[CVE-2026-90051][MODERATE REGULAR] tcp: reject non zerocopy devmem tx [ Upstream

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

CVE: CVE-2026-90051
Priority: MODERATE REGULAR
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: tcp: reject non zerocopy devmem tx [ Upstream
Commit: 2151b2bcf6fcec52665f606eed20da068447f0b7
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=2151b2bcf6fcec52665f606eed20da068447f0b7
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90051
Analysis date: Thu, 17 Sep 2026 16:03:28 -0400
ActionableScore: 4
ActionableScore lower bound: 3
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-2026-90051 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90051

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-2026-90051	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-90051 CHECK WITH IMPACT FROM ORIG NN MODERATE	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:H';*CWE-20;CWE-754;CWE-670;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '5.8';DESCR 'TCP devmem TX requests could pass validation even when the send path was not actually using zero copy or no devmem binding was present. On devices without NETIF_F_SG this can select the copy path for an iovec that contains DMA buffer offsets and can also allow net_iov fragments and ordinary page fragments to be mixed in one skb. This breaks networking invariants and may cause a kernel crash or other incorrect skb processing. For the CVSS the PR:L reflects that a local user process must submit the malformed io_uring or TCP send request through an accessible socket and devmem TX interface. The issue is not directly network reachable by sending packets to the host. The base impact is local denial of service. In the paranoid assessment limited confidentiality or integrity impact is retained because later skb processing may interpret mixed fragment types incorrectly, although the patch does not establish a specific UAF, OOB write, or arbitrary write primitive.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline / manual review recommended (with actual score 4)  SKIP CVE-2026-90051 UNKNOWN SKIP The Fixes patch not applied yet, so unlikely that actual: url YES  NO NO unknown 	MAYBE	REMOTE SIMPLEFIX NETWORK IO_URING MEMORY  DECREASED_TO_MODERATEREG_BASED_ON_FALSEPOSCHECKOFKP	-	-	checked

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

ActionableScore=4
ActionableScoreLower=3
PrimitiveClass=CLASS_C
RealUAF=NO
HostAdminRequired=NO
NamespaceScopedPrivilege=NO
ImportantCandidate=NO
ManualReview=RECOMMENDED
AutoDowngradeAllowed=YES

1. ActionableScore

* Raw conservative score: 3.
* Conservative score: 3. No floor or cap changed the result.
* Raw paranoid score: 4.
* Paranoid score: 4. The validation-only paranoid cap of 5 was not reached.
* Final recommended bucket: **Borderline / manual review recommended**::

2. Signal breakdown

Conservative signals:

* +1 Local unprivileged trigger. An ordinary process can construct the malformed io_uring or TCP send request with `dmabuf_id` and zero-copy flags but without a valid binding.
* +1 Weak or indirect corruption candidate. The accepted state can mix `net_iov` and page fragments in one skb, violating the expected fragment representation.
* +1 Common networking core path. The invalid state is admitted through `tcp_sendmsg_locked()` and can escape into normal skb processing.
* Total: 3.

Paranoid signals:

* The same +3 signals.
* +1 Reliable kernel crash or strong DoS concern. Later skb processing relies on homogeneous fragment semantics, so the documented invariant violation plausibly causes an invalid interpretation or kernel fault. A specific crashing consumer is not shown, so this point is paranoid only.
* Total: 4.

Not awarded:

* No UAF, stale lifetime, OOB write, arbitrary write, controlled reclaim, callback influence, or concrete disclosure is established.
* Strong semantic reinterpretation was not awarded because the visible evidence does not identify a downstream consumer that turns the mixed fragments into an independently useful memory-access primitive.
* No C or I impact was awarded from the previous CVSS hint alone.
* No remote-reachability point was awarded. Network packets alone do not create the malformed local send request.
* No extra Availability point was awarded beyond the paranoid crash concern.
* No privilege penalty was applied because the malformed request itself does not require host-admin-equivalent authority.
* No niche-subsystem or hardware penalty was applied. `NETIF_F_SG` absence is a device-path condition, not evidence of rare physical hardware.
* No race penalty applies.

3. Reachability analysis

A local process can submit the malformed send operation through io_uring or the applicable TCP send interface. The old check accepted `dmabuf_id` when zero-copy flags and `SOCK_ZEROCOPY` were present, even if the actual selected path was not `MSG_ZEROCOPY` or no devmem binding existed.

The trigger therefore does not inherently require host root or a capability. It does require a kernel with TCP devmem TX support, an accessible socket, and a route through a device path where zero-copy is unavailable, notably when `NETIF_F_SG` is absent. These are environmental prerequisites rather than attacker privileges.

No namespace-scoped capability is required for the malformed send itself. No prior valid devmem binding is required because the missing binding is part of the invalid state. The issue is local rather than directly network-triggerable, and no timing race is involved.

4. Primitive and severity interpretation

The exact primitive is acceptance of an inconsistent TCP devmem TX state. DMA-buffer offsets may reach a normal copy path, and `net_iov` fragments may be mixed with ordinary page fragments in one skb.

This is active invalid-state behavior, but the supplied code does not establish an actual UAF, OOB access, write primitive, or controlled type confusion. The copy-path misuse would normally fail, while the stronger concern is that mixed skb fragment types can later be interpreted under incompatible assumptions.

The demonstrated class is therefore CLASS_C rather than CLASS_A or CLASS_B. The practical consequence is failed transmission and a plausible kernel crash or networking disruption. Stronger memory corruption remains speculative because the relevant downstream skb consumer is not supplied.

ImportantCandidate is NO because no Important-preserving primitive is established. AutoDowngradeAllowed is YES because the available evidence supports invalid-state or DoS behavior without unresolved concrete lifetime corruption, controlled semantic reinterpretation, or a security-boundary bypass. Manual review remains recommended because violating skb fragment-type invariants can have consequences not visible in this one-line validation fix.

5. One-sentence report phrase

TCP devmem TX validation could admit a non-zero-copy request without a valid binding and mix net_iov with page fragments in one skb, creating a local invalid-state condition with plausible kernel DoS impact but no demonstrated UAF, arbitrary write, or privilege-escalation primitive.

6. Manual review recommendation

MANUAL CHECK RECOMMENDED

Review downstream skb fragment consumers to determine whether mixed `net_iov` and page fragments cause only a fault or enable a concrete type-confusion, disclosure, or write primitive.

7. Call-site confidence

Call-site confidence: medium

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

Patch: tcp: reject non zerocopy devmem tx [ Upstream
Commit: 2151b2bcf6fcec52665f606eed20da068447f0b7
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=2151b2bcf6fcec52665f606eed20da068447f0b7

Changed files:
  net/ipv4/tcp.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=2151b2bcf6fcec52665f606eed20da068447f0b7

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

The original Linux kernel CVE announcement for CVE-2026-90051 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90051

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:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:H
  The Best / paranoid CVSS score: 5.8

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: 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).