From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-97603][MODERATE REGULAR] idpf: disable DIM work before freeing q_vectors [ Upstream Date: Fri, 25 Sep 2026 07:12:36 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-97603 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: 4 X-AL-KERNEL-ActionableScore-Lower: unknown X-AL-KERNEL-Commit: 7e4312953788244d47074987d40d76702cf929f1 List-Id: CVE: CVE-2026-97603 Priority: MODERATE REGULAR AL-KERNEL base severity: MODERATE KPANIC flag: NO Patch: idpf: disable DIM work before freeing q_vectors [ Upstream Commit: 7e4312953788244d47074987d40d76702cf929f1 Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=7e4312953788244d47074987d40d76702cf929f1 Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97603 Analysis date: Fri, 25 Sep 2026 07:12:36 -0400 ActionableScore: 4 ActionableScore lower bound: unknown Actionable bucket: Borderline / manual review recommended Manual review required: YES Summary: No one-sentence report phrase was found. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2026-97603 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97603 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-2026-97603 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-97603 CHECK WITH IMPACT FROM ORIG NN MODERATE 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-667;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '7';DESCR 'A use after free in the Intel idpf driver can occur when NAPI queues TX or RX DIM work through net_dim and the associated q_vector is freed before the queued work completes. The stale work item can access fields from freed q_vector memory and write interrupt moderation values through MMIO pointers obtained from that memory. This is a race condition and reliable triggering requires DIM activity together with a concurrent interface stop, vport teardown, or similar control plane operation. For the CVSS the PR:L is used because an attacker typically needs CAP_NET_ADMIN or delegated control over the affected network interface to reliably initiate teardown. Remote packet traffic may help generate DIM events but cannot by itself reliably free the q_vector. The base impact is local denial of service through a kernel crash. The paranoid impact includes confidentiality and integrity because this is a real use after free with stale pointer based MMIO access, although reliable privilege escalation is not demonstrated by the patch. YES REQUIRES MANUAL CHECK';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline / manual review recommended (with actual score 4) YES REMOTE DANGER INIT RACE UAF NETWORK TEST DECREASED_TO_MODERATEREG_BASED_ON_FALSEPOSCHECKOFKP YES NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=4 PrimitiveClass=CLASS_B RealUAF=YES HostAdminRequired=UNKNOWN NamespaceScopedPrivilege=UNKNOWN ImportantCandidate=NO ManualReview=RECOMMENDED AutoDowngradeAllowed=YES 1. ActionableScore * Raw conservative score: 3 * Conservative score: 4 * Raw paranoid score: 3 * Paranoid score: 4 * Applied conservative floor: 3 -> 4. A real asynchronous UAF accesses an active q_vector structure after its lifetime ends. The fixed-storage exception is not fully applicable because deferred DIM work consumes q_vector fields and performs an MMIO register write, even though meaningful post-free object replacement is not established. * Applied paranoid floor: 3 -> 4. * Final recommended bucket: **Borderline / manual review recommended**:: 2. Signal breakdown Positive signals: * +2 Real UAF or active lifetime violation. TX and RX DIM work is embedded in q_vector, can remain queued after q_vectors is freed, and later executes using the freed q_vector address. * +1 Realistic kernel crash or strong DoS. The stale work dereferences fields from freed q_vector memory, including MMIO register pointers. A KASAN ifup ifdown loop under network load is identified as the reproduction condition. A stale dereference or invalid MMIO access can cause an Oops, KASAN failure, or kernel instability. Negative signals: * No timing penalty applied. This is a race, but the supplied context states that repeated ifup ifdown loops under load can expose it. The attacker controlling the relevant lifecycle operation can repeatedly attempt the sequence, so the race is not treated as a one-shot externally uncontrollable event. * No host-admin penalty applied. Interface teardown likely involves CAP_NET_ADMIN, but the supplied evidence does not establish whether the complete trigger requires host namespace CAP_NET_ADMIN over a host-owned IDPF device. * No rare-hardware penalty applied. The bug requires an affected IDPF device and driver, but driver code alone is not evidence of rare or physical-only attacker access. Important non-awarded signals: * Controlled reclaim or replacement of freed q_vector memory is not demonstrated. * Arbitrary or attacker-directed kernel write is not demonstrated. The stale path writes through an MMIO pointer read from freed memory, but attacker control of the pointer or written value is not established. * Structured stale-object exploitation leverage is not awarded. q_vector is an active driver structure, but there is no evidence that the attacker controls replacement contents or can make the stale work consume attacker-shaped fields after reuse. * Weak LPE concern is not awarded. A generic UAF and theoretical allocator reuse are insufficient. * Confidentiality and integrity impact are not awarded. No disclosure, attacker-controlled MMIO target, protected-state modification, or privilege escalation chain is established. * Remote reachability is not established. Network traffic can help create DIM events, but it cannot itself perform the required q_vector free or vport teardown. * No concrete repeated release or refcount mutation exists. The bug is queued deferred work accessing freed storage, not a second put, release, free, or refcount operation on the same dangling reference. * No extra Availability point is awarded because the possible crash is already counted by the reliable DoS signal. 3. Reachability analysis The affected path requires an IDPF network device with dynamic interrupt moderation enabled. The commit states that dynamic ITR is initialized by default, so no special DIM configuration is needed. A practical trigger needs network activity sufficient to accumulate DIM events and a concurrent interface stop, vport teardown, reset, or comparable path that frees q_vectors. Incoming traffic can contribute to DIM scheduling, but it does not by itself free q_vectors. The complete attack sequence is therefore local control-plane plus traffic activity, not directly remote. Changing interface state generally requires CAP_NET_ADMIN in the relevant network namespace. However, the provided patch does not show the netlink capability checks, ownership namespace of the affected device, whether the IDPF device can be moved to an attacker-controlled namespace, or whether the complete stop and teardown path works for a namespace-owned device. Therefore: * HostAdminRequired is UNKNOWN. * NamespaceScopedPrivilege is UNKNOWN. * Local unprivileged reachability is not established. * A delegated network-management service account or constrained container with relevant CAP_NET_ADMIN may be able to participate, but this is not proven for the complete trigger chain. * Prior availability of affected Intel IDPF hardware and the driver is required, but there is no evidence that the attacker needs physical access or firmware compromise. The race is repeatable under load and interface lifecycle cycling, but successful triggering still depends on DIM event accumulation and the workqueue versus teardown timing. 4. Primitive and severity interpretation The exact primitive is an asynchronous workqueue use-after-free. The TX and RX DIM work structures are embedded in q_vector. NAPI polling can queue them through net_dim. The driver can then free the q_vectors array without disabling and synchronizing those work items. The queued work subsequently dereferences stale q_vector fields. This is closer to a fixed asynchronous driver-context UAF than to controlled active-object substitution. The stale object is structured and its fields are actively used, including interrupt register addresses and moderation state. However, the supplied evidence does not show attacker-controlled reclamation, replacement with an attacker-shaped object, callback dispatch, controlled pointer substitution, or a useful post-free kernel data primitive. The demonstrated effect is kernel instability or denial of service through stale-memory dereference or invalid MMIO access. The stale worker can perform an MMIO write, but neither the MMIO target nor the value is shown to be attacker-controlled after free. Therefore this is CLASS_B rather than CLASS_A. ImportantCandidate is NO because the available evidence supports a real UAF and potential crash, but does not establish meaningful exploitation leverage beyond the stale asynchronous access. AutoDowngradeAllowed is YES because the issue is currently a constrained async driver UAF with no demonstrated controlled reclaim, repeated lifetime mutation, disclosure, authorization bypass, or directed corruption primitive. It should nevertheless receive manual review because the UAF accesses a structured q_vector and reaches an MMIO write path. 5. One-sentence report phrase A race in the Intel IDPF driver can allow queued TX or RX DIM work to access a freed q_vector after interface teardown, potentially causing a kernel crash or device instability through stale MMIO-related state. 6. Manual review recommendation MANUAL CHECK RECOMMENDED A real workqueue UAF is established and the stale path reads structured driver state before an MMIO write, but attacker-controlled reclamation, target selection, and below-host-admin reachability are not established. 7. Call-site confidence Call-site confidence: medium ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: idpf: disable DIM work before freeing q_vectors [ Upstream Commit: 7e4312953788244d47074987d40d76702cf929f1 Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=7e4312953788244d47074987d40d76702cf929f1 Changed files: drivers/net/ethernet/intel/idpf/idpf_txrx.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=7e4312953788244d47074987d40d76702cf929f1 ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-97603 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97603 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:N/I:N/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: 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: 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).