From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-98016][MODERATE 7.0] net/mlx5e: Fix use-after-free race in sample_restore_put() [ Upstream Date: Sat, 26 Sep 2026 16:02:59 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-98016 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: unknown X-AL-KERNEL-Commit: 3efd1a1938cbb33c53b0d75e55b6c0fe2ebad79a List-Id: CVE: CVE-2026-98016 Priority: MODERATE 7.0 AL-KERNEL base severity: MODERATE KPANIC flag: YES Patch: net/mlx5e: Fix use-after-free race in sample_restore_put() [ Upstream Commit: 3efd1a1938cbb33c53b0d75e55b6c0fe2ebad79a Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=3efd1a1938cbb33c53b0d75e55b6c0fe2ebad79a Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98016 Analysis date: Sat, 26 Sep 2026 16:02:59 -0400 ActionableScore: 5 ActionableScore lower bound: unknown 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-2026-98016 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98016 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-2026-98016 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-98016 CHECK WITH IMPACT FROM ORIG NN IMPORTANT Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H';*CWE-416;CWE-362;CWE-415;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '7';DESCR 'A race in sample_restore_put can access a shared mlx5e sample restore object after another teardown thread has removed and freed it. One thread can decrement restore.count and release restore_lock while another thread reaches zero, destroys the associated rule and modify header, and frees the restore object before the first thread reads the count again. This creates an explicit use after free read and may also cause repeated cleanup or a double free if the stale count is observed as zero. For the CVSS the PR:L reflects that a local process normally needs CAP_NET_ADMIN or delegated TC control to create and concurrently remove the affected sampling rules, but it does not require full control of the host. The issue is not directly network reachable through ordinary packet traffic because the vulnerable path is TC rule teardown in the network control plane. Impact is at least a local denial of service through kernel memory corruption and a crash. In the worst defensible case the possible double free may affect confidentiality and integrity or support privilege escalation. The race and memory corruption require manual review.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 5) SKIP CVE-2026-98016 UNKNOWN SKIP No affected files built, so skip this CVE NO - - unknown MAYBE REMOTE RACE UAF NETWORK KPANIC KPANIC INCREASED_TO_HIGH_BASED_ON_GUESSCVSS DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFDH - - checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=5 PrimitiveClass=CLASS_B RealUAF=YES HostAdminRequired=UNKNOWN NamespaceScopedPrivilege=UNKNOWN ImportantCandidate=YES ManualReview=REQUIRED AutoDowngradeAllowed=NO 1. ActionableScore * Raw conservative score: 4 * Conservative score: 5 * Raw paranoid score: 5 * Paranoid score: 5 * Conservative floor applied: 4 -> 5 under the concrete repeated-lifetime-operation UAF floor. * No paranoid floor or cap changed the raw paranoid score. * Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**:: 2. Signal breakdown * +2 Real UAF. One teardown thread can free `restore` after another thread drops `restore_lock` but before that thread rereads `restore->count`. * +1 Concrete repeated release. After the first thread frees `restore`, the exact stale reference can reach repeated destruction of `rule`, `modify_hdr`, and `restore`. * +1 Reliable kernel crash or strong DoS. Repeated destruction and `kfree()` can produce double-free corruption, allocator failure, or a kernel crash. * +1 Paranoid repeated-lifetime leverage. The stale path can perform actual destructor and free operations after the original object lifetime has ended. This is more concrete than theoretical allocator reuse. * No privilege penalty. Host-admin-equivalent privilege is not positively established by the supplied patch. * No race penalty. Concurrent teardown is required, but the evidence does not establish a narrow, one-shot, or externally uncontrollable race. * No remote-reachability bonus. Ordinary packet traffic does not directly reach the vulnerable teardown operation. * No structured stale-object leverage bonus. Although stale `rule` and `modify_hdr` fields are consumed, attacker-controlled reclaim or replacement contents are not established. * No Weak LPE bonus. It would duplicate the concrete repeated-release signal. * No strong exploitation-leverage bonus. Controlled reclaim, directed corruption, arbitrary write, callback control, and refcount takeover are not demonstrated. * No separate C or I bonuses. Double-free corruption creates unresolved risk, but no disclosure or unauthorized state modification is independently demonstrated. * No extra Availability point. The possible allocator corruption and crash are the same availability consequence already counted. Repeated-release checks: * The first cleanup can end the lifetime of `restore`: yes, through `kfree(restore)`. * The exact same reference survives in another invocation: yes. * The later operation mutates lifetime state: yes, through repeated destructor operations and `kfree(restore)`. * Concrete repeated release/refcount +1 awarded: yes. * Paranoid repeated-refcount/lifetime leverage +1 awarded: yes. * FLOOR 5 applied: yes, because the exact dangling reference can undergo a second concrete lifetime-ending operation. 3. Reachability analysis The trigger requires concurrent teardown of TC sample rules that share one mlx5e restore context. An actor able to configure and remove the affected hardware-offloaded TC rules can plausibly drive both teardown operations, but the supplied evidence does not resolve whether this always requires host `CAP_NET_ADMIN`, can be delegated to a network-management service, or works completely within an attacker-created namespace. `HostAdminRequired` is therefore UNKNOWN rather than YES or NO. `NamespaceScopedPrivilege` is also UNKNOWN because namespace-aware TC control alone does not prove that mlx5 eswitch configuration, rule creation, and teardown are all available below host administration. The path requires mlx5 hardware and the relevant TC sampling/offload configuration. No rare-hardware penalty is applied merely because this is a driver path. Prior device and offload setup may require administration, but the supplied evidence does not establish the complete deployment boundary. Direct remote triggering through packet traffic is not shown. The race may be repeatable by issuing concurrent rule-removal operations. Nothing in the supplied material establishes sufficiently narrow or uncontrollable timing to justify a race penalty. 4. Primitive and severity interpretation The exact primitive is a real UAF followed by a possible double cleanup and double free. `restore` is an active structured object containing a reference count, hash membership, a flow rule, and a modify-header resource. This is not merely an ordinary stale cleanup dereference that reads freed memory and immediately faults. Another thread can complete the final cleanup and free the object, after which the stale invocation rereads `restore->count`. If that stale read produces zero, the invocation proceeds to destroy nested resources and call `kfree()` on the same object again. This is a second lifetime-ending operation through a potentially dangling reference. The primitive remains CLASS_B because controlled reclaim, arbitrary read/write, callback hijacking, and reliable object substitution are not demonstrated. The demonstrated practical consequence is kernel corruption and likely crash or DoS. Privilege escalation is an unresolved possibility rather than an established consequence. `ImportantCandidate=YES` because the concrete post-lifetime destructor and free operations leave meaningful exploitation leverage unresolved. `AutoDowngradeAllowed=NO` because automatically treating the issue as an ordinary crash-only cleanup bug would discard the established double-free mechanism. The likely operational severity is Actionable Moderate, with Important classification requiring analysis of allocator reuse, nested-resource destruction, and the actual TC privilege boundary. 5. One-sentence report phrase A race in mlx5e TC sample rule teardown can reread and destroy an already freed restore context, creating a concrete UAF and possible double free that requires manual exploitation and privilege-boundary review. 6. Manual review recommendation MANUAL CHECK REQUIRED The exact stale reference can perform repeated destructor and free operations, while controlled reclaim and the complete TC privilege boundary remain unresolved. 7. Call-site confidence Call-site confidence: high ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: net/mlx5e: Fix use-after-free race in sample_restore_put() [ Upstream Commit: 3efd1a1938cbb33c53b0d75e55b6c0fe2ebad79a Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=3efd1a1938cbb33c53b0d75e55b6c0fe2ebad79a Changed files: drivers/net/ethernet/mellanox/mlx5/core/en/tc/sample.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=3efd1a1938cbb33c53b0d75e55b6c0fe2ebad79a ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-98016 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98016 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:L/UI:N/S:U/C:L/I:L/A:H The Best / paranoid CVSS vector: AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/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: unknown 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).