public inbox for [email protected]
 help / color / mirror / Atom feed
This is experimental automated Linux kernel CVE triage research. Results are heuristic and may be incorrect. This site is not an official vendor advisory or severity source.
From: AL-KERNEL <[email protected]>
To: [email protected]
Subject: [CVE-2026-97408][MODERATE REGULAR] Bluetooth: L2CAP: validate connectionless PSM length [ Upstream
Date: Thu, 24 Sep 2026 15:11:24 -0400	[thread overview]
Message-ID: <[email protected]> (raw)

CVE: CVE-2026-97408
Priority: MODERATE REGULAR
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: Bluetooth: L2CAP: validate connectionless PSM length [ Upstream
Commit: 3502ede8aed3c3ee1786d952a11054ac737d1a38
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=3502ede8aed3c3ee1786d952a11054ac737d1a38
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97408
Analysis date: Thu, 24 Sep 2026 15:11:24 -0400
ActionableScore: 2
ActionableScore lower bound: unknown
Actionable bucket: Ordinary Moderate
Manual review required: NO

Summary:
No one-sentence report phrase was found.

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

The original Linux kernel CVE announcement for CVE-2026-97408 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97408

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: NO

A detailed explanation of the methodology and priority rules is included
at the end of this message.

======================================================================
AL-KERNEL CLASSIFICATION RESULT
======================================================================

CVE-2026-97408	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-97408 CHECK WITH IMPACT FROM ORIG NN IMPORTANT	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:A/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L';CWE-125;*CWE-20;CWE-1284;Other CVSS 'AV:A/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:N';BEST CVSS score: '4.3';DESCR 'A Bluetooth L2CAP connectionless frame can cause an out of bounds read because l2cap_recv_frame reads a two byte PSM from skb data without first verifying that the connectionless payload contains two bytes. A nearby Bluetooth attacker can send a malformed frame with a zero or one byte payload, causing the parser to use bytes beyond the advertised skb payload as part of the PSM. For the CVSS the PR:N because no local account or local code execution on the target is required. The attack is adjacent network reachable over Bluetooth and requires radio range rather than public Internet access. The patch context does not demonstrate an information disclosure primitive, an out of bounds write, use after free, or privilege escalation. Impact is primarily malformed packet handling and a limited denial of service risk in the conservative interpretation.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Ordinary Moderate (with actual score 2)	YES	REMOTE NETWORK SKB PACKET  LOWERED_FROM_HIGH_BASED_ON_GUESSCVSS  DECREASED_TO_MODERATEREG_BASED_ON_ACTIONABLESCORELESSTHAN4	NO	NO	checked

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

ActionableScore=2
PrimitiveClass=CLASS_C
RealUAF=NO
HostAdminRequired=NO
NamespaceScopedPrivilege=UNKNOWN
ImportantCandidate=NO
ManualReview=RECOMMENDED
AutoDowngradeAllowed=YES

1. ActionableScore

* Raw conservative score: 2
* Conservative score: 2
* Raw paranoid score: 2
* Paranoid score: 2
* Final recommended bucket: **Ordinary Moderate**::

The conservative and paranoid scores are identical. The patch establishes a limited out of bounds read of skb tailroom, but does not establish disclosure, memory modification, crash, or a reusable corruption primitive.

2. Signal breakdown

* +2 Directly adjacent network reachable input. A nearby Bluetooth peer can supply the malformed L2CAP connectionless frame. The trigger is Bluetooth adjacent network traffic and does not require a local account on the victim.
* +0 Weak invalid read concern. The parser reads two bytes for the PSM when the advertised skb payload has zero or one byte. The extra byte can originate from skb tailroom.
* +0 No demonstrated information disclosure. The tailroom byte is consumed internally as part of a PSM lookup value and is not shown to be copied back to the attacker, userspace, or a network response.
* +0 No demonstrated write primitive. The patch adds a length check before get_unaligned and skb_pull. It does not indicate an out of bounds write, arbitrary write, refcount corruption, or object replacement.
* +0 No demonstrated reliable kernel crash or strong DoS. The supplied reproducer reaches the unchecked read, but does not demonstrate an Oops, panic, persistent resource exhaustion, or host-wide unavailability.
* +0 No UAF or repeated release/refcount operation. The skb is read and then passed to the connectionless channel path. There is no evidence of a freed object being reused or released again.
* +0 No privilege escalation, security-boundary bypass, or attacker-controlled active-object reuse is established.
* No host-admin penalty was applied. Host administrator privilege is not required for a nearby Bluetooth attacker to transmit the malformed frame.
* No niche subsystem penalty was applied. Bluetooth is not treated as rare solely because it is not universally enabled. Actual reachability still depends on Bluetooth being enabled and accepting the relevant traffic.

3. Reachability analysis

The evaluated attacker is a nearby Bluetooth peer, not a local host user. The supplied context states that a VHCI-backed QEMU reproducer can inject a one-byte connectionless payload and reach the unchecked read. The same malformed L2CAP frame format is attacker controlled over a Bluetooth link.

No host administrator privilege, local code execution, CAP_NET_ADMIN, or namespace capability is required on the target to send Bluetooth traffic. Therefore HostAdminRequired is NO.

NamespaceScopedPrivilege is UNKNOWN because namespaces are not relevant to the supplied Bluetooth receive-path trigger. There is no requirement to create a network namespace, Bluetooth namespace, virtual interface, or privileged controller configuration in the evidence.

The practical prerequisite is Bluetooth connectivity to the target and reachability of the affected L2CAP connectionless receive path. This is adjacent network exposure, not public Internet exposure. No special hardware beyond an available Bluetooth controller and radio-range access is established. No race condition is involved.

4. Primitive and severity interpretation

The exact primitive is a bounded logical out of bounds read. l2cap_recv_frame reads a two-byte PSM through get_unaligned when skb len is shorter than L2CAP_PSMLEN_SIZE. A zero-byte or one-byte connectionless payload can therefore cause the PSM value to include bytes beyond the advertised skb payload, potentially from allocated skb tailroom.

This is not a UAF, active structured-object lifetime violation, refcount mutation, out of bounds write, type confusion, or authorization bypass. The stale or extra byte is used as part of a PSM value and the supplied patch does not establish that this value creates attacker-controlled memory access, disclosure, or persistent corruption.

The appropriate class is CLASS_C because the demonstrated behavior is malformed parser state and an invalid read without an independently useful corruption primitive. The demonstrated consequence is that a malformed Bluetooth frame reaches an unchecked read. A kernel crash or limited denial of service is possible only as an unresolved downstream concern and is not demonstrated by the supplied evidence.

ImportantCandidate is NO because no meaningful exploitation leverage beyond the limited invalid read is established. AutoDowngradeAllowed is YES because there is no demonstrated UAF, write primitive, disclosure path, repeated lifetime mutation, security-boundary bypass, or controlled active-object corruption.

Likely operational severity is low to ordinary Moderate. Manual review is still recommended to confirm whether the affected Bluetooth configuration exposes connectionless L2CAP traffic to untrusted nearby devices and whether downstream l2cap_conless_channel handling has additional consequences for malformed short skbs.

5. One-sentence report phrase

A nearby Bluetooth attacker can send a malformed connectionless L2CAP frame with a payload shorter than the required two byte PSM, causing l2cap_recv_frame to read beyond the advertised skb payload, but the supplied patch does not establish information disclosure, memory corruption, privilege escalation, or a reliable kernel crash.

6. Manual review recommendation

MANUAL CHECK RECOMMENDED

The issue is externally reachable over Bluetooth and contains an unchecked out of bounds read, but the supplied evidence does not demonstrate a useful disclosure, write, crash, or lifetime-corruption primitive.

7. Call-site confidence

Call-site confidence: medium

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

Patch: Bluetooth: L2CAP: validate connectionless PSM length [ Upstream
Commit: 3502ede8aed3c3ee1786d952a11054ac737d1a38
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=3502ede8aed3c3ee1786d952a11054ac737d1a38

Changed files:
  net/bluetooth/l2cap_core.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=3502ede8aed3c3ee1786d952a11054ac737d1a38

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

The original Linux kernel CVE announcement for CVE-2026-97408 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97408

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:L/PR:N/UI:N/S:U/C:N/I:N/A:N
  The Best / paranoid CVSS vector: AV:A/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L
  The Best / paranoid CVSS score: 4.3

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: 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: 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 [email protected] (and both
send reply to CVE record itself too and see "reply" button below for howto reply).

                 reply	other threads:[~2026-09-24 19:11 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --from, and --in-reply-to
  switches of git-send-email(1) and next cmd tested by kernelcve.org admin:

  git send-email --smtp-server=mail.kernelcve.org --smtp-server-port=25 --smtp-auth=none --from='Your Name <youremail@domain.is>' --suppress-cc=all --no-cc  \
    --in-reply-to=cve-2026-97408.d3470d5d9f69a6401037e153@kernelcve.org \
    [email protected] \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
The file with msg could look like this then:
cat YOUR_REPLY
Subject: Re: [CVE-2026-64206][MODERATE REGULAR] Bluetooth: L2CAP test

Just testing public-inbox replies.

Thanks,
MyName


This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox