From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-90049][IMPORTANT] net: skbuff: don't skb_tx_error() the source skb in skb_zerocopy() [ Upstream Date: Wed, 16 Sep 2026 09:22:34 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-90049 X-AL-KERNEL-Priority: IMPORTANT X-AL-KERNEL-Severity: IMPORTANT X-AL-KERNEL-Base-Severity: IMPORTANT X-AL-KERNEL-KPANIC: YES X-AL-KERNEL-ActionableScore: 8 X-AL-KERNEL-ActionableScore-Lower: 7 X-AL-KERNEL-Commit: 849bdb83123760a865bcb2970127f4c0b9423ba3 List-Id: CVE: CVE-2026-90049 Priority: IMPORTANT AL-KERNEL base severity: IMPORTANT KPANIC flag: YES Patch: net: skbuff: don't skb_tx_error() the source skb in skb_zerocopy() [ Upstream Commit: 849bdb83123760a865bcb2970127f4c0b9423ba3 Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=849bdb83123760a865bcb2970127f4c0b9423ba3 Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90049 Analysis date: Wed, 16 Sep 2026 09:22:34 -0400 ActionableScore: 8 ActionableScore lower bound: 7 Actionable bucket: Strong Important candidate Manual review required: YES Summary: No one-sentence report phrase was found. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2026-90049 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90049 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: IMPORTANT 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-2026-90049 IMPORTANT https://www.kernelcve.org/list/?q=CVE-2026-90049 CHECK WITH IMPACT FROM ORIG NN LOW Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H';*CWE-416;*CWE-664;*CWE-787;Other CVSS 'AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '8.1';DESCR 'skb_zerocopy() could call skb_tx_error() on the source skb after skb_orphan_frags() failed, even though the helper does not own that skb. This prematurely completes the zerocopy uarg and clears zerocopy ownership flags while the same skb may still be forwarded by Open vSwitch after a failed userspace upcall. A later local ESP delivery may then decrypt in place over shared frags that the skb no longer owns privately, creating a plausible memory corruption or use after free style condition. For the CVSS the PR:N is selected for the paranoid score because a remote packet may reach the affected OVS and ESP processing path when the vulnerable configuration is already present, without attacker privileges on the target. The issue is network reachable only in specific deployments using Open vSwitch userspace actions, zerocopy skb handling, and local ESP delivery. Impact is at least denial of service and may include confidentiality or integrity impact due to corrupted or reused packet buffers.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Strong Important candidate (with actual score 8) YES REMOTE ERRORPATH UAF SKB KPANIC INCREASED_TO_MODERATE7_FROM_LOW_BASED_ON_GUESSCVSS_AND_ACTIONABLESCOREHIGHER4 KPANIC INCREASED_TO_HIGH_BASED_ON_ACTIONABLESCOREHIGHERTHAN5 NO NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=8 ActionableScoreLower=7 PrimitiveClass=CLASS_A RealUAF=YES HostAdminRequired=NO NamespaceScopedPrivilege=UNKNOWN ImportantCandidate=YES ManualReview=REQUIRED AutoDowngradeAllowed=NO 1. ActionableScore * Raw conservative score: 7 * Conservative score: 7. No floor or cap applied. * Raw paranoid score: 8 * Paranoid score: 8. No floor or cap applied. * Final recommended bucket: **Strong Important candidate**:: 2. Signal breakdown Conservative signals: * +2 Real UAF / active lifetime violation. `skb_tx_error(from)` completes the zerocopy `uarg` for the source skb while that skb may still be in flight through remaining Open vSwitch actions. * +2 Strong ownership / shared-frag primitive. The same destructive operation clears `SKBFL_SHARED_FRAG`, which can make later ESP processing decrypt in place over fragments the skb does not own privately. * +2 Network-triggerable packet-processing path. A packet can reach the affected OVS userspace-action and later ESP delivery path when the vulnerable deployment is already configured. * +1 Reliable strong DoS. The established consequence includes corrupted skb/page ownership state and plausible kernel crash or fatal packet-processing failure. Paranoid additional signal: * +1 Integrity plausible. The in-place ESP write over non-private shared fragments is a concrete shared-page ownership violation and can plausibly corrupt data owned outside the skb. Important non-awarded signals: * Controlled reclaim is not demonstrated. * Arbitrary kernel read/write is not demonstrated. * Callback or function-pointer control is not demonstrated. * Confidentiality impact is not separately awarded because no concrete read-back or disclosure path is shown. * Extra availability is not awarded because the same crash/DoS consequence is already counted. * No race penalty is applied because the described issue is an ordered ownership/lifetime error path, not a narrow timing race. * No host-admin penalty is applied because the evaluated attacker can be a remote sender once the vulnerable OVS/ESP configuration exists. * No rare-configuration penalty is applied. The path is deployment-specific, but OVS, skb zerocopy handling, and ESP/XFRM are normal networking facilities rather than rare hardware or experimental-only features. Repeated release/refcount check: * First release can end lifetime: yes for the zerocopy completion semantics, because `skb_tx_error(from)` can complete the producer-visible zerocopy state. * Exact same reference survives: yes, the same skb may continue through remaining OVS actions. * Second operation is a real lifetime/refcount mutation: no concrete second put/free/refcount mutation on the exact same reference is shown. * Concrete repeated release/refcount +1 awarded: no. * Paranoid repeated-refcount leverage +1 awarded: no. * FLOOR 5 applied: no explicit floor was needed because raw scoring already exceeds it. 3. Reachability analysis The practical trigger is a packet-processing path involving Open vSwitch `OVS_ACTION_ATTR_USERSPACE`, an error from `skb_orphan_frags()`, continued forwarding because `do_execute_actions()` ignores the `output_userspace()` return value, and later local ESP delivery. The attacker does not need host administrative privileges to send traffic into an already configured vulnerable datapath, so `HostAdminRequired=NO` for the evaluated trigger. The issue is not universally Internet-reachable. It depends on a deployment where Open vSwitch userspace actions, zerocopy skb fragment handling, and ESP/XFRM processing can be reached by attacker-controlled packets. In many real deployments this is more likely inside a virtualized, container, VPN, overlay, or data-center network than on a default host. Namespace scope is unresolved from the supplied evidence. The patch does not prove that the complete setup and trigger chain can be instantiated entirely inside attacker-created namespaces, so `NamespaceScopedPrivilege=UNKNOWN`. No special physical hardware is required. No hard race is established. The failure depends on a specific packet-processing sequence and allocation/orphaning failure, but not on a narrow uncontrollable timing window. 4. Primitive and severity interpretation The exact primitive is a destructive lifetime and ownership operation on a source skb that `skb_zerocopy()` does not own. Calling `skb_tx_error(from)` completes the source skb zerocopy state and clears zerocopy/shared-frag ownership flags while the skb can still be processed by the networking stack. This is an active structured skb/page-fragment ownership violation, not fixed stale storage and not a simple NULL dereference. The most important consequence is that ESP can later decrypt in place instead of forcing copy-on-write through `skb_cow_data()`, causing writes into shared fragments that are not privately owned by the skb. This supports `PrimitiveClass=CLASS_A` because the patch establishes a concrete shared-frag ownership/COW violation and active packet-buffer lifetime corruption. Demonstrated impact is at least DoS. Speculative but technically defensible impact includes integrity corruption of reused or externally owned packet buffers. Full privilege escalation or arbitrary kernel R/W is not established. `ImportantCandidate=YES` because the primitive is stronger than ordinary crash-only behavior and crosses skb fragment ownership/lifetime boundaries in a network packet-processing path. `AutoDowngradeAllowed=NO` because automatic closure would lose a concrete shared-page ownership violation and unresolved memory-lifetime corruption primitive. Likely operational severity is Strong Important candidate, especially for systems using Open vSwitch plus ESP/XFRM paths reachable by untrusted traffic. 5. One-sentence report phrase A destructive skb_zerocopy error path can prematurely complete the source skb zerocopy state and clear shared fragment ownership flags, allowing later ESP processing to write in place to non-private skb fragments and creating a network-reachable DoS with plausible memory ownership corruption. 6. Manual review recommendation MANUAL CHECK REQUIRED The patch establishes a concrete skb lifetime and shared-frag ownership violation in a network packet path, with plausible corruption beyond ordinary crash-only behavior. 7. Call-site confidence Call-site confidence: high ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: net: skbuff: don't skb_tx_error() the source skb in skb_zerocopy() [ Upstream Commit: 849bdb83123760a865bcb2970127f4c0b9423ba3 Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=849bdb83123760a865bcb2970127f4c0b9423ba3 Changed files: net/core/skbuff.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=849bdb83123760a865bcb2970127f4c0b9423ba3 ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-90049 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90049 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:N/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:H The Best / paranoid CVSS vector: AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H The Best / paranoid CVSS score: 8.1 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: 7 Paranoid ActionableScore: 8 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: IMPORTANT KPANIC detected for this report: YES Published priority for this report (same as in Subject): IMPORTANT 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).