[CVE-2026-90091][MODERATE REGULAR] Bluetooth: L2CAP: fix race l2cap_sock_cleanup_listen() vs. put_chan [ Upstream

[AL-KERNEL <alkernel@kernelcve.org>]

CVE: CVE-2026-90091
Priority: MODERATE REGULAR
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: Bluetooth: L2CAP: fix race l2cap_sock_cleanup_listen() vs. put_chan [ Upstream
Commit: 827de6bd2865b22aaabd554540def3b8a33018ab
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=827de6bd2865b22aaabd554540def3b8a33018ab
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90091
Analysis date: Fri, 18 Sep 2026 02:32:16 -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-2026-90091 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90091

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-2026-90091	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-90091 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:H';CWE-362;**CWE-476;CWE-667;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '5.3';DESCR 'A race in Bluetooth L2CAP socket teardown allows l2cap_sock_cleanup_listen to read the channel pointer while l2cap_sock_kill concurrently clears it through l2cap_sock_put_chan. The observed primitive is a NULL pointer dereference rather than a demonstrated use after free or controlled memory corruption, and the resulting kernel Oops or panic causes denial of service. A nearby Bluetooth peer may contribute to the trigger by causing a disconnect confirmation while listener cleanup is running, but exploitation depends on an active L2CAP connection and a narrow timing window. For the CVSS the PR:N value is used in the paranoid adjacent network scenario because a nearby Bluetooth peer does not need an account or local execution context on the target. The realistic local scenario instead requires a local process with access to Bluetooth L2CAP sockets and is represented by PR:L in the base vector. No confidentiality or integrity impact is supported because the analyzed path provides only a NULL dereference and no stale object access, information disclosure, or write primitive.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline / manual review recommended (with actual score 4)  SKIP CVE-2026-90091 UNKNOWN SKIP The Fixes patch not applied yet, so unlikely that actual: url YES  NO NO unknown 	MAYBE	REMOTE LOCK DEADLOCK DANGER NULLPTR RACE UAF NETWORK SYZBOT  DECREASED_TO_MODERATE_REGULAR_FROM_MODERATE7_BASED_ON_GUESSCVSS_AND_CIANNH	-	-	checked

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

ActionableScore=4
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=NO
NamespaceScopedPrivilege=UNKNOWN
ImportantCandidate=NO
ManualReview=RECOMMENDED
AutoDowngradeAllowed=YES

1. ActionableScore

* Raw conservative score: 3.
* Conservative score: 4.
* Raw paranoid score: 4.
* Paranoid score: 4.
* Applied conservative REAL UAF FLOOR 4: 3 -> 4.
* Final recommended bucket: **Borderline / manual review recommended**::

2. Signal breakdown

Conservative signals:

* +2 Real UAF or lifetime violation. Task 1 can load the old non-NULL `chan` immediately before Task 2 clears the field and drops its owning reference. If that put ends the channel lifetime, `l2cap_chan_hold_unless_zero()` subsequently accesses the freed `kref`.
* +1 Reliable kernel crash. The explicitly reported outcome is a NULL pointer dereference when Task 1 observes the cleared pointer.
* +1 Local unprivileged trigger. Normal L2CAP socket creation, connection handling, and socket release can be exercised below host-admin privilege when a Bluetooth adapter is available.
* -1 Narrow teardown race. The vulnerable result requires listener cleanup to overlap channel teardown between loading or clearing the pointer and acquiring a reference.

Paranoid-only signal:

* +1 Firmware-mediated external influence. A nearby Bluetooth peer can contribute to the required interleaving by causing the disconnect-confirmation path through the Bluetooth controller.
* The same -1 timing penalty remains because remote disconnection alone does not guarantee the required overlap.

Important non-awarded signals:

* Controlled reclaim or attacker-shaped replacement contents are not demonstrated.
* Structured stale-object leverage is not awarded because the stale access is limited to acquiring a channel reference during cleanup.
* Strong or weak LPE leverage is not awarded merely from the possible UAF.
* Direct remote reachability +2 is not awarded because the complete trigger also includes local listener release and controller-mediated processing.
* Extra Availability is not awarded because the Oops or panic is the same crash already counted.
* Confidentiality and integrity signals are not awarded because no disclosure, directed write, callback use, or policy bypass is established.
* No concrete repeated release/refcount signal applies. The first `l2cap_put_chan()` may end lifetime, but the later operation is an attempted reference acquisition by another path, not a second put or release of the exact dangling reference.
* The repeated-refcount paranoid bonus and FLOOR 5 are therefore withheld.

3. Reachability analysis

A local process can create and release an L2CAP listening socket while disconnect processing runs on the Bluetooth device workqueue. A nearby peer may influence the second path by disconnecting at the relevant time, but the supplied evidence does not establish a peer-only trigger.

Host-admin-equivalent privilege is not required for the ordinary L2CAP socket lifecycle once an accessible Bluetooth adapter is configured. Prior system setup is normally required to enable and expose the adapter, but that is not the same as requiring the attacker to hold host-admin privilege.

The evidence does not establish that the complete Bluetooth operation is available through attacker-created user or network namespaces, so `NamespaceScopedPrivilege` remains UNKNOWN. No rare hardware penalty is applied because ordinary Bluetooth hardware is sufficient. The race is narrow, although repeated connection and teardown attempts may improve reproducibility.

4. Primitive and severity interpretation

The explicit demonstrated primitive is a NULL pointer dereference after `chan` is cleared. The same unsynchronized sequence also directly implies a possible UAF: Task 1 may retain the previous non-NULL pointer while Task 2 clears it and drops the final owning reference, after which `kref_get_unless_zero()` accesses lifetime metadata through stale storage.

The channel is an active structured object, but the stale operation is narrowly confined to cleanup-time reference acquisition. There is no demonstrated controlled reclaim, structured field consumption after replacement, second release, arbitrary read or write, callback dispatch, or privilege escalation.

This is therefore constrained CLASS_B rather than CLASS_A. The practical consequence is kernel Oops or panic and denial of service. Stronger exploitation is theoretically possible only if additional reclaim or reuse control exists, which the supplied evidence does not establish.

`ImportantCandidate=NO` because meaningful exploitation leverage beyond the cleanup fault is not currently demonstrated. `AutoDowngradeAllowed=YES` because this remains a constrained crash-oriented UAF without repeated lifetime mutation or an unresolved boundary primitive. Manual review is still recommended because the old non-NULL pointer outcome is more serious than the commit message explicit NULL observation.

5. One-sentence report phrase

A race between L2CAP listener cleanup and channel release can cause either a NULL pointer dereference or a stale refcount access after the channel lifetime ends, resulting primarily in a local or Bluetooth-adjacent kernel denial of service.

6. Manual review recommendation

MANUAL CHECK RECOMMENDED

Confirm whether the affected kernel permits a reliable non-NULL stale-pointer interleaving and whether unprivileged users can exercise the complete L2CAP listening and teardown sequence under the product Bluetooth policy.

7. Call-site confidence

Call-site confidence: high

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

Patch: Bluetooth: L2CAP: fix race l2cap_sock_cleanup_listen() vs. put_chan [ Upstream
Commit: 827de6bd2865b22aaabd554540def3b8a33018ab
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=827de6bd2865b22aaabd554540def3b8a33018ab

Changed files:
  include/net/bluetooth/l2cap.h
  net/bluetooth/l2cap_sock.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=827de6bd2865b22aaabd554540def3b8a33018ab

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

The original Linux kernel CVE announcement for CVE-2026-90091 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90091

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:A/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H
  The Best / paranoid CVSS score: 5.3

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).