From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-74736][MODERATE REGULAR] net/sched: cls_bpf: reject dev-bound programs bound to a different device [ Upstream Date: Wed, 26 Aug 2026 12:45:47 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-74736 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: 1 X-AL-KERNEL-Commit: ec5a552f4b2d841c6c021752450716e1a9676661 List-Id: CVE: CVE-2026-74736 Priority: MODERATE REGULAR AL-KERNEL base severity: MODERATE KPANIC flag: NO Patch: net/sched: cls_bpf: reject dev-bound programs bound to a different device [ Upstream Commit: ec5a552f4b2d841c6c021752450716e1a9676661 Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=ec5a552f4b2d841c6c021752450716e1a9676661 Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-74736 Analysis date: Wed, 26 Aug 2026 12:45:47 -0400 ActionableScore: 3 ActionableScore lower bound: 1 Actionable bucket: Borderline, manual review recommended Manual review required: YES Summary: A cls_bpf validation bug allowed a device-bound offloaded SCHED_CLS BPF program for one netdev to be attached to another netdev, leaving offload state inconsistent after the original device is removed and potentially causing a WARN or kernel panic under panic_on_warn. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2026-74736 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-74736 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-2026-74736 MODERATE CHECK WITH IMPACT FROM ORIG NN MODERATE Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H';CWE-863;CWE-284;*CWE-664;Other CVSS 'AV:L/AC:L/PR:H/UI:N/S:U/C:N/I:N/A:L';BEST CVSS score: '5.5';DESCR 'A TC cls_bpf logic bug allowed a device bound offloaded SCHED_CLS BPF program loaded for device A to be attached with skip_sw to device B without checking that the bound device matched the target TC device. When device A is deleted, the program offload state can be destroyed while the program is still attached to device B, which can trigger a netdevsim WARN and can become a kernel panic when panic_on_warn is enabled. For the CVSS the PR:L is used for the paranoid score because TC BPF offload control may be available to a reduced capability root, container root, or delegated network administration service rather than only to a full host administrator. The issue is not directly network reachable and is triggered through local control plane operations for BPF and traffic control configuration. Impact is denial of service, with no concrete evidence in the patch of information disclosure or privilege escalation.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline, manual review recommended (with actual score 2) YES REMOTE BPF SIMPLEFIX NETWORK HARDWARE QDISC YES NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=3 ActionableScoreLower=1 ## 1. ActionableScore * Conservative score: 1 * Paranoid score: 3 * Final recommended bucket: **Borderline, manual review recommended**:: ## 2. Signal breakdown Conservative signals: * Real lifetime corruption: +1. A device-bound BPF program can remain attached to device B after device A deletion destroys the program offload state. * Reliable kernel crash / strong DoS: +1. The demonstrated failure is a netdevsim WARN that can become a panic when panic_on_warn=1. * Requires admin/root/CAP_* in typical deployments: -2. Loading and attaching TC BPF offload programs normally requires CAP_NET_ADMIN and BPF-related privileges. * Rare or difficult-to-reach configuration: -1. The path depends on device-bound offloaded BPF, skip_sw, and an offload-capable or simulated netdev setup. * Floor applied at 1 because this is a real kernel lifetime/state mismatch in BPF/netdev offload code, not a pure cosmetic validation fix. Paranoid signals: * Real lifetime corruption: +1. * Reliable kernel crash / strong DoS: +1. * Privileged device-management lifetime/state corruption path: +1. The mismatch affects netdev offload state lifetime during device teardown. * Privilege penalty reduced in paranoid interpretation: 0 instead of -2. TC/BPF network control may be delegated to container root, reduced-capability root, or a network-management service account. * Rare/config-dependent exposure: 0 in paranoid interpretation, because BPF and TC offload bugs deserve manual review even when not broadly exposed. Not awarded: * No remote/network-triggerable score. This is a local control-plane operation. * No local unprivileged trigger. Unprivileged BPF should not be assumed. * No generic strong memory corruption. The patch shows stale offload state and WARN, but not attacker-controlled reclaim, arbitrary write, type confusion, or object replacement. * No privilege escalation score. LPE is not supported by the patch context. ## 3. Reachability analysis The bug is triggered by loading a device-bound SCHED_CLS BPF program for one netdev and attaching it through cls_bpf with skip_sw to another TC netdev. In typical deployments this requires CAP_NET_ADMIN and BPF/offload privileges, so conservative scoring treats it as privileged local. Namespaces or delegated network administration can lower the practical privilege level if a container root or service account can manage TC and BPF offload objects. The path is not remotely reachable by packet traffic and depends on device-bound offload support, making broad default exposure limited. ## 4. Severity interpretation This behaves more like a borderline actionable Moderate than an Important-class issue. The realistic demonstrated impact is DoS through WARN or panic_on_warn after offload state is destroyed while still referenced by another device attachment. Theoretical lifetime corruption exists, but the patch does not show a concrete memory corruption primitive such as UAF reclaim, controlled replacement, arbitrary write, type confusion, or callback control. Manual review is still useful because BPF plus netdev offload lifetime mismatches are historically sensitive and may have product-specific reachability through delegated network control. ## 5. One-sentence report phrase A cls_bpf validation bug allowed a device-bound offloaded SCHED_CLS BPF program for one netdev to be attached to another netdev, leaving offload state inconsistent after the original device is removed and potentially causing a WARN or kernel panic under panic_on_warn. ## 6. Manual review recommendation MANUAL CHECK RECOMMENDED Manual review is recommended because the issue combines BPF, TC, device-bound offload state, and netdev teardown lifetime behavior. It should not be treated as a strong Important candidate without evidence of real memory corruption or privilege escalation, but it is also not ideal for automatic closure in environments that delegate TC/BPF control to reduced-privilege services or containers. ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: net/sched: cls_bpf: reject dev-bound programs bound to a different device [ Upstream Commit: ec5a552f4b2d841c6c021752450716e1a9676661 Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=ec5a552f4b2d841c6c021752450716e1a9676661 Changed files: net/core/dev.c net/sched/cls_bpf.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=ec5a552f4b2d841c6c021752450716e1a9676661 ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-74736 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-74736 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:N/A:L The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H The Best / paranoid CVSS score: 5.5 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: 1 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).