[CVE-2026-72444][MODERATE 7.0] flow_dissector: check device type before reading ETH_ADDRS [ Upstream

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

CVE: CVE-2026-72444
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: flow_dissector: check device type before reading ETH_ADDRS [ Upstream
Commit: 0fe6455b8e1a22414f39c07bc90a1df12b52ec74
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=0fe6455b8e1a22414f39c07bc90a1df12b52ec74
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-72444
Analysis date: Sun, 16 Aug 2026 03:55:16 -0400
ActionableScore: 6
ActionableScore lower bound: 3
Actionable bucket: Actionable Moderate at minimum
Manual review required: YES

Summary:
A local attacker with control over TUN and TC flower configuration may cause the flow dissector to read Ethernet address fields from a non Ethernet skb, leading to uninitialized kernel memory use in classifier lookup and possible local DoS or limited packet classification integrity impact.

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

The original Linux kernel CVE announcement for CVE-2026-72444 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-72444

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-2026-72444	MODERATE	CHECK WITH IMPACT FROM ORIG NN LOW	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:H';CWE-125;CWE-908;*CWE-20;Other CVSS 'AV:L/AC:L/PR:H/UI:N/S:U/C:L/I:L/A:L';BEST CVSS score: '6.6';DESCR 'The flow dissector can read Ethernet address fields from packets that do not have an Ethernet MAC header when FLOW_DISSECTOR_KEY_ETH_ADDRS is requested. A local attacker with control over TUN and tc flower configuration can trigger this on an L3 TUN device with a shared multiq qdisc and a flower rule matching eth_src, then send traffic through AF_PACKET. The read can consume uninitialized skb memory or in a pathological unset mac_header case read from an invalid header location, causing unpredictable classifier keys and possible kernel crash. For the CVSS the PR:L is used in the paranoid score because CAP_NET_ADMIN like control may be delegated to containers or network namespaces even when the attacker is not full host root. The issue is not directly network reachable because it depends on local device and qdisc setup. Impact is at least local denial of service and may include limited confidentiality or integrity impact through uninitialized data use in classifier lookup.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate at minimum (with actual score 5)	YES	REMOTE KASAN OOB SKB HARDWARE LINUS QDISC SYZBOT PACKET  DECREASED_TO_MODERATEREG_BASED_ON_FALSEPOSCHECKOFKP	NO	NO	checked

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

ActionableScore=6
ActionableScoreLower=3

## 1. ActionableScore

* Conservative score: 3
* Paranoid score: 6
* Final recommended bucket: **Actionable Moderate at minimum**

## 2. Signal breakdown

Conservative signals:

* Weak or indirect corruption candidate: +1
  `eth_hdr(skb)` is interpreted as an Ethernet header even when the skb belongs to an L3 TUN device or has an unset MAC header. This is an invalid header interpretation and uninitialized or invalid kernel memory read, not a proven write primitive.

* Confidentiality impact plausible: +1
  The 12 bytes read from uninitialized skb memory are propagated into classifier key material. There is no direct copy to userspace, but classifier behavior may become externally observable through rule matching behavior.

* Integrity impact plausible: +1
  The uninitialized key can influence `rhashtable` lookup and flower classifier matching, causing packets to match unintended filters or actions.

* Common networking core path or packet-processing path: +1
  `flow_dissector` and TC flower classification are central packet-processing components, even though this trigger needs a specific TUN plus qdisc setup.

* Requires CAP_NET_ADMIN or equivalent in typical deployments: -2
  Creating or configuring the relevant TUN device, qdisc, and flower filter typically requires network administration privileges.

Paranoid additional interpretation:

* Local unprivileged trigger: +1
  In environments with unprivileged user namespaces, delegated network namespaces, or containerized CAP_NET_ADMIN, a non-host-root attacker may be able to create the needed local setup.

* Reliable kernel crash or strong DoS concern: +1
  The commit explicitly notes the unset `mac_header` sentinel case where `eth_hdr()` may return a wild pointer. This makes a crash path plausible, although the demonstrated report is KMSAN uninitialized read.

Paranoid score rationale: 6 is defensible because the bug is an invalid kernel memory read in a common networking classifier path, with possible classifier integrity effects and a plausible crash path, but it is not a strong memory corruption or LPE primitive.

## 3. Reachability analysis

The realistic trigger is local. The described reproducer needs an L3 TUN device, a multiq qdisc, a flower filter matching `eth_src`, and traffic injected through `AF_PACKET`. In ordinary host deployments this usually requires CAP_NET_ADMIN and likely CAP_NET_RAW, so the conservative score applies a privilege penalty.

Namespaces and containers matter. If unprivileged user namespaces are enabled and a user can obtain CAP_NET_ADMIN inside a network namespace with access to `/dev/net/tun`, the practical privilege requirement may be closer to PR:L rather than PR:H. This is why the paranoid score is higher.

The path is not directly remote-network reachable. Packet traffic alone is not enough unless the attacker can also arrange the affected TC/TUN classification setup. Call-site confidence: high, because the commit describes the concrete TC flower and TUN path and the changed code is in the actual flow dissector read site.

## 4. Severity interpretation

This is not an Important-class vulnerability by default because the patch does not show OOB write, UAF reclaim, object replacement, arbitrary write, or privilege escalation. It behaves more like an actionable Moderate issue: invalid kernel memory read plus uninitialized data use in a packet classifier, with possible crash and limited classifier integrity impact.

The theoretical risk is higher than a pure warning or cosmetic KMSAN fix because the read result affects lookup keys and because an unset MAC header can produce a wild pointer. The realistic exploitation evidence still points mainly to local DoS or limited policy/classification misbehavior, not reliable LPE.

## 5. One-sentence report phrase

A local attacker with control over TUN and TC flower configuration may cause the flow dissector to read Ethernet address fields from a non Ethernet skb, leading to uninitialized kernel memory use in classifier lookup and possible local DoS or limited packet classification integrity impact.

## 6. Manual review recommendation

MANUAL CHECK REQUIRED

YES REQUIRES MANUAL CHECK. The issue is not a proven LPE, but it is an invalid kernel memory read in a common networking classification path with namespace-dependent reachability, possible wild-pointer crash behavior, and limited confidentiality or integrity implications through classifier key use.

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

Patch: flow_dissector: check device type before reading ETH_ADDRS [ Upstream
Commit: 0fe6455b8e1a22414f39c07bc90a1df12b52ec74
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=0fe6455b8e1a22414f39c07bc90a1df12b52ec74

Changed files:
  net/core/flow_dissector.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=0fe6455b8e1a22414f39c07bc90a1df12b52ec74

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

The original Linux kernel CVE announcement for CVE-2026-72444 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-72444

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:L/PR:H/UI:N/S:U/C:L/I:L/A:L
  The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:H
  The Best / paranoid CVSS score: 6.6

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: 3
  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: NO
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).