behavior. Speculative consequence is limited traffic misdelivery or disclosure within the same L2 domain if the torn address maps to another host. There is no patch-supported path to kernel memory corruption, LPE, or host-wide DoS. ImportantCandidate is NO because no strong corruption, ownership, authorization, or lifetime primitive is established. AutoDowngradeAllowed is YES because the unresolved risk does not include a concrete kernel corruption or security-boundary primitive. Likely operational severity is low to moderate, but adjacent network reachability makes manual review recommended rather than fully automatic closure. 5. One-sentence report phrase A bridge ARP and ND proxy race can read a torn neighbour MAC address during concurrent neighbour updates, causing incorrect FDB lookup or proxy replies and possible L2 traffic disruption or limited misdelivery when neighbour suppression is enabled. 6. Manual review recommendation MANUAL CHECK RECOMMENDED Reason: the issue is network-adjacent and can affect ARP or ND proxy behavior, but the patch supports only invalid MAC use and limited L2 impact, not kernel memory corruption or privilege escalation. 7. Call-site confidence Call-site confidence: high ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: net: bridge: arp/nd proxy: fix reading neigh ha [ Upstream Commit: 291c9e137b7c74e8a512cded4845352da4c4adda Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=291c9e137b7c74e8a512cded4845352da4c4adda Changed files: net/bridge/br_arp_nd_proxy.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=291c9e137b7c74e8a512cded4845352da4c4adda ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-90106 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90106 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:N/I:L/A:L The Best / paranoid CVSS vector: AV:A/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:L The Best / paranoid CVSS score: 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: 3 Paranoid ActionableScore: 4 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).