[CVE-2026-98462][MODERATE 7.0] Bluetooth: hci_conn: fix CIS hold ownership on reuse

[CVE-2026-98462][MODERATE 7.0] Bluetooth: hci_conn: fix CIS hold ownership on reuse

[AL-KERNEL]

CVE: CVE-2026-98462
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: Bluetooth: hci_conn: fix CIS hold ownership on reuse
Commit: 10f959cb435217b4b407864e5be0eb779e389490
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=10f959cb435217b4b407864e5be0eb779e389490
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98462
Analysis date: Mon, 19 Oct 2026 02:28:00 -0400
ActionableScore: 6
ActionableScore lower bound: 5
Actionable bucket: Actionable Moderate / Strong Important candidate requiring manual review
Manual review required: YES

Summary:
Bluetooth CIS reuse can return an under-referenced active connection and later perform another refcount mutation through the potentially dangling object, creating a local use-after-free and kernel crash risk that requires manual exploitation review.

======================================================================
ABOUT THIS REPORT
======================================================================

The original Linux kernel CVE announcement for CVE-2026-98462 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98462

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-2026-98462	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-98462 CHECK WITH IMPACT FROM ORIG NN IMPORTANT | AEGIS IMPACT MODERATE_KPANIC AEGIS_AS=1 LEGACY=MODERATE_KPANIC RESULT=SAME (final reconciliation agrees; DECISION: both MODERATE_KPANIC)	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-911;*CWE-672;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '7';DESCR 'Bluetooth CIS connection reuse violates the documented reference ownership contract because hci_bind_cis can return an existing connection without taking the ISO layer hold. hci_connect_cis can then drop that reference after hci_conn_link returns an existing link without acquiring another CIS hold. This can cause premature object release and creates a potential use-after-free condition in later Bluetooth connection handling. For the CVSS the PR:L is selected because a local user or process must normally access Bluetooth ISO socket functionality, while full administrative privileges are not clearly required. AC:H is selected because triggering depends on CIS reuse, an existing link or bound state, and a suitable object lifetime window. The issue is not proven to be remotely triggerable by a nearby Bluetooth peer without local participation. Impact is at least a local denial of service via kernel crash and in the worst case may allow confidentiality or integrity impact through use-after-free based memory corruption or privilege escalation. Automated impact decision path. The original SimpleNet classifier selected IMPORTANT. The primary LLM estimated paranoid CVSS 7 and the production ActionableScore returned 6 with recommendation Actionable Moderate / Strong Important candidate requiring manual review. The production rules then evaluated CVSS thresholds and ActionableScore thresholds. Recorded rule transitions were KP and FD specialist checks stage IMPORTANT to MODERATE_KPANIC because KP specialist did not establish a qualifying panic. The legacy KP specialist produced a review artifact and its decision was processed by the applicable KP rules. The legacy FD or DH downgrade specialist has no review artifact for this run. Exact triggered rule markers were DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFKP. The independent weaker Aegis shadow evaluated CVSS 0.0 and ActionableScore 1 with primitive class CLASS_C and manual review RECOMMENDED. Its independently reconciled impact was MODERATE_KPANIC. The shadow agreed with the final production impact and aggregation did not change that impact. The final reported impact is MODERATE_KPANIC.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 6)	MAYBE	REMOTE SIMPLEFIX NETWORK LINUS KPANIC  DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFKP	-	-	checked

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

ActionableScore=6
ActionableScoreLower=5
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=NO
NamespaceScopedPrivilege=NO
ImportantCandidate=YES
ManualReview=REQUIRED
AutoDowngradeAllowed=NO
RequiresHostUID0=NO
AuthorizationDecisionBypass=NO
ProtectedResourceImpact=NONE
B4PendingReview=NO
B4PendingConfirmed=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 or equivalent lifetime violation. On CIS reuse, `hci_bind_cis()` can return a bare lookup result without the hold promised to the ISO caller. If `hci_conn_link()` finds an existing link, it does not acquire the additional CIS hold assumed by `hci_connect_cis()`, but the latter still executes `hci_conn_drop(cis)`.
* +1 Concrete repeated release or refcount operation. The erroneous `hci_conn_drop(cis)` can remove a reference that the current path does not own and may end the CIS lifetime. The same pointer is returned under a contract stating that the ISO layer owns a hold, so its later normal release can perform another lifetime mutation through the under-referenced or potentially dangling object.
* +1 Local unprivileged trigger. Bluetooth ISO socket connection operations are exposed to a local process and do not inherently require host UID 0 or host-administrator-equivalent privilege. Controller availability and a suitable CIS reuse state remain prerequisites.
* +1 Plausible kernel crash or strong DoS. Premature destruction followed by ISO-layer use or release can produce a stale-object dereference, refcount fault, or kernel crash.

Paranoid-only signal:

* +1 Repeated-refcount post-lifetime corruption concern. The later balancing drop is not merely a read of stale storage. It can modify lifetime metadata at an address whose original CIS lifetime may already have ended. This is unresolved Class B leverage, not proof of controlled reclaim or arbitrary write.

Repeated-release checks:

* The first erroneous drop can decrement the CIS lifetime reference and may initiate destruction.
* The exact `cis` pointer remains reachable because it is returned to the ISO caller.
* The caller is expected to release the hold documented by the ownership contract, creating a second lifetime/refcount mutation.
* Concrete repeated release/refcount +1 was awarded.
* Paranoid repeated-refcount leverage +1 was awarded because the later operation may mutate post-lifetime refcount storage.
* FLOOR 5 was not needed because the conservative arithmetic already reached 5.

Not awarded:

* No controlled reclaim, attacker-shaped replacement object, callback control, arbitrary kernel read/write, or demonstrated privilege escalation.
* No separate Structured stale-object leverage point. The relevant structured operation is already represented by the repeated refcount mutation.
* No Weak LPE point for the same repeated-release fact.
* No additional Availability point because it would duplicate the potential kernel crash.
* No Confidentiality or Integrity points merely from theoretical reuse of freed storage.
* No direct remote or Bluetooth-adjacent trigger point. The supplied evidence does not show that a remote peer can complete the trigger without a local ISO operation.
* No race penalty. This is an ownership error on state-dependent reuse paths, not a demonstrated narrow timing race.
* NEW52 B5-R does not apply because there is no retry-accumulated heap OOB write.
* NEW53 B3-PENDING does not apply because there is no shared-backing write.
* NEW54 B4-PENDING does not apply because no authorization decision changes from DENY to ALLOW.
* V33 and V32 do not apply because no attacker-influenced checksum metadata or active packet-header corruption is involved.

3. Reachability analysis

A local process using Bluetooth ISO connection functionality can reach the relevant CIS bind and connect paths. The trigger requires an existing connected or bound CIS, or an already linked CIS that is reused. It also requires a Bluetooth controller and a usable ISO or LE Audio configuration.

Host UID 0 is not inherently required by the ISO socket interface, and the vulnerable ownership operation is below host-admin control. Therefore `HostAdminRequired=NO` and `RequiresHostUID0=NO`.

No namespace-scoped capability is part of the demonstrated trigger. Bluetooth HCI devices are not made independently attacker-controlled merely by creating a user or network namespace, so `NamespaceScopedPrivilege=NO`.

A nearby Bluetooth peer may help create connection state, but the supplied evidence does not establish a peer-only trigger. Direct remote or adjacent-network reachability is therefore not scored.

The bug is state-dependent but not shown to require a narrow race. Repeated bind or connect activity can plausibly recreate CIS reuse states, although reliable post-free heap shaping is not established.

4. Primitive and severity interpretation

The exact primitive is a CIS reference-ownership violation. A borrowed or already-linked connection is treated as if a new hold had been acquired. `hci_connect_cis()` then drops the nonexistent caller-owned hold and returns the same CIS pointer under an ownership contract that permits later use and release.

The CIS is an active structured kernel object rather than fixed stale storage. The vulnerable behavior is not limited to an immediate destructor read. It allows a potentially under-referenced object to escape back to the ISO layer.

This is category B from the repeated-lifetime distinction:

* It is not merely an ordinary stale cleanup dereference.
* The first `hci_conn_drop(cis)` may end the lifetime.
* The same pointer survives and is returned.
* A later caller release can perform another lifetime/refcount mutation through potentially stale storage.

The primitive is CLASS_B, not CLASS_A. Controlled reclaim, refcount takeover, directed corruption, callback influence, and a completed privilege-escalation chain are not demonstrated.

The demonstrated root cause is incorrect ownership and unbalanced reference handling. Kernel crash and DoS are plausible practical consequences. Use-after-free based memory corruption and privilege escalation remain unresolved worst cases rather than established outcomes.

`ImportantCandidate=YES` because a concrete repeated lifetime mutation can occur through an escaping active CIS reference. `AutoDowngradeAllowed=NO` because automatically reducing this to an ordinary crash-only issue would discard unresolved post-lifetime refcount corruption. The likely operational severity is Actionable Moderate, with Important-class review required before downgrading.

5. One-sentence report phrase

Bluetooth CIS reuse can return an under-referenced active connection and later perform another refcount mutation through the potentially dangling object, creating a local use-after-free and kernel crash risk that requires manual exploitation review.

6. Manual review recommendation

MANUAL CHECK REQUIRED

Confirm all ISO callers that consume and release the returned CIS, determine whether the first unbalanced drop can reach the final reference in practical states, and inspect whether CIS slab reuse permits a useful post-free refcount mutation or only an immediate crash.

7. Call-site confidence

Call-site confidence: medium

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

Patch: Bluetooth: hci_conn: fix CIS hold ownership on reuse
Commit: 10f959cb435217b4b407864e5be0eb779e389490
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=10f959cb435217b4b407864e5be0eb779e389490

Commit description:

Commit 69997d5 ("Bluetooth: ISO: handle bound CIS cleanup via
hci_conn") made hci_bind_cis() and hci_connect_cis() return a
connection with one hold for the ISO layer.  hci_bind_cis() currently
takes that hold only after configuring a CIS, so its BT_CONNECTED and
matching BT_BOUND paths return a bare lookup result.  Its configuration
failure path can likewise call hci_conn_drop() before taking a hold.

Take the hold before any state-dependent return or configuration error
so every successful return follows the documented ownership contract
and every error drop is balanced.

hci_connect_cis() also assumes hci_conn_link() always takes a new CIS
hold before dropping the one returned by hci_bind_cis().  However, the
helper returns an existing link without taking another hold.  In that
case, preserve the CIS hold for the caller and drop the redundant LE
hold because the existing link already owns its parent hold.  Returning
early also avoids changing an existing CIS back to BT_CONNECT.

Fixes: 69997d5 ("Bluetooth: ISO: handle bound CIS cleanup via hci_conn")
Cc: stable@vger.kernel.org
Signed-off-by: Aldo Ariel Panzardo <qwe.aldo@gmail.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed files:
  net/bluetooth/hci_conn.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=10f959cb435217b4b407864e5be0eb779e389490

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

The original Linux kernel CVE announcement for CVE-2026-98462 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98462

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: 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).