[CVE-2026-90146][MODERATE 7.0] bpf, xdp: move offload check into dev_xdp_install() [ Upstream

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

CVE: CVE-2026-90146
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: bpf, xdp: move offload check into dev_xdp_install() [ Upstream
Commit: ea7b35dcc9430293b861bc7bad0c546f193c85f9
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=ea7b35dcc9430293b861bc7bad0c546f193c85f9
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90146
Analysis date: Thu, 17 Sep 2026 19:09:51 -0400
ActionableScore: 5
ActionableScore lower bound: 4
Actionable bucket: Actionable Moderate
Manual review required: NO

Summary:
No one-sentence report phrase was found.

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

The original Linux kernel CVE announcement for CVE-2026-90146 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90146

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: NO

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

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

CVE-2026-90146	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-90146 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:N/I:N/A:H';CWE-693;*CWE-20;CWE-284;Other CVSS 'AV:L/AC:L/PR:H/UI:N/S:U/C:N/I:N/A:L';BEST CVSS score: '5.5';DESCR 'An XDP link update validation bypass exists because bpf_xdp_link_update calls dev_xdp_install directly and did not run the compatibility checks kept in dev_xdp_attach. A user could create an XDP link with a normal program and then replace it with an offloaded or device-bound program through BPF_LINK_UPDATE, causing a program that should be restricted to hardware offload or a specific device to be placed on the software XDP path. This is not directly network reachable by packet traffic alone, but packets may exercise the bad attachment after a local control plane update. For the CVSS the PR:L is used for the paranoid score because exploitation can be performed by a local user or service that already has delegated BPF/XDP update capability, such as CAP_BPF and CAP_NET_ADMIN or an equivalent capability in the target environment. Impact is mainly denial of service or validation policy bypass. The patch context does not show a concrete arbitrary write or information leak primitive.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate (with actual score 5)	YES	REMOTE BPF SKB HARDWARE LINUS  KPANIC INCREASED_FROM_LOW_BASED_ON_REQUIREMANUALCHECK	NO	NO	checked

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

ActionableScore=5
ActionableScoreLower=4
PrimitiveClass=CLASS_C
RealUAF=NO
HostAdminRequired=UNKNOWN
NamespaceScopedPrivilege=UNKNOWN
ImportantCandidate=NO
ManualReview=RECOMMENDED
AutoDowngradeAllowed=NO

1. ActionableScore

* Raw conservative score: 4
* Conservative score: 4. No floor or cap applied.
* Raw paranoid score: 5
* Paranoid score: 5. No floor or cap applied.
* Final recommended bucket: **Actionable Moderate**::

2. Signal breakdown

* +2: Concrete validation and policy bypass. `BPF_LINK_UPDATE` reached `dev_xdp_install()` directly and skipped checks that would otherwise reject offloaded, device-bound, wrong-device, generic-mode-incompatible, `DEVMAP`, and `CPUMAP` XDP programs.
* +1: Externally meaningful BPF/XDP attach-state impact. The bypass can place a program into a software XDP path where the normal attach model says it must not run.
* +1: BPF/XDP packet-processing relevance. The bad state is created through control-plane update, but packets can later exercise the incorrectly installed program on the data path.
* Paranoid +1: The wrong-mode attachment creates an unresolved BPF/XDP runtime semantic mismatch between offload/device-bound assumptions and software execution. This is not scored as memory corruption, but it is enough to avoid automatic closure.

Non-awarded signals:

* Real UAF not awarded. No stale pointer, freed object, double release, or lifetime violation is shown.
* Controlled reclaim not demonstrated.
* Arbitrary write not demonstrated.
* Confidentiality and integrity not awarded because the patch does not establish a disclosure path, protected-object write, or concrete kernel memory corruption primitive.
* Reliable kernel crash not conservatively awarded because the commit establishes invalid attachment state, not a specific observed panic.
* Remote reachability not awarded. Packet traffic alone does not create the bad attachment.
* Local unprivileged trigger not awarded because BPF/XDP load, link creation, link update, and netdevice control privileges are not resolved as ordinary-user reachable.
* Host-admin penalty not applied because host-admin-equivalent privilege is not positively established from the supplied evidence.
* Namespace-scoped privilege not awarded because complete BPF load, XDP attach, link update, and target netdevice control are not proven to be namespace-scoped.

3. Reachability analysis

The trigger is local control-plane manipulation of BPF/XDP state. An attacker must create an XDP link with a normal program and then use `BPF_LINK_UPDATE` to replace it with an offloaded or device-bound program that normal attach checks would reject.

This is not directly network-triggerable. Network packets may exercise the invalid state after the local update, but packets alone do not perform the vulnerable operation.

For BPF/XDP, the complete privilege chain is unresolved. Program loading and XDP attachment commonly require `CAP_BPF`, `CAP_NET_ADMIN`, `CAP_SYS_ADMIN`, ownership of relevant BPF objects, or an equivalent delegated service context. The supplied patch does not prove that an ordinary local user can perform the full sequence.

Namespace scope is also unresolved. Some netdevice operations may be namespace-aware, but the complete chain including BPF program creation, link creation, link update, and target device control is not proven to work entirely inside attacker-created namespaces.

No special hardware is strictly required for the generic bypass pattern, although offloaded or device-bound program cases may depend on device/offload support. There is no race or timing dependency.

4. Primitive and severity interpretation

The exact primitive is a BPF/XDP validation bypass. A program that should be rejected for the selected XDP mode or device can be installed through the `BPF_LINK_UPDATE` path because checks were placed in `dev_xdp_attach()` while `bpf_xdp_link_update()` called `dev_xdp_install()` directly.

This is not an active-object UAF and not fixed stale storage. There is no cleanup/destructor stale use and no concrete repeated release/refcount operation.

Primitive class is `CLASS_C`: invalid kernel control-plane state with possible DoS or policy-bypass consequences, but without an independently established memory-corruption primitive.

Demonstrated consequence is incorrect acceptance of unsupported XDP program combinations. Speculative consequence is a stronger runtime failure if the wrongly installed program violates assumptions of the software XDP path. The evidence does not establish arbitrary read/write, privilege escalation, or information disclosure.

ImportantCandidate is `NO` because the patch does not establish Class A corruption, LPE, protected resource access, or a strong Class B lifetime primitive. AutoDowngradeAllowed is `NO` because this is a meaningful BPF/XDP validation bypass affecting packet-processing attachment semantics, so automatic closure as an ordinary low-impact issue could miss environment-specific impact.

Likely operational severity is Actionable Moderate.

5. One-sentence report phrase

A BPF/XDP link update validation bypass allowed `BPF_LINK_UPDATE` to install offloaded or device-bound XDP programs into an unsupported software path, requiring local BPF/XDP update capability and potentially causing denial of service or policy bypass when the invalid attachment is exercised.

6. Manual review recommendation

MANUAL CHECK RECOMMENDED

Reason: no concrete UAF or arbitrary write is shown, but this is a BPF/XDP validation bypass on a packet-processing attachment path with unresolved environment-dependent impact and unresolved privilege scope.

7. Call-site confidence

Call-site confidence: high

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

Patch: bpf, xdp: move offload check into dev_xdp_install() [ Upstream
Commit: ea7b35dcc9430293b861bc7bad0c546f193c85f9
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=ea7b35dcc9430293b861bc7bad0c546f193c85f9

Changed files:
  net/core/dev.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=ea7b35dcc9430293b861bc7bad0c546f193c85f9

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

The original Linux kernel CVE announcement for CVE-2026-90146 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90146

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:N/I:N/A:L
  The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
  The Best / paranoid CVSS score: 5.5

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: 4
  Paranoid ActionableScore: 5

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).