From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-98153][LOW] nvme: fix racy access to FDP placement id array [ Upstream Date: Fri, 25 Sep 2026 10:44:49 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-98153 X-AL-KERNEL-Priority: LOW X-AL-KERNEL-Severity: LOW X-AL-KERNEL-Base-Severity: LOW X-AL-KERNEL-KPANIC: NO X-AL-KERNEL-ActionableScore: 2 X-AL-KERNEL-ActionableScore-Lower: 1 X-AL-KERNEL-Commit: e0a23e3c8c4c3ce11529fa70d2292b4e56d143ba List-Id: CVE: CVE-2026-98153 Priority: LOW AL-KERNEL base severity: LOW KPANIC flag: NO Patch: nvme: fix racy access to FDP placement id array [ Upstream Commit: e0a23e3c8c4c3ce11529fa70d2292b4e56d143ba Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=e0a23e3c8c4c3ce11529fa70d2292b4e56d143ba Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98153 Analysis date: Fri, 25 Sep 2026 10:44:49 -0400 ActionableScore: 2 ActionableScore lower bound: 1 Actionable bucket: Ordinary Moderate Manual review required: NO Summary: A race during NVMe FDP discovery can concurrently initialize shared namespace placement ID state and cause inconsistent write stream configuration or a leaked allocation, but no UAF, OOB access, reliable crash, or privilege-escalation primitive is established. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2026-98153 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98153 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: LOW Manual review required: NO A detailed explanation of the methodology and priority rules is included at the end of this message. ====================================================================== AL-KERNEL CLASSIFICATION RESULT ====================================================================== CVE-2026-98153 LOW https://www.kernelcve.org/list/?q=CVE-2026-98153 CHECK WITH IMPACT FROM ORIG NN MODERATE Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:A/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:L';CWE-362;CWE-667;CWE-364;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:L';BEST CVSS score: '3.1';DESCR 'A race in NVMe Flexible Data Placement discovery occurs because nvme_query_fdp_info can run concurrently for multiple paths of the same shared namespace while updating head nr_plids and head plids. Concurrent path discovery can cause inconsistent shared placement ID state, leaked allocations, or incorrect write stream limits. The patch context does not demonstrate a use after free, out of bounds access, information leak, data corruption, or a reliable kernel crash. For the CVSS the PR:N in the paranoid AV:A scenario reflects an attacker that can reach and control an NVMe over Fabrics target on the storage network without local code execution on the host. Exploitation requires a multipath NVMe configuration and a timing race during namespace discovery. This is not a public Internet issue in typical deployments because NVMe over Fabrics is normally limited to a local or trusted storage network.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Ordinary Moderate (with actual score 2) YES READ INIT SIMPLEFIX NVME RACE HARDWARE LINUS DECREASED_TO_LOW_FROM_MODERATE_BASED_ON_GUESSCVSS YES NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=2 ActionableScoreLower=1 PrimitiveClass=CLASS_C RealUAF=NO HostAdminRequired=UNKNOWN NamespaceScopedPrivilege=UNKNOWN ImportantCandidate=NO ManualReview=NO AutoDowngradeAllowed=YES ## 1. ActionableScore * Raw conservative score: 1 * Conservative score: 1 * Raw paranoid score: 2 * Paranoid score: 2 * Final recommended bucket: **Ordinary Moderate**:: No floor or cap was applied. ## 2. Signal breakdown Conservative signals: * +1 Concrete persistent kernel resource leak. Concurrent executions of `nvme_query_fdp_info()` can each allocate and assign `head->plids`. One assignment can overwrite the other pointer in the shared namespace head, making the earlier allocation unreachable by normal cleanup. * +0 No demonstrated kernel memory corruption. The patch does not establish a UAF, double free, stale dereference, OOB access, invalid free, arbitrary write, or refcount mutation. * +0 No demonstrated reliable kernel crash or host-wide DoS. Inconsistent FDP parameters or a lost allocation are possible outcomes, but the supplied patch does not show a crash path. * +0 No hard-race penalty. The bug is a race, but the supplied evidence does not establish a narrow, externally uncontrollable, or one-shot timing condition. * +0 No privilege penalty. The complete privilege boundary for initiating multipath namespace discovery is not established. * +0 No niche or hardware penalty. FDP and NVMe multipath are feature-dependent, but no evidence shows that the attacker-facing path requires rare physical hardware. Paranoid-only signal: * +1 Firmware-mediated external influence. FDP data is retrieved from an NVMe controller through `nvme_get_features()`. A malicious or untrusted NVMe over Fabrics endpoint could potentially influence controller-provided FDP information, but the supplied evidence does not establish an attacker-controlled endpoint as a complete trigger path. Important non-awarded signals: * Controlled reclaim is not demonstrated. * Arbitrary or directed kernel write is not demonstrated. * A UAF or post-lifetime object access is not demonstrated. * An OOB read or write is not demonstrated. * Integrity impact is not awarded. Incorrect FDP placement IDs or write-stream limits do not by themselves show modification of protected data. * Confidentiality impact is not awarded. No disclosure path is shown. * Reliable kernel crash or distinct availability impact is not awarded. * Direct remote reachability is not established. NVMe over Fabrics deployment alone does not prove an attacker can reach or control the affected controller. * Local unprivileged reachability is not established. * No attacker-controlled payload, target, reuse, callback, or active-object replacement primitive is shown. ## 3. Reachability analysis The affected code runs while NVMe namespace information is discovered and a shared `nvme_ns_head` is created for an NVMe multipath namespace. The vulnerable condition requires at least two paths concurrently scanning the same namespace while FDP is enabled by the controller. The supplied patch does not establish that an ordinary local user can create the required NVMe multipath topology, connect host paths to the controller, trigger namespace discovery, or repeatedly cause the affected initialization sequence. Therefore, below-host-admin reachability is not positively established. An NVMe over Fabrics controller can provide FDP feature information over a storage transport. However, a complete attacker model is missing. The evidence does not establish that an attacker can operate a reachable malicious NVMe target, cause the host to connect to it, configure multiple paths for the same namespace, and trigger the concurrent discovery sequence without prior privileged setup. The relevant capability checks and device-management stages are not visible. Namespace-scoped privilege is therefore unknown. The existence of network namespaces or NVMe network transports would not by itself make the complete trigger namespace-scoped. Race repeatability is also unresolved. The patch confirms unsynchronized concurrent initialization, but does not show that an attacker can reliably induce repeated scans of the same namespace. ## 4. Primitive and severity interpretation The exact primitive is an unsynchronized concurrent initialization of shared namespace FDP state: `head->nr_plids` `head->plids` `head->write_stream_granularity` Before the fix, `nvme_query_fdp_info()` was called per path and could concurrently populate fields belonging to one shared `nvme_ns_head`. One path could observe or publish an incomplete state while another path performs the same initialization. Multiple allocations may also race, with one `head->plids` assignment replacing another and leaving an allocation unreachable. This is not a demonstrated active stale-object, UAF, double release, refcount corruption, OOB access, or ownership violation. The patch moves initialization to `nvme_alloc_ns_head()`, which is called once per shared namespace head, and stores write-stream granularity in that same shared object. The demonstrated consequence is inconsistent FDP configuration and a possible leaked placement ID allocation. The speculative consequence is incorrect block-layer write-stream configuration. A kernel crash, data corruption, privilege escalation, or arbitrary kernel read/write is not supported by the supplied patch. The issue is classified as CLASS_C because it is a shared-state race producing invalid or inconsistent state without an independently useful corruption primitive. It is not an Important candidate. Automatic downgrade is allowed because no meaningful unresolved memory-corruption, ownership, authorization, or security-boundary primitive is established. Likely operational severity is Low-like to ordinary Moderate. Review should focus on systems using both NVMe multipath and FDP, but the patch alone does not justify treating this as a kernel memory-corruption vulnerability. ## 5. One-sentence report phrase A race during NVMe FDP discovery can concurrently initialize shared namespace placement ID state and cause inconsistent write stream configuration or a leaked allocation, but no UAF, OOB access, reliable crash, or privilege-escalation primitive is established. ## 6. Manual review recommendation NO MANUAL CHECK NEEDED The supplied patch establishes a configuration-state race and possible allocation leak, but does not establish memory corruption, a security-boundary bypass, a reliable host DoS, or a complete attacker-controlled NVMe fabrics trigger path. ## 7. Call-site confidence Call-site confidence: medium ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: nvme: fix racy access to FDP placement id array [ Upstream Commit: e0a23e3c8c4c3ce11529fa70d2292b4e56d143ba Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=e0a23e3c8c4c3ce11529fa70d2292b4e56d143ba Changed files: drivers/nvme/host/core.c drivers/nvme/host/nvme.h 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=e0a23e3c8c4c3ce11529fa70d2292b4e56d143ba ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-98153 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98153 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:L The Best / paranoid CVSS vector: AV:A/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:L The Best / paranoid CVSS score: 3.1 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: 1 Paranoid ActionableScore: 2 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: LOW KPANIC detected for this report: NO Published priority for this report (same as in Subject): LOW 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).