From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-92485][MODERATE 7.0] bpf: Fix WARNING in bpf_tracing_link_release [ Upstream Date: Fri, 18 Sep 2026 04:30:06 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-92485 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: 48a0209d8da0d90a7b0a0db19d1ff88027b13781 List-Id: CVE: CVE-2026-92485 Priority: MODERATE 7.0 AL-KERNEL base severity: MODERATE KPANIC flag: YES Patch: bpf: Fix WARNING in bpf_tracing_link_release [ Upstream Commit: 48a0209d8da0d90a7b0a0db19d1ff88027b13781 Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=48a0209d8da0d90a7b0a0db19d1ff88027b13781 Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-92485 Analysis date: Fri, 18 Sep 2026 04:30:06 -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-2026-92485 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-92485 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-2026-92485 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-92485 CHECK WITH IMPACT FROM ORIG NN LOW Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:H';CWE-670;CWE-667;CWE-362;Other CVSS 'AV:L/AC:L/PR:H/UI:N/S:U/C:N/I:L/A:L';BEST CVSS score: '6.6';DESCR 'BPF trampoline flags can be corrupted because the verifier overwrites the existing trampoline flags when attaching fexit programs to a tail_call_reachable target. This can remove BPF_TRAMP_F_CALL_ORIG from a shared trampoline and later make link release fail to restore the patched instruction state correctly, producing a kernel WARNING and potentially leaving the trampoline in an inconsistent state. For the CVSS the PR:L is used for the paranoid score because some deployments delegate BPF capabilities such as CAP_BPF, CAP_PERFMON, or service level BPF loading without full host administration. The issue is not network reachable and requires local ability to load suitable BPF tracing programs. Impact is at least local denial of service or BPF control flow integrity degradation. In the worst still defensible interpretation it may have limited confidentiality or integrity impact because executable trampoline state is corrupted and 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 WARNONLY BPF HARDWARE TEST KPANIC INCREASED_FROM_LOW_BASED_ON_REQUIREMANUALCHECK YES NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=5 ActionableScoreLower=4 PrimitiveClass=CLASS_B RealUAF=NO HostAdminRequired=UNKNOWN NamespaceScopedPrivilege=UNKNOWN ImportantCandidate=YES ManualReview=REQUIRED AutoDowngradeAllowed=NO 1. ActionableScore * Raw conservative score: 4 * Conservative score: 4 * Raw paranoid score: 5 * Paranoid score: 5 * Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**:: No floor or cap changed the conservative score. No floor or cap changed the paranoid score. The BPF privileged control-plane cap does not lower the result because the score remains within the allowed 5 to 6 paranoid range for unresolved privileged BPF control-plane corruption. 2. Signal breakdown Conservative positive signals: * +1: Constrained active-object corruption. The verifier overwrites `tr->flags` with `BPF_TRAMP_F_TAIL_CALL_CTX` instead of preserving existing flags, which can remove `BPF_TRAMP_F_CALL_ORIG` from an already shared BPF trampoline. * +1: Active executable trampoline state impact. The corrupted flag state affects whether trampoline patching/restoration uses the correct call/jmp instruction semantics. * +1: Demonstrated inconsistent release behavior. Closing the first fexit link can fail to restore the target program instruction correctly and triggers a WARNING in `bpf_tracing_link_release`. * +1: Concrete local trigger sequence is described. The commit gives a specific ordered sequence using two fexit attachments to the same `tail_call_reachable` target and then closing one link. Paranoid additional signal: * +1: Unresolved BPF trampoline code-patching consequence. The corrupted state affects executable trampoline / target-program patching, so a manual review is needed to determine whether impact is limited to WARN/DoS or can influence BPF control-flow integrity more broadly. Negative or non-awarded signals: * +0: No Real UAF is established. * +0: No concrete repeated release/refcount operation is established. * +0: No controlled reclaim or attacker-shaped replacement object is demonstrated. * +0: No arbitrary or directed kernel write is demonstrated. * +0: No callback/function-pointer control is demonstrated. * +0: No information disclosure path is demonstrated. * +0: No direct privilege escalation chain is demonstrated. * +0: No remote reachability is established. * +0: No extra Availability point is awarded because the visible symptom is a WARN and inconsistent trampoline state, not a separately demonstrated persistent host-wide failure. * +0: Weak LPE is not separately awarded because the evidence supports unresolved trampoline state corruption, not a concrete LPE primitive. * +0: No race/timing penalty is applied. The trigger is an ordered state sequence, not a narrow externally uncontrollable race. Repeated release/refcount checklist: * First release can end lifetime: not established. * Exact same dangling reference survives: not established. * Second operation mutates lifetime/refcount: not established. * Concrete repeated release/refcount +1 awarded: no. * Paranoid repeated-refcount leverage +1 awarded: no. * FLOOR 5 for repeated release/refcount applied: no. 3. Reachability analysis The relevant attacker action is local BPF control-plane activity: loading/attaching fexit tracing programs to a `tail_call_reachable` BPF program and then closing one BPF link. Ordinary unprivileged reachability is not established. BPF tracing program loading and attachment normally require privileged BPF capabilities such as `CAP_BPF`, `CAP_PERFMON`, or older `CAP_SYS_ADMIN`-style permission depending on kernel and configuration. Because delegation of BPF capabilities is possible in some environments, but complete below-host-admin reachability is not proven from the supplied patch, `HostAdminRequired=UNKNOWN` is used rather than YES or NO. Namespace-scoped privilege is also unresolved. The patch does not establish that all required stages, including BPF program load, fexit attach, target program setup, and link release, are fully available inside attacker-created namespaces. Therefore `NamespaceScopedPrivilege=UNKNOWN`. No network trigger is indicated. No special hardware is required beyond BPF trampoline/JIT support and a kernel configuration that supports the affected BPF tracing path. No hard race is required by the described sequence. 4. Primitive and severity interpretation The exact primitive is corruption of BPF trampoline state through incorrect overwrite of `tr->flags`. The affected object is an active structured kernel object involved in executable trampoline and target-program instruction patching. This is not a heap UAF, not a double free, and not an ordinary cleanup-only stale dereference. The demonstrated consequence is a WARN during `bpf_tracing_link_release` and failure to restore the expected patched instruction state because `BPF_TRAMP_F_CALL_ORIG` was lost. The speculative consequence is broader BPF control-flow integrity degradation if the inconsistent trampoline state persists or affects later trampoline updates. The patch does not establish arbitrary kernel read/write, controlled reclaim, callback hijack, or LPE. PrimitiveClass is CLASS_B because a real active kernel state corruption primitive exists and affects executable trampoline patching, but strong exploitation leverage is not directly established. It is not CLASS_A because there is no controlled reclaim, arbitrary write, callback control, or demonstrated privilege escalation. It is not CLASS_C because the issue is more than a passive WARN-only value mismatch: the corrupted flag controls trampoline instruction patching/restoration behavior. ImportantCandidate is YES because BPF trampoline code-patching state corruption has meaningful unresolved exploitation implications that should not be automatically downgraded from a small patch alone. AutoDowngradeAllowed is NO because automatic closure would lose an unresolved active kernel code-patching corruption primitive. Likely operational severity is Actionable Moderate, with manual review needed before ruling out Important-class impact. 5. One-sentence report phrase Incorrect BPF trampoline flag handling can corrupt shared trampoline state during fexit attachment to tail call reachable programs, causing failed instruction restoration on link release and requiring manual review for possible BPF control flow integrity impact beyond the demonstrated WARNING. 6. Manual review recommendation MANUAL CHECK REQUIRED Reason: the patch affects BPF trampoline instruction patching/restoration state, and although only a WARNING is demonstrated, the corrupted active executable state is security-relevant enough that automatic downgrade would be unsafe. 7. Call-site confidence Call-site confidence: high ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: bpf: Fix WARNING in bpf_tracing_link_release [ Upstream Commit: 48a0209d8da0d90a7b0a0db19d1ff88027b13781 Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=48a0209d8da0d90a7b0a0db19d1ff88027b13781 Changed files: kernel/bpf/syscall.c include/linux/bpf.h kernel/bpf/trampoline.c kernel/bpf/verifier.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=48a0209d8da0d90a7b0a0db19d1ff88027b13781 ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-92485 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-92485 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:L/PR:H/UI:N/S:U/C:N/I:L/A:L The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:H The Best / paranoid CVSS score: 6.6 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).