From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2025-38590][MODERATE 7.0] net/mlx5e: Remove skb secpath if xfrm state is not found [ Upstream Date: Wed, 16 Sep 2026 12:49:21 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2025-38590 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: 4 X-AL-KERNEL-ActionableScore-Lower: 3 X-AL-KERNEL-Commit: 314f568b84b01f6eac1e4313ca47f9ade4349443 List-Id: CVE: CVE-2025-38590 Priority: MODERATE 7.0 AL-KERNEL base severity: MODERATE KPANIC flag: YES Patch: net/mlx5e: Remove skb secpath if xfrm state is not found [ Upstream Commit: 314f568b84b01f6eac1e4313ca47f9ade4349443 Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=314f568b84b01f6eac1e4313ca47f9ade4349443 Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2025-38590 Analysis date: Wed, 16 Sep 2026 12:49:21 -0400 ActionableScore: 4 ActionableScore lower bound: 3 Actionable bucket: Borderline / manual review recommended Manual review required: YES Summary: A remote peer reaching an mlx5e IPsec RX offload endpoint may trigger a zero-length secpath invalid dereference in XFRM policy checking, causing a kernel crash without demonstrated information disclosure or privilege escalation. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2025-38590 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2025-38590 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-2025-38590 MODERATE https://www.kernelcve.org/list/?q=CVE-2025-38590 CHECK WITH IMPACT FROM ORIG NN IMPORTANT Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:H';CWE-125;**CWE-476;CWE-362;Other CVSS 'AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '7';DESCR 'A crash in the mlx5e IPsec RX offload path can occur when hardware reports a decrypted packet whose xfrm state is no longer found in the driver xarray. The skb can keep a zero-length secpath and later XFRM policy code can evaluate xfrm_input_state(skb), which effectively dereferences sp xvec at len minus 1 and reaches an invalid state pointer. A remote IPsec peer or a host able to send traffic to the offloaded IPsec endpoint may be able to trigger the receive path, but reliable triggering depends on mlx5e IPsec offload being enabled and on timing around xfrm state removal or rekey. For the CVSS the PR:N is used because no local account on the target is required when the affected IPsec receive path is reachable over the network. Impact is at least denial of service via kernel crash. For the paranoid score, limited confidentiality or integrity impact is kept for manual review because the bug includes an invalid read in policy handling, but the patch context does not support a strong write primitive or confirmed privilege escalation.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline / manual review recommended (with actual score 4) MAYBE REMOTE DANGER UAF NETWORK SKB KERNEL_PANIC_PLUS_UAF HARDWARE KPANIC PACKET - - 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 signals: * +2 Remote reachable packet-processing path. The vulnerable code is in mlx5e IPsec RX offload and can be reached by received IPsec traffic when the endpoint is configured and reachable. * +1 Reliable kernel crash / strong DoS. The supplied trace shows a kernel Oops in `__xfrm_policy_check()` from RX softirq context. * +1 Weak / indirect corruption candidate. A zero-length `secpath` makes downstream XFRM code evaluate `sp->xvec[sp->len - 1]`, causing an invalid structured-object read / invalid pointer dereference. * -1 Timing / lifecycle condition. The miss depends on the xfrm state no longer being found after hardware decryption, likely around state removal, rekey, or teardown timing. Paranoid difference: * The paranoid score withholds the -1 timing penalty because the exact difficulty and repeatability are not fully established, and repeated network traffic during normal IPsec state churn may make the condition more reachable. Material non-awarded signals: * RealUAF not awarded. The xfrm state may have been freed before lookup, but the demonstrated failing operation is not a dereference of that freed state. It is an invalid `secpath` / zero-length `xvec` access. * Controlled reclaim not demonstrated. * Arbitrary write not demonstrated. * Confidentiality not awarded. The invalid read is internal and no disclosure path to userspace or network output is shown. * Integrity not awarded. No unauthorized modification of kernel state is established beyond crash-inducing invalid state. * Strong LPE / strong exploitation leverage not awarded. No controlled replacement, callback control, refcount takeover, or directed corruption is shown. * Extra Availability not awarded because the same kernel Oops is already counted as reliable crash / DoS. * Concrete repeated release/refcount signal not awarded. There is no visible second `put`, `free`, or refcount mutation through the same dangling reference. ## 3. Reachability analysis The evaluated attacker is a remote network peer or host able to send traffic to the affected IPsec endpoint. No local account on the target is required for the packet receive path, so `HostAdminRequired=NO`. The vulnerable path requires mlx5e / ConnectX IPsec RX offload to be configured and active. That is a prior deployment condition, normally set by an administrator, but it is not a privilege the attacker needs once the endpoint is exposed. This is not namespace-scoped privilege. The issue is in a physical NIC offload receive path, not an attacker-created namespace-only object. The complete trigger likely requires traffic to arrive after hardware decryption when the corresponding xfrm state is no longer found in the driver's xarray. That suggests a lifecycle/timing dependency around state removal or rekey. The patch and trace establish the crash path, but not full attacker control over the state removal event. Special hardware is involved, but the driver and feature are not treated as physical-only or rare enough for a hardware penalty. Exposure is deployment-specific and mainly relevant to systems using mlx5e IPsec offload. ## 4. Primitive and severity interpretation The exact primitive is an invalid structured read / invalid pointer dereference caused by leaving a zero-length `secpath` attached to an skb after the offloaded xfrm state lookup fails. This is active packet-processing state, but the demonstrated stale condition is not a confirmed use-after-free of `struct xfrm_state`. The downstream crash occurs because XFRM policy code assumes a valid secpath entry and indexes `xvec[len - 1]` when `len == 0`. Primitive class is `CLASS_C`: demonstrated behavior is primarily crash / DoS / invalid-state behavior without an independently useful corruption primitive. Demonstrated consequence is kernel crash / denial of service. A speculative consequence would be limited C/I concern from invalid structured state processing, but the supplied evidence does not show a disclosure path, write primitive, controlled reclaim, or privilege escalation. `ImportantCandidate=NO` because the patch supports remote DoS and invalid read, but not a strong Class B or Class A primitive. `AutoDowngradeAllowed=YES` because no meaningful unresolved exploitation leverage beyond crash is established, although manual review is still recommended due to network reachability and security-sensitive XFRM/IPsec context. Likely operational severity: actionable network-triggered DoS in affected mlx5e IPsec offload deployments, but not an Important-class memory-corruption candidate on the supplied evidence. ## 5. One-sentence report phrase A remote peer reaching an mlx5e IPsec RX offload endpoint may trigger a zero-length secpath invalid dereference in XFRM policy checking, causing a kernel crash without demonstrated information disclosure or privilege escalation. ## 6. Manual review recommendation MANUAL CHECK RECOMMENDED Reason: the issue is network-reachable in affected IPsec offload deployments and occurs in security-sensitive XFRM packet processing, but the established primitive is crash-only invalid dereference rather than confirmed UAF, controlled corruption, or LPE. ## 7. Call-site confidence Call-site confidence: high ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: net/mlx5e: Remove skb secpath if xfrm state is not found [ Upstream Commit: 314f568b84b01f6eac1e4313ca47f9ade4349443 Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=314f568b84b01f6eac1e4313ca47f9ade4349443 Changed files: drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.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=314f568b84b01f6eac1e4313ca47f9ade4349443 ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2025-38590 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2025-38590 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:N/I:N/A:H The Best / paranoid CVSS vector: AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:H The Best / paranoid CVSS score: 7 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: 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).