[CVE-2026-98125][LOW] smb/client: fix stale page cache in insert/collapse range [ Upstream

[CVE-2026-98125][LOW] smb/client: fix stale page cache in insert/collapse range [ Upstream

[AL-KERNEL]

CVE: CVE-2026-98125
Priority: LOW
AL-KERNEL base severity: LOW
KPANIC flag: NO
Patch: smb/client: fix stale page cache in insert/collapse range [ Upstream
Commit: 76222e76d08fbae4092ee9da84ded2b4c24ee598
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=76222e76d08fbae4092ee9da84ded2b4c24ee598
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98125
Analysis date: Fri, 25 Sep 2026 12:58:39 -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-2026-98125 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98125

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: LOW
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-2026-98125	LOW	https://www.kernelcve.org/list/?q=CVE-2026-98125 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:N';*CWE-664;CWE-682;Other CVSS 'AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:L/A:N';BEST CVSS score: '4.4';DESCR 'A stale page-cache issue in the SMB client can occur after SMB3 insert range or collapse range operations because partially covered boundary pages remain uptodate after the server moves file data. A local process with access to an SMB mounted file can trigger the condition using a non page aligned range operation and may subsequently read stale data from the client cache. For the CVSS the PR:L is used because the attacker needs local access to the mounted share and permission to operate on the affected file. The issue is not directly network reachable because the SMB server cannot independently invoke the vulnerable client operation. Impact is primarily low integrity impact from incorrect file contents returned to applications. Limited confidentiality impact is included only in the paranoid score because previously cached data could theoretically be returned after server side data movement.';NO MANUAL CHECK; ,and ActionableScore result is Ordinary Moderate (with actual score 2)	YES	DANGER DISK SERVERTOCLIENT LINUS MEMORY  DECREASED_TO_MODERATE_REGULAR_FROM_MODERATE7_BASED_ON_GUESSCVSS DECREASED_TO_LOW_FROM_MODERATE_BASED_ON_ACTIONABLESCORELESSTHAN3	YES	NO	checked

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

ActionableScore=2
PrimitiveClass=NONE
RealUAF=NO
HostAdminRequired=NO
NamespaceScopedPrivilege=UNKNOWN
ImportantCandidate=NO
ManualReview=NO
AutoDowngradeAllowed=YES

1. ActionableScore

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

No floor or cap was applied.

2. Signal breakdown

* +1 Local unprivileged trigger. A local user with write access to a file on an already mounted SMB share can perform a non-page-aligned insert or collapse range operation and read the resulting stale client cache data.
* +1 Important filesystem/storage path. The defect affects SMB client page-cache coherency during file range modification operations.

Important non-awarded signals:

* No UAF, OOB access, invalid free, refcount misuse, or kernel memory corruption is established.
* No controlled reclaim, arbitrary write, callback influence, or attacker-controlled kernel-object replacement is demonstrated.
* No remote reachability is established. An SMB server cannot independently invoke smb3_insert_range or smb3_collapse_range on the client.
* No confidentiality point is awarded. The supplied reproducer primes the same clients cache before the range operation, and no cross-user or permission-bypass disclosure is established.
* No integrity point is awarded. The patch demonstrates stale data returned to the local reader, not unauthorized modification of server data, another principals data, filesystem metadata, or protected shared pages.
* No Dirty-Pipe-like ownership or COW violation is established. There is no evidence that an authorized write to file A modifies independently permissioned file B.
* No reliable kernel crash, persistent filesystem failure, or host-wide DoS is established.
* No negative privilege penalty is applied. The input establishes a local file operation available to a user with ordinary write access to the affected SMB file, not a host-administration-only operation.
* No niche-subsystem penalty is applied. SMB mount availability and server COPYCHUNK support are deployment prerequisites, but no evidence shows they are attacker-inaccessible once the vulnerable SMB operation is available.

3. Reachability analysis

The triggering actor is a local process with access to an existing CIFS or SMB mount and sufficient file permissions to perform the relevant range operation. The supplied reproduction uses fallocate with insert range against a writable file, which is below host-admin privilege when the user owns or may modify that file.

The operation requires an SMB server and client configuration supporting the relevant SMB3 range operation and COPYCHUNK behavior. Mount creation may have required prior administrator setup, but the attacker does not need to create the mount or hold host-admin privileges if a writable affected share is already available.

No network namespace or namespace-scoped capability behavior is shown. NamespaceScopedPrivilege is therefore UNKNOWN. The SMB server is part of the storage deployment, but it cannot directly trigger the vulnerable client-side range operation through network traffic alone.

There is no race requirement. The condition is deterministic when the affected range boundary is not page-aligned and the cache is primed before the server-side data move.

4. Primitive and severity interpretation

The exact primitive is stale page-cache state. Boundary folios remain marked uptodate after partial zeroing, while SMB3 COPYCHUNK or related server-side range movement changes the authoritative file layout. Later local reads can return obsolete bytes from the client cache.

This is a cache-coherency and filesystem semantic-consistency bug. It is not an active-object lifetime violation, UAF, stale freed-object access, refcount mutation, COW ownership violation, or authorization bypass. The stale state is page-cache content associated with the same inode and client mount, not reclaimed kernel storage or an attacker-shaped replacement object.

The demonstrated consequence is incorrect file content returned to a local application after a successful insert or collapse range operation. A speculative consequence is application-level malfunction if an application acts on stale bytes. The supplied evidence does not establish data disclosure across access-control boundaries, unauthorized persistent modification, privilege escalation, or kernel compromise.

PrimitiveClass is NONE because no meaningful corruption, lifetime, ownership, authorization, or similar security primitive is established beyond stale local cache semantics. ImportantCandidate is NO because the patch-supported behavior is limited to stale reads in a local file operation. AutoDowngradeAllowed is YES because no unresolved kernel-corruption or security-boundary primitive would be discarded by ordinary Moderate handling.

Likely operational severity is ordinary Moderate or low-priority correctness impact. Systems relying on SMB3 insert or collapse range operations should apply the fix to preserve correct file readback behavior.

5. One-sentence report phrase

An SMB client page-cache coherency flaw can return stale file data after a local non-page-aligned SMB3 insert or collapse range operation, but no kernel memory corruption, privilege escalation, or direct remote trigger is established.

6. Manual review recommendation

NO MANUAL CHECK NEEDED

The patch establishes a deterministic stale-cache correctness issue without a memory-corruption, ownership-boundary, authorization-bypass, or cross-principal disclosure primitive.

7. Call-site confidence

Call-site confidence: medium

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

Patch: smb/client: fix stale page cache in insert/collapse range [ Upstream
Commit: 76222e76d08fbae4092ee9da84ded2b4c24ee598
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=76222e76d08fbae4092ee9da84ded2b4c24ee598

Changed files:
  fs/smb/client/smb2ops.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=76222e76d08fbae4092ee9da84ded2b4c24ee598

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

The original Linux kernel CVE announcement for CVE-2026-98125 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98125

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:L/UI:N/S:U/C:N/I:L/A:N
  The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:N
  The Best / paranoid CVSS score: 4.4

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: LOW
KPANIC detected for this report: NO
Published priority for this report (same as in Subject): LOW

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).