[CVE-2026-89970][MODERATE 7.0] nvmet-auth: Synchronize timeout work during SQ teardown [ Upstream

[CVE-2026-89970][MODERATE 7.0] nvmet-auth: Synchronize timeout work during SQ teardown [ Upstream

[AL-KERNEL]

CVE: CVE-2026-89970
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: nvmet-auth: Synchronize timeout work during SQ teardown [ Upstream
Commit: 664022fa1c93f4eba09ef5ee02411b9709dd7a93
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=664022fa1c93f4eba09ef5ee02411b9709dd7a93
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89970
Analysis date: Wed, 16 Sep 2026 08:17:56 -0400
ActionableScore: 6
ActionableScore lower bound: 5
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-2026-89970 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89970

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-2026-89970	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-89970 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:H/I:H/A:H';*CWE-416;CWE-362;CWE-667;Other CVSS 'AV:A/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '7.5';DESCR 'A race in nvmet auth SQ teardown can leave auth_expired_work running after the SQ owner starts freeing authentication state. The old path used non synchronous delayed work cancellation, so a callback that already started may still access struct nvmet_sq after transport teardown has freed or reused the queue. This is a use-after-free candidate in the NVMe target authentication path and can at least cause a kernel crash or denial of service. For the CVSS the PR:N is used because an attacker may only need reachability to the NVMe target authentication flow and no local account on the victim host. The attack is adjacent network scoped because NVMe targets are normally exposed inside storage or data center fabrics rather than the public Internet. Impact is at least adjacent network denial of service via kernel crash and in worst case may allow confidentiality or integrity impact due to use-after-free memory corruption, so manual review is required.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 6)	YES	INIT SIMPLEFIX NVME LINUS  KPANIC INCREASED_TO_MODERATE7_BASED_ON_GUESSCVSS KPANIC INCREASED_TO_HIGH_BASED_ON_ACTIONABLESCOREHIGHERTHAN5 DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFDH	YES	NO	checked

======================================================================
ACTIONABLESCORE ANALYSIS
======================================================================

ActionableScore=6
ActionableScoreLower=5
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=NO
NamespaceScopedPrivilege=NO
ImportantCandidate=YES
ManualReview=REQUIRED
AutoDowngradeAllowed=NO

1. ActionableScore

* Raw conservative score: 5
* Conservative score: 5. No floor or cap changed the score.
* Raw paranoid score: 6
* Paranoid score: 6. No floor or cap changed the score.
* Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**::

2. Signal breakdown

Conservative positive signals:

* +2 Real UAF / active lifetime violation. `cancel_delayed_work()` does not wait for an already running `auth_expired_work`, while SQ teardown can free or reuse the queue containing `struct nvmet_sq`.
* +2 Adjacent/network reachable target protocol path. The vulnerable object belongs to NVMe target authentication state and can plausibly be reached by a remote NVMe initiator with access to the exposed NVMe target fabric.
* +1 Reliable kernel crash / strong DoS. A delayed work callback can access `struct nvmet_sq` after teardown, which directly supports kernel fault or crash.

Conservative negative signals:

* -1 Hard timing / async teardown race. The stale use requires `auth_expired_work` to have already started while SQ teardown proceeds, so the conservative score treats the timing as nontrivial.

Paranoid difference:

* +1 relative to conservative by not applying the timing penalty. A remote initiator can plausibly repeat authentication/session setup and teardown attempts, so the race is not treated as a one-shot or practically unreachable condition in the paranoid score.

Important non-awarded signals:

* Controlled reclaim or attacker-shaped replacement is not demonstrated.
* Arbitrary kernel read/write is not demonstrated.
* Confidentiality and integrity points are not awarded as independent ActionableScore signals merely because UAF memory corruption is theoretically possible.
* Structured stale-object leverage is not separately awarded because the supplied patch does not show attacker-controlled post-free fields, callback control, or meaningful replacement-object consumption.
* Weak LPE is not separately awarded solely from the UAF class.
* Extra Availability is not awarded because the same kernel crash is already counted once.
* Concrete repeated release/refcount is not awarded. The fix is synchronization of delayed work, not a visible first put followed by a second put on the exact same dangling reference.
* FLOOR 5 is effectively satisfied by raw arithmetic, but no separate floor adjustment is needed.

Repeated-release/refcount checklist:

* First release can end lifetime: not the relevant primitive shown here.
* Exact same reference survives to a second lifetime mutation: not established.
* Second operation is a real lifetime/refcount mutation: not established.
* +1 Concrete repeated release/refcount awarded: no.
* Paranoid repeated-refcount leverage awarded: no.
* FLOOR 5 applied: no score change needed.

3. Reachability analysis

The evaluated attacker is a remote or adjacent NVMe initiator that can reach the NVMe target authentication flow. The vulnerable host must already be configured as an NVMe target with authentication and an exposed transport, which is prior deployment setup rather than a host-admin action required from the attacker at exploit time.

Host-admin-equivalent privilege is not required for the attacker once the target service is reachable. This supports `HostAdminRequired=NO`. This is not a namespace-scoped local capability issue, so `NamespaceScopedPrivilege=NO`.

The likely network scope is adjacent or internal rather than Internet-wide. NVMe targets are normally deployed in storage fabrics or data-center networks. The trigger depends on an async race between authentication timeout work and SQ teardown. Conservative scoring treats this as hard timing, while paranoid scoring considers repeated connection/authentication/teardown attempts sufficient to keep the issue actionable.

No special rare hardware requirement is established beyond use of the NVMe target stack and its authentication support.

4. Primitive and severity interpretation

The exact primitive is an async post-teardown use-after-free candidate: `auth_expired_work` may keep running after the queue owner begins SQ destruction, and the work can access `struct nvmet_sq` after the queue has been freed or reused.

This is an active structured kernel object, not fixed stale storage. However, the supplied diff does not show controlled reclaim, attacker-controlled replacement contents, function-pointer use, arbitrary R/W, or a concrete second refcount/lifetime mutation. Therefore the primitive is `CLASS_B`, not `CLASS_A`.

The demonstrated consequence is denial of service via kernel crash or fault. The speculative consequence is stronger exploitation from UAF memory corruption, but the supplied evidence does not prove a usable confidentiality, integrity, or privilege-escalation primitive.

`ImportantCandidate=YES` because this is a real network-adjacent active-object UAF in kernel NVMe target authentication code and the post-lifetime behavior remains security-relevant enough to require Important-vs-Moderate review. `AutoDowngradeAllowed=NO` because automatic closure would discard an unresolved concrete lifetime corruption primitive reachable through a storage protocol path.

Likely operational severity is actionable Moderate at minimum, with Strong Important candidacy depending on whether callback internals or allocator behavior provide useful post-free field consumption or reuse leverage.

5. One-sentence report phrase

A race in NVMe target authentication SQ teardown can leave `auth_expired_work` running after `struct nvmet_sq` teardown, allowing an adjacent network initiator with access to the NVMe target to trigger a use-after-free and kernel crash, with stronger memory-corruption impact requiring manual review.

6. Manual review recommendation

MANUAL CHECK REQUIRED

Reason: real async UAF of an active structured kernel object is established, and the issue is reachable through the NVMe target authentication path. DoS is directly supported, while exploitation leverage beyond crash is unresolved and should not be auto-downgraded.

7. Call-site confidence

Call-site confidence: medium

======================================================================
UPSTREAM PATCH SUMMARY
======================================================================

Patch: nvmet-auth: Synchronize timeout work during SQ teardown [ Upstream
Commit: 664022fa1c93f4eba09ef5ee02411b9709dd7a93
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=664022fa1c93f4eba09ef5ee02411b9709dd7a93

Changed files:
  drivers/nvme/target/auth.c
  drivers/nvme/target/core.c
  drivers/nvme/target/nvmet.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=664022fa1c93f4eba09ef5ee02411b9709dd7a93

======================================================================
DETAILED REPORT METHODOLOGY
======================================================================

The original Linux kernel CVE announcement for CVE-2026-89970 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89970

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:A/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H
  The Best / paranoid CVSS vector: AV:A/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H
  The Best / paranoid CVSS score: 7.5

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: 5
  Paranoid ActionableScore: 6

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).