From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-93108][MODERATE REGULAR] RDMA/ipoib: Drain RCU callbacks during module teardown [ Upstream Date: Thu, 17 Sep 2026 23:27:22 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-93108 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: 1 X-AL-KERNEL-ActionableScore-Lower: unknown X-AL-KERNEL-Commit: bb9c23733a0c50e136bd40c27cb60023cc58e22c List-Id: CVE: CVE-2026-93108 Priority: MODERATE REGULAR AL-KERNEL base severity: MODERATE KPANIC flag: NO Patch: RDMA/ipoib: Drain RCU callbacks during module teardown [ Upstream Commit: bb9c23733a0c50e136bd40c27cb60023cc58e22c Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=bb9c23733a0c50e136bd40c27cb60023cc58e22c Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-93108 Analysis date: Thu, 17 Sep 2026 23:27:22 -0400 ActionableScore: 1 ActionableScore lower bound: unknown Actionable bucket: Ordinary Moderate Manual review required: NO Summary: No one-sentence report phrase was found. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2026-93108 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-93108 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: NO A detailed explanation of the methodology and priority rules is included at the end of this message. ====================================================================== AL-KERNEL CLASSIFICATION RESULT ====================================================================== CVE-2026-93108 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-93108 CHECK WITH IMPACT FROM ORIG NN MODERATE Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:H/PR:H/UI:N/S:U/C:H/I:H/A:H';*CWE-416;CWE-362;*CWE-664;Other CVSS 'AV:L/AC:H/PR:H/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '6.4';DESCR 'RDMA/IPoIB module teardown can race with pending RCU callbacks because IPoIB reclamation completions may still execute from an RCU callback while the ib_ipoib module is being removed. If teardown continues before the callback returns, module code can be unloaded while it is still executing, which creates a use-after-free style condition and can crash the kernel. For the CVSS the PR:H is used because reliable triggering normally requires loading or unloading the ib_ipoib kernel module or controlling RDMA client teardown with administrator privileges. The issue is not directly network reachable because RDMA or IPoIB traffic alone does not unload the module. Impact is at least denial of service via kernel crash and in the paranoid case may include confidentiality or integrity impact if the stale module code execution window can be turned into controllable memory corruption.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Ordinary Moderate (with actual score 1) YES SIMPLEFIX NETWORK HARDWARE DEBUGFS LINUS DECREASED_TO_MODERATEREG_BASED_ON_ACTIONABLESCORELESSTHAN4 NO NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=1 PrimitiveClass=CLASS_B RealUAF=YES HostAdminRequired=YES NamespaceScopedPrivilege=NO ImportantCandidate=NO ManualReview=RECOMMENDED AutoDowngradeAllowed=YES 1. ActionableScore * Raw conservative score: 1 * Conservative score: 1 * Raw paranoid score: 1 * Paranoid score: 1 * Final recommended bucket: **Ordinary Moderate**:: No floor was applied. Real UAF FLOOR 4 does not apply because host-admin-equivalent privilege is positively established for module load or unload. No paranoid escalation was applied because the patch supports a module teardown lifetime bug and crash risk, but not controlled reclaim, arbitrary read/write, callback hijack from reclaimed data, or a below-host-admin trigger. 2. Signal breakdown Positive signals: * +2 Real UAF / lifetime violation: RCU callback code may still be executing while the ib_ipoib module is unloaded, creating a post-lifetime module text or callback execution condition. * +1 Reliable kernel crash / strong DoS: unloading module code while an RCU callback is still executing can realistically crash or fault the kernel. Negative signals: * -2 Full host-admin-equivalent privilege: reliable triggering requires loading, unloading, or failing initialization of a kernel module, which is host root / CAP_SYS_MODULE equivalent. Non-awarded material signals: * Remote reachability not awarded: RDMA/IPoIB traffic alone does not unload the kernel module. * Local unprivileged trigger not awarded: no evidence shows that an ordinary local user can perform the complete module teardown or registration-failure path. * Namespace-scoped privilege not awarded: module unload is not a network namespace scoped operation. * Controlled reclaim not demonstrated. * Arbitrary write not demonstrated. * Confidentiality and integrity not awarded: the established consequence is stale execution during teardown and crash risk, not a demonstrated disclosure or modification primitive. * Structured stale-object leverage not separately awarded: the stale use is teardown/RCU callback lifetime related and no attacker-shaped replacement object or meaningful field consumption is established. * Concrete repeated release/refcount signal not awarded: the patch does not show first put/free followed by a second lifetime/refcount mutation on the exact same dangling reference. * Extra Availability not awarded: the same crash/fault is already counted once. * Hard race penalty not awarded: although this is a teardown/RCU ordering bug, the supplied evidence does not establish a materially hard attacker-controlled timing requirement beyond the administrative teardown sequence. Repeated-release/refcount checklist: * First release can end lifetime: not established for an object reference in the repeated-release sense. * Exact same reference survives: not established. * Second operation is a real lifetime/refcount mutation: not established. * +1 Concrete repeated release/refcount awarded: no. * Paranoid repeated-refcount leverage +1 awarded: no. * FLOOR 5 applied: no. 3. Reachability analysis The complete trigger appears to require host administrator control over the ib_ipoib kernel module lifecycle. The vulnerable paths are module initialization failure cleanup and module exit cleanup. Both are reached through module loading/unloading or RDMA client teardown controlled by host-level kernel management. Ordinary local users are not shown to be able to unload ib_ipoib or force the relevant client registration failure. Namespace-scoped privilege is not established because module teardown is global host state, not per-network-namespace state. Prior host setup exists in the sense that the module and RDMA/IPoIB stack must be present, but the decisive operation is still host-admin module management. Special RDMA-capable hardware or configuration may be relevant for real deployment, but no separate rare-hardware penalty is needed because the host-admin requirement already dominates the practical security exposure. The race is between RCU callback completion and module teardown. The fix uses rcu_barrier() after client/workqueue teardown, which confirms that pending callbacks must be drained before module text can safely disappear. 4. Primitive and severity interpretation The exact primitive is a real RCU/module lifetime violation: callback code associated with IPoIB reclamation may still execute after teardown has progressed far enough that ib_ipoib can be unloaded. This is a Class B lifetime/UAF primitive, not Class A, because there is no evidence of controlled reclaim, object replacement, directed corruption, arbitrary kernel read/write, or callback/function-pointer control by a below-host-admin attacker. This is closer to an asynchronous teardown UAF than a general attacker-controlled active-object UAF. The demonstrated practical consequence is kernel crash or host DoS. A speculative consequence would be stronger memory corruption if module text reuse or callback execution after unload could be shaped, but the supplied patch does not establish such leverage. ImportantCandidate=NO because the only established attacker model is host-admin-equivalent module lifecycle control, and the demonstrated consequence is crash/DoS. AutoDowngradeAllowed=YES because this is a constrained teardown/RCU lifetime issue with no unresolved below-host-admin exploitation primitive, no repeated refcount mutation, and no demonstrated security-boundary bypass. Likely operational severity is ordinary Moderate or Low-like for security triage, despite being a real correctness and stability bug. 5. One-sentence report phrase RDMA/IPoIB module teardown can race with pending RCU callbacks, allowing ib_ipoib code to be unloaded while callback code is still executing and causing a host kernel crash, but reliable triggering appears to require host-admin module lifecycle control. 6. Manual review recommendation MANUAL CHECK RECOMMENDED Reason: this is a real RCU lifetime/UAF-style module teardown bug, but current evidence supports host-admin-only reachability and crash-only impact, so it does not require Important-class escalation without additional exploit or below-host-admin reachability evidence. 7. Call-site confidence Call-site confidence: high ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: RDMA/ipoib: Drain RCU callbacks during module teardown [ Upstream Commit: bb9c23733a0c50e136bd40c27cb60023cc58e22c Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=bb9c23733a0c50e136bd40c27cb60023cc58e22c Changed files: drivers/infiniband/ulp/ipoib/ipoib_main.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=bb9c23733a0c50e136bd40c27cb60023cc58e22c ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-93108 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-93108 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:H/UI:N/S:U/C:N/I:N/A:H The Best / paranoid CVSS vector: AV:L/AC:H/PR:H/UI:N/S:U/C:H/I:H/A:H The Best / paranoid CVSS score: 6.4 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: 1 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).