[CVE-2026-97558][MODERATE REGULAR] smb: client: fix cifsFileInfo reference leak in deferred close

[CVE-2026-97558][MODERATE REGULAR] smb: client: fix cifsFileInfo reference leak in deferred close

[AL-KERNEL]

CVE: CVE-2026-97558
Priority: MODERATE REGULAR
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: smb: client: fix cifsFileInfo reference leak in deferred close
Commit: ea93759d6c2789924f4e557aed5be532207715ea
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=ea93759d6c2789924f4e557aed5be532207715ea
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97558
Analysis date: Fri, 25 Sep 2026 13:43:32 -0400
ActionableScore: 3
ActionableScore lower bound: 2
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-2026-97558 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97558

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-2026-97558	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-97558 CHECK WITH IMPACT FROM ORIG NN LOW	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H';*CWE-401;CWE-362;*CWE-772;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:L';BEST CVSS score: '4.7';DESCR 'A race in the SMB client deferred close path can leak a cifsFileInfo reference when queue_delayed_work fails because the same deferred work item is already pending. The leaked reference can retain related tlink, dentry, and server handle resources. Repeated successful triggering may cause resource exhaustion on a client that mounts and uses an SMB share. For the CVSS the PR:L metric is used because an attacker needs local access to the mounted SMB filesystem and permission to perform file open and close operations. The race requires a narrow timing window, therefore AC:H is used. The issue is not directly network reachable because it is triggered by local client file operations. Impact is denial of service through accumulated resource leakage. No confidentiality or integrity impact is supported because the patch indicates a reference leak and does not demonstrate a memory corruption primitive.';NO MANUAL CHECK; ,and ActionableScore result is Ordinary Moderate (with actual score 3)	YES	LOCK DISK LEAK SERVERTOCLIENT LINUS  INCREASED_FROM_LOW_TO_MODERATE70_BASED_ON_ACTIONABLESCOREHIGHEREQTHAN3 DECREASED_TO_MODERATEREG_BASED_ON_ACTIONABLESCORELESSTHAN4	YES	NO	checked

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

ActionableScore=3
ActionableScoreLower=2
PrimitiveClass=NONE
RealUAF=NO
HostAdminRequired=NO
NamespaceScopedPrivilege=NO
ImportantCandidate=NO
ManualReview=NO
AutoDowngradeAllowed=YES

1. ActionableScore

* Raw conservative score: 2
* Conservative score: 2
* Raw paranoid score: 3
* Paranoid score: 3
* Final recommended bucket: **Ordinary Moderate**::

No floor or cap was applied.

2. Signal breakdown

Conservative signals:

* +1 Local unprivileged trigger. A user with ordinary access to an already mounted SMB filesystem can perform the required local file open and close operations. Mount creation itself is not required for the trigger.
* +1 Concrete persistent kernel resource/reference leak. The patch directly establishes that a cifsFileInfo reference from the closing struct file is not dropped when queue_delayed_work fails because deferred work is already pending. The retained reference pins associated tlink, dentry, and server handle resources.

Paranoid-only signal:

* +1 Sustained resource exhaustion plausibility. The same sequence can recur during repeated file open and deferred close activity. Each successful occurrence can retain another reference, so a long-running workload can accumulate leaked SMB client resources and potentially cause availability degradation.

Non-awarded material signals:

* +0 Real UAF. Not established. The defect leaks an extra reference instead of dropping a reference too early. The object remains pinned rather than becoming dangling.
* +0 Concrete repeated release/refcount mutation. Not established. There is no first put that may free cifsFileInfo followed by a second put through the same dangling reference. The missing put is the defect.
* +0 Controlled reclaim, arbitrary write, callback control, or post-free structured-object reuse. Not established.
* +0 Reliable kernel crash. Not established by the patch. The demonstrated result is resource retention.
* +0 Additional confidentiality or integrity impact. The patch provides no disclosure, unauthorized modification, or security-boundary bypass primitive.
* +0 Direct remote reachability. Not established. Network SMB connectivity may be needed for the mount, but the vulnerable operation is local client-side file close handling.
* -0 Hard race penalty. The bug involves a workqueue ordering window, but the attacker can repeatedly perform local open and close operations. The supplied evidence does not establish a one-shot or externally uncontrollable race.
* -0 host-admin penalty. Host-admin-equivalent privileges are not needed to trigger the bug once an accessible SMB mount already exists.
* -0 niche subsystem penalty. SMB client mounts are optional, but an attacker who already has access to the mounted filesystem can exercise the affected path normally.

3. Reachability analysis

The evaluated attacker is a local user with permission to access an existing mounted SMB CIFS filesystem. The attacker can repeatedly open and close files whose close path uses deferred close handling.

The complete trigger does not require the attacker to mount the SMB share, modify mount options, manage the SMB client module, or obtain host administrative capabilities. A host administrator or another trusted component must have configured the SMB mount beforehand, but that is prior environmental setup rather than a required attacker privilege for the vulnerable close sequence.

No capability check or namespace-scoped privileged operation is shown in the vulnerable cifs_close path. NamespaceScopedPrivilege is therefore NO rather than YES or UNKNOWN. The issue is not directly network-triggerable by a remote SMB server alone because the relevant reference leak requires local VFS file lifecycle operations on the client.

The workqueue state sequence is repeatable under file activity. However, it requires a specific ordering involving a running deferred callback, requeueing, flag clearing, handle reuse, and a later failed queue_delayed_work call. This supports AC:H-style CVSS reasoning, but it does not establish memory corruption or a strong exploitation primitive.

4. Primitive and severity interpretation

The exact primitive is a missing reference release. cifs_close transfers ownership of the closing file reference to deferred work only when queue_delayed_work successfully queues that work. Before the patch, cifs_close returned success even if queue_delayed_work failed because the work item was already pending. The closing file reference was then not released and remained permanently pinned.

This is neither an active structured-object UAF nor fixed stale storage. The cifsFileInfo object is retained by excess references instead of being released too early. The queued deferred work owns one valid reference, while the failed queue attempt incorrectly leaves an additional closing-file reference alive.

This is not an ordinary stale cleanup dereference and not a second lifetime mutation through a potentially dangling object. The first lifetime operation does not free the object. The missing _cifsFileInfo_put prevents the expected reference decrement. No double put, double free, stale dereference, post-free refcount mutation, or controlled object replacement is established.

PrimitiveClass is NONE because the patch establishes a resource ownership leak without memory corruption, authorization bypass, semantic reinterpretation, or exploitable lifetime violation. The demonstrated consequence is accumulating SMB client resource usage. The speculative worst case is local denial of service through sustained memory or resource exhaustion. No practical leverage beyond accumulation of leaked references is shown.

ImportantCandidate is NO because this is a concrete but bounded resource leak without a UAF, corruption primitive, cross-boundary effect, or isolation bypass. AutoDowngradeAllowed is YES because automatic closure as a non-urgent resource leak would not discard an unresolved kernel corruption or security-boundary primitive.

Likely operational severity is Low to ordinary Moderate. Prioritize the fix normally on systems that run SMB client mounts with deferred close enabled and permit untrusted users to perform sustained filesystem workloads.

5. One-sentence report phrase

A race in the SMB client deferred close path can leak cifsFileInfo references when deferred work is already pending, allowing repeated local file activity on an accessible SMB mount to accumulate kernel resources and potentially cause denial of service.

6. Manual review recommendation

NO MANUAL CHECK NEEDED

The patch establishes a reference leak rather than a dangling-reference use, repeated release, memory corruption, authorization bypass, or remotely reachable attack path. Routine remediation is appropriate, but urgent manual exploitation analysis is not required.

7. Call-site confidence

Call-site confidence: high

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

Patch: smb: client: fix cifsFileInfo reference leak in deferred close
Commit: ea93759d6c2789924f4e557aed5be532207715ea
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=ea93759d6c2789924f4e557aed5be532207715ea

Commit description:

When cifs_close() defers a close, it hands the cifsFileInfo reference
of the closing struct file to the queued work. Each execution of
smb2_deferred_work_close() drops one such reference.

deferred_close_scheduled can be false while the work is pending: the
workqueue clears PENDING when the callback starts to run, before the
callback clears the flag under deferred_lock. A close in that
interval requeues the running work, and the callback then clears the
flag, leaving the requeued work pending with the flag down. A later
cifs_open() can reuse the handle and its cifs_close() reaches the
same branch: queue_delayed_work() fails because the work is still
pending, but cifs_close() returns without dropping the closing file's
reference. The cifsFileInfo count stays pinned and its tlink, dentry
and server handle are leaked.

Check the return value and hand off the reference only when work was
actually queued. Otherwise, use the shared _cifsFileInfo_put(), like
the mod_delayed_work() branch above: the pending execution already
owns its reference.

This issue was found by an in-house static analysis tool.

Fixes: c3f207a ("cifs: Deferred close for files")
Cc: stable@vger.kernel.org
Assisted-by: Codex:gpt-5.6
Co-developed-by: Song Li <songl@zju.edu.cn>
Signed-off-by: Song Li <songl@zju.edu.cn>
Signed-off-by: Fan Wu <fanwu01@zju.edu.cn>
Signed-off-by: Paulo Alcantara <pc@manguebit.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed files:
  fs/smb/client/file.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=ea93759d6c2789924f4e557aed5be532207715ea

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

The original Linux kernel CVE announcement for CVE-2026-97558 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97558

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:L
  The Best / paranoid CVSS vector: AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H
  The Best / paranoid CVSS score: 4.7

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: 2
  Paranoid ActionableScore: 3

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).