[CVE-2026-90036][IMPORTANT] NFSD: Prevent client use-after-free during blocked-lock reaping [ Upstream

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

CVE: CVE-2026-90036
Priority: IMPORTANT
AL-KERNEL base severity: IMPORTANT
KPANIC flag: YES
Patch: NFSD: Prevent client use-after-free during blocked-lock reaping [ Upstream
Commit: cd489b03587378645fe0d20142a33f1ed60bac98
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=cd489b03587378645fe0d20142a33f1ed60bac98
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90036
Analysis date: Wed, 16 Sep 2026 09:17:20 -0400
ActionableScore: 6
ActionableScore lower bound: 5
Actionable bucket: Actionable Moderate / Strong Important candidate requiring manual review
Manual review required: YES

Summary:
No one-sentence report phrase was found.

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

The original Linux kernel CVE announcement for CVE-2026-90036 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90036

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: IMPORTANT
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-2026-90036	IMPORTANT	https://www.kernelcve.org/list/?q=CVE-2026-90036 CHECK WITH IMPACT FROM ORIG NN LOW	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H';*CWE-416;CWE-362;CWE-667;Other CVSS 'AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '7.5';DESCR 'A use-after-free can occur in the NFSD NFSv4 blocked lock path because a bare lock owner can keep a raw pointer to an nfs4_client without holding a reference that pins the client lifetime. During blocked lock reaping, nfs4_laundromat() can detach the blocked lock before the related client expiration path observes it, allowing force_expire_client() to free the client while later state owner cleanup still dereferences client cl_lock. For the CVSS the PR:L is used because practical triggering normally requires the attacker to act as an NFSv4 client with access to an export and the ability to create lock state, but not root on the NFS server. The issue is network reachable when the NFS server is exposed to the attacker, although exploitation requires a timing race with client expiration or lock reaping and is therefore AC:H. Impact is at least denial of service via kernel crash. For the paranoid score, the UAF bug class supports C:H/I:H/A:H because freed kernel memory is dereferenced in a concurrent lifetime race and should be reviewed manually for possible privilege escalation or stronger memory corruption effects.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 6)	YES	LOCK DANGER DISK UAF NETWORK LINUS KPANIC  KPANIC INCREASED_TO_HIGH_BASED_ON_GUESSCVSS	NO	NO	checked

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

ActionableScore=6
ActionableScoreLower=5
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=NO
NamespaceScopedPrivilege=NO
ImportantCandidate=YES
ManualReview=REQUIRED
AutoDowngradeAllowed=NO

1. ActionableScore

* Raw conservative score: 4
* Conservative score: 5
* Applied conservative floor/cap: Real UAF FLOOR 5 applied because this is a below-host-admin reachable active structured NFSD client lifetime bug with stale use of freed nfs4_client state.
* Raw paranoid score: 6
* Paranoid score: 6
* Applied paranoid floor/cap: none.
* Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**::

2. Signal breakdown

Conservative signals:

* +2 Real UAF / active lifetime violation: the commit directly states that force_expire_client() can free nfs4_client before nfs4_put_stateowner() later dereferences cl_lock in freed memory.
* +2 Remote reachable / network-triggerable surface: the vulnerable state is created through NFSD NFSv4 client and blocked-lock handling, reachable by an NFSv4 client with access to the server export.
* +1 Reliable kernel crash / strong DoS: dereferencing cl_lock in freed kernel memory can realistically cause Oops, panic, or host-wide NFSD/kernel failure.
* -1 Hard timing race: the UAF requires concurrency between blocked-lock reaping in nfs4_laundromat() and client expiration through force_expire_client().
* Conservative raw total: 4.
* Floor to 5: the object is an active structured nfs4_client, the stale field is used in state-owner cleanup, and the path is reachable below host-admin through NFSv4 client state operations.

Paranoid signals:

* +2 Real UAF / active lifetime violation: same as conservative.
* +2 Remote reachable / network-triggerable surface: same as conservative.
* +1 Reliable kernel crash / strong DoS: same as conservative.
* +1 Structured stale-object exploitation leverage: stale use reaches an active synchronization and client-state field, cl_lock, in a freed nfs4_client while cleanup is still processing state ownership. This is more than passive metadata, although controlled reclaim is not demonstrated.
* +0 Race penalty withheld in paranoid score: repeated NFSv4 client and lock-state activity can plausibly provide repeated attempts, so the race remains AC:H-like for CVSS but is not used as an ActionableScore reducing signal in the paranoid interpretation.
* Paranoid raw total: 6.

Important non-awarded signals:

* Strong LPE +2 not awarded: no controlled reclaim, object replacement, arbitrary write, callback control, credential manipulation, or demonstrated privilege escalation is supplied.
* Confidentiality and integrity points not awarded separately: the patch establishes UAF and stale cl_lock dereference, not a concrete leak or unauthorized modification primitive.
* Concrete repeated release/refcount +1 not awarded: this is not shown as the exact same nfs4_client reference receiving a second lifetime-mutating put after a first put may have freed it. The visible stale operation is dereference or lock use of cl_lock.
* Extra Availability +1 not awarded: the kernel crash/DoS is already counted once.
* Namespace-scoped privilege not awarded: this is an NFSD network server path, not a proven attacker-created namespace capability path.
* Auto-downgrade not allowed: the network-reachable active-object UAF leaves meaningful Important-vs-Moderate exploitation leverage unresolved.

3. Reachability analysis

The likely attacker is a remote NFSv4 client that can connect to the NFSD server and access an export sufficiently to create lock-owner and blocked-lock state. The supplied context indicates PR:L for CVSS-style analysis: the attacker needs NFS client access and the ability to create lock state, but not root or host-admin privileges on the NFS server.

HostAdminRequired=NO because the vulnerable objects are driven by client-visible NFSv4 state handling rather than by a host-admin-only local control-plane operation, based on the supplied triage context. NamespaceScopedPrivilege=NO because the assessed trigger is remote NFSD protocol activity, not a namespaced local capability.

There is prior setup in the ordinary deployment sense: NFSD and NFSv4 exports must be configured and reachable. That is not scored as host-admin required for the attacker, because exposed NFS service is the victim configuration. No special hardware, boot mode, firmware condition, or rare CPU feature is required.

The race is real and nontrivial: blocked-lock reaping must overlap with client expiration. Conservative scoring treats this as a hard timing condition. Paranoid scoring does not apply a timing penalty because repeated creation of NFSv4 client and lock state may make repeated attempts feasible, but this does not prove reliable exploitation beyond DoS.

4. Primitive and severity interpretation

The exact primitive is a use-after-free of struct nfs4_client. A bare lock owner keeps a raw pointer to its nfs4_client without holding a lifetime reference. The laundromat can detach and reap a blocked lock, while force_expire_client() can concurrently free the client. Later state-owner cleanup can still dereference cl_lock from the freed nfs4_client.

This is an active structured object UAF, not fixed stale storage. The stale object is part of NFSD NFSv4 client and lock-owner state. The stale use is partly cleanup-related, but it is not merely passive bookkeeping because it dereferences an active client lock field during state-owner teardown.

There is no concrete repeated release/refcount operation established for the same dangling nfs4_client reference. The first relevant lifetime loss is client freeing by force_expire_client(). The later stale operation is use of cl_lock in freed memory, not a clearly shown second put/free/refcount mutation of the same exact client reference.

PrimitiveClass=CLASS_B because there is a real kernel lifetime violation with active structured state, but no controlled reclaim, arbitrary read/write, callback hijack, or demonstrated LPE. Demonstrated consequence is at least kernel crash or DoS. Speculative consequence is stronger memory corruption or privilege escalation if freed nfs4_client storage can be usefully reclaimed or if stale lock operations can be shaped, but that is not established by the supplied patch.

ImportantCandidate=YES because the bug is a network-reachable active-object UAF in NFSD and exploitation leverage beyond crash remains unresolved. AutoDowngradeAllowed=NO because automatic closure as ordinary Moderate could discard a meaningful unresolved kernel lifetime-corruption primitive.

Likely operational severity is Actionable Moderate at minimum, with Strong Important review warranted. The evidence is not strong enough for a direct Strong Important score of 7 because no controlled reclaim, arbitrary write, information disclosure, or demonstrated privilege escalation is shown.

5. One-sentence report phrase

A network-reachable NFSD NFSv4 blocked-lock lifetime race can free an nfs4_client while later state-owner cleanup still dereferences its cl_lock, causing a kernel UAF and at least a DoS, with stronger exploitation impact requiring manual review.

6. Manual review recommendation

MANUAL CHECK REQUIRED

Reason: this is a real active structured kernel UAF in a remotely reachable NFSD path, and while only crash/DoS is directly demonstrated, the post-free client-state use is significant enough that Important-vs-Moderate exploitation leverage should not be automatically downgraded.

7. Call-site confidence

Call-site confidence: high

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

Patch: NFSD: Prevent client use-after-free during blocked-lock reaping [ Upstream
Commit: cd489b03587378645fe0d20142a33f1ed60bac98
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=cd489b03587378645fe0d20142a33f1ed60bac98

Changed files:
  fs/nfsd/nfs4state.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=cd489b03587378645fe0d20142a33f1ed60bac98

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

The original Linux kernel CVE announcement for CVE-2026-90036 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90036

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:N/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H
  The Best / paranoid CVSS vector: AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H
  The Best / paranoid CVSS score: 7.5

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: 5
  Paranoid ActionableScore: 6

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: IMPORTANT
KPANIC detected for this report: YES
Published priority for this report (same as in Subject): IMPORTANT

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).