[CVE-2026-90029][MODERATE REGULAR] usb: storage: realtek_cr: fix use-after-free on disconnect

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

CVE: CVE-2026-90029
Priority: MODERATE REGULAR
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: usb: storage: realtek_cr: fix use-after-free on disconnect
Commit: 7c4e2f964c65dea4ea22386799d5fb10ef1e3e54
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=7c4e2f964c65dea4ea22386799d5fb10ef1e3e54
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90029
Analysis date: Wed, 16 Sep 2026 09:12:44 -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-2026-90029 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90029

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-2026-90029	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-90029 CHECK WITH IMPACT FROM ORIG NN MODERATE	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:P/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H';*CWE-416;CWE-362;CWE-667;Other CVSS 'AV:P/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '6.4';DESCR 'A use after free can occur in the Realtek USB card reader driver because realtek_cr_destructor used timer_delete before freeing the chip object that contains the autosuspend timer. The timer callback may already be running and may rearm itself, so disconnect or teardown can leave a live timer path referencing freed kernel memory. For the CVSS the PR:N is used for the paranoid physical attack case because a malicious actor with physical access can trigger USB device disconnect without a local account on the victim. The issue is not network reachable and normally requires access to the affected USB device or equivalent privileged local control over USB driver teardown. Impact is at least denial of service via kernel crash. In the worst defensible case, the explicit kernel use after free may allow confidentiality or integrity impact and should be reviewed manually.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline / manual review recommended (with actual score 4)	YES	USB INIT UAF TIMER HARDWARE LINUS TEST 	NO	NO	checked

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

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

1. ActionableScore

* Raw conservative score: 3
* Conservative score: 3
* Raw paranoid score: 4
* Paranoid score: 4
* Final recommended bucket: **Borderline / manual review recommended**::

No floor or cap changed the conservative score.
No floor or cap changed the paranoid score.

2. Signal breakdown

Conservative signals:

* +2 Real UAF. The patch directly states that `realtek_cr_destructor()` can free the `chip` object while `rts51x_suspend_timer_fn()` may still be running or may rearm itself.
* +1 Reliable kernel crash / strong DoS. A kernel timer callback using a freed driver object can realistically cause a kernel fault or crash.
* +0 Remote reachability not awarded. This is a USB storage driver disconnect or teardown path, not a network packet path.
* +0 Local unprivileged code trigger not awarded. The supplied evidence supports physical disconnect or privileged local driver teardown, not an ordinary local syscall trigger.
* +0 C/I impact not awarded. No controlled reclaim, arbitrary write, information disclosure, callback hijack, or credential/security-object manipulation is demonstrated.

Paranoid additional signal:

* +1 Active async timer rearm concern. The timer callback may rearm itself after teardown, so the stale use is not merely a single shallow cleanup dereference. This supports a higher manual-review-sensitive score, but not Class A.

Non-awarded material signals:

* Controlled reclaim not demonstrated.
* Arbitrary or directed kernel write not demonstrated.
* Function pointer or callback pointer overwrite not demonstrated.
* Concrete repeated release/refcount mutation not established.
* Structured stale-object leverage not separately awarded because there is no evidence that attacker-shaped replacement fields are consumed.
* Weak LPE not awarded merely because this is a UAF.
* Extra Availability not awarded because the same kernel crash is already counted once.
* Confidentiality and integrity not awarded because the patch establishes a lifetime bug, not a disclosure path or unauthorized modification primitive.
* Physical or hardware rarity penalty not applied as a hard numeric reduction because Realtek USB card-reader hardware is hardware-specific but not proven rare. The reachability limitation is reflected by absence of remote or ordinary local-user bonuses.

Repeated release/refcount checklist:

* First release can end lifetime: not applicable to a refcounted object release sequence. The relevant lifetime end is `kfree()` of `chip`.
* Exact same reference survives: the timer stores the relevant `chip` context and may still run after `chip` teardown.
* Second operation is a real lifetime/refcount mutation: no.
* Concrete repeated release/refcount +1 awarded: no.
* Paranoid repeated-refcount leverage +1 awarded: no.
* FLOOR 5 applied: no.

3. Reachability analysis

The realistic trigger is USB device disconnect or driver teardown while the autosuspend timer callback is running or capable of rearming. A malicious physical actor may be able to trigger disconnect without host administrator privileges, so `HostAdminRequired=NO` is used. A local software-only trigger would usually require privileged control over USB driver binding, device removal, module unloading, or equivalent teardown paths, but that is not the only supported attacker model.

This is not namespace-scoped. No network namespace, user namespace, or namespaced capability path is involved. `NamespaceScopedPrivilege=NO`.

Prior setup requires the affected Realtek USB storage/card-reader driver and `CONFIG_REALTEK_AUTOPM` support. The attacker does not necessarily control this setup. It is a deployment prerequisite, not an exploitation primitive.

The race is between destructor/disconnect teardown and a kernel timer callback. Timing is not fully attacker-controlled, but repeated physical or management-triggered teardown attempts may make the condition testable. No separate hard-race penalty was applied because the commit itself establishes the lifetime bug and the scoring already avoids remote/local-control bonuses.

4. Primitive and severity interpretation

The exact primitive is an async timer use-after-free. The `chip` object contains `rts51x_suspend_timer`. The destructor used `timer_delete()` and then freed `chip`, but `timer_delete()` does not guarantee that an already running callback has completed and does not prevent the callback from rearming the timer. Therefore a live timer path may later reference freed kernel memory.

This is best classified as `PrimitiveClass=CLASS_B`. It is a real kernel lifetime violation, but it is constrained: the stale storage is a fixed USB driver context, the trigger is teardown/disconnect, and the demonstrated consequence is crash/DoS. There is no evidence of controlled replacement, attacker-controlled post-free contents, arbitrary read/write, function-pointer control, or a security-boundary bypass.

This is not an ordinary repeated-release/refcount case. The stale operation is a timer callback using freed storage, not a second `put`, `release`, `kref_put`, or refcount mutation on the same dangling reference.

Demonstrated consequence is denial of service via kernel crash or fault. Speculative consequences include broader memory corruption if the freed object is reused before the callback runs, but the supplied patch does not establish a useful exploitation primitive beyond the UAF itself.

`ImportantCandidate=NO` because this looks like a constrained physical/device teardown UAF with no concrete exploitation leverage beyond crash. `AutoDowngradeAllowed=YES` because the unresolved primitive is not strong enough to require Important preservation, although manual review remains recommended due to real UAF.

Operational severity is borderline Moderate. It should not be treated like a pure warning or harmless cleanup issue, but it also should not be promoted to a strong Important candidate without evidence of controlled reuse or privilege escalation.

5. One-sentence report phrase

A use-after-free in the Realtek USB card-reader autosuspend timer can occur during disconnect because the timer callback may still be running or may rearm itself after the driver object is freed, leading primarily to a physical or privileged-local kernel crash DoS.

6. Manual review recommendation

MANUAL CHECK RECOMMENDED

Reason: this is a real kernel UAF in an async timer path, but the available evidence supports a constrained disconnect/teardown crash scenario rather than controlled corruption, privilege escalation, or a network-reachable attack.

7. Call-site confidence

Call-site confidence: high

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

Patch: usb: storage: realtek_cr: fix use-after-free on disconnect
Commit: 7c4e2f964c65dea4ea22386799d5fb10ef1e3e54
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=7c4e2f964c65dea4ea22386799d5fb10ef1e3e54

Commit description:

realtek_cr_destructor() calls timer_delete() before the chip containing
the timer is freed. The timer callback may still be running and can
rearm itself, resulting in a use-after-free.

Use timer_shutdown_sync() to wait for the callback and prevent further
rearming. Do this unconditionally because ss_en may be changed after
the timer is armed.

Move timer_setup() into init_realtek_cr() so the timer is initialized
before any failure path can invoke the destructor.

Found by static analysis.

Fixes: e931830 ("Realtek cr: Add autosuspend function.")
Cc: stable <stable@kernel.org>
Co-developed-by: Ijae Kim <ae878000@gmail.com>
Signed-off-by: Ijae Kim <ae878000@gmail.com>
Signed-off-by: Myeonghun Pak <mhun512@gmail.com>
Link: https://patch.msgid.link/20260727123414.44700-1-mhun512@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed files:
  drivers/usb/storage/realtek_cr.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=7c4e2f964c65dea4ea22386799d5fb10ef1e3e54

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

The original Linux kernel CVE announcement for CVE-2026-90029 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90029

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:P/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H
  The Best / paranoid CVSS vector: AV:P/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H
  The Best / paranoid CVSS score: 6.4

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: 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).