[CVE-2026-90038][MODERATE REGULAR] NFSD: Prevent client use-after-free during export state revocation [ Upstream

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

CVE: CVE-2026-90038
Priority: MODERATE REGULAR
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: NFSD: Prevent client use-after-free during export state revocation [ Upstream
Commit: c05ae58fe06ce2ac34312f649e71c3596bdde358
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=c05ae58fe06ce2ac34312f649e71c3596bdde358
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90038
Analysis date: Wed, 16 Sep 2026 09:36:00 -0400
ActionableScore: 3
ActionableScore lower bound: 2
Actionable bucket: Borderline / manual review recommended
Manual review required: YES

Summary:
NFSD export state revocation can race client expiry because `nfsd4_revoke_export_states()` drops `client_lock` without pinning the `nfs4_client`, allowing a freed client object to be used during `revoke_one_stid()` or a later `cl_minorversion` read and causing a kernel DoS.

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

The original Linux kernel CVE announcement for CVE-2026-90038 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90038

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-2026-90038	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-90038 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:H/I:H/A:H';*CWE-416;CWE-362;CWE-667;Other CVSS 'AV:L/AC:H/PR:H/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '7';DESCR 'NFSD export state revocation can hit a use after free because nfsd4_revoke_export_states() drops nn client_lock while holding only a stateid reference, and that reference does not pin the nfs4_client object. A concurrent client teardown or expiry can free the client while revoke_one_stid() or the following cl_minorversion read still uses it, which can crash the kernel and may create memory corruption risk. The main trigger path is local control plane export removal through exportfs -u or NFSD_CMD_UNLOCK_EXPORT, with a remote NFS client mainly contributing the racing expiry condition. For the CVSS the PR:L in the paranoid score is used for environments where export management is delegated to a service account or reduced capability root rather than full host administration. The issue is not directly reachable by ordinary NFS packet traffic alone. Impact is at least denial of service and in the paranoid interpretation may include privilege escalation due to UAF in kernel state handling.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline / manual review recommended (with actual score 3)  SKIP CVE-2026-90038 UNKNOWN SKIP The Fixes patch not applied yet, so unlikely that actual: url YES  NO NO unknown 	MAYBE	DANGER DISK RACE UAF NETWORK  DECREASED_TO_MODERATEREG_BASED_ON_ACTIONABLESCORELESSTHAN4	-	-	checked

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

ActionableScore=3
ActionableScoreLower=2
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=UNKNOWN
NamespaceScopedPrivilege=UNKNOWN
ImportantCandidate=NO
ManualReview=RECOMMENDED
AutoDowngradeAllowed=YES

## 1. ActionableScore

* Raw conservative score: 2
* Conservative score: 2
* Raw paranoid score: 3
* Paranoid score: 3
* Final recommended bucket: **Borderline / manual review recommended**::

No floor or cap was applied. The paranoid score differs because the timing difficulty is not applied there: the race is real, but repeatability under export revocation plus client expiry is not fully resolved from the supplied context.

## 2. Signal breakdown

Conservative signals:

* +2 Real UAF / lifetime violation: `nfsd4_revoke_export_states()` drops `nn->client_lock` while holding only a stateid reference, and that does not pin `struct nfs4_client`.
* +1 Reliable kernel crash / strong DoS: a racing teardown can free `clp` while `revoke_one_stid()` or the later `clp->cl_minorversion` read still uses it.
* -1 Hard or special timing race: the described trigger requires export state revocation to race client expiry while the lock is dropped.

Paranoid signals:

* +2 Real UAF / lifetime violation: same as above.
* +1 Reliable kernel crash / strong DoS: same as above.
* 0 race penalty withheld: the race is explicit, but the supplied data does not prove it is one-shot or practically unrepeatable when an operator repeatedly removes exports and clients expire.

Important non-awarded signals:

* Controlled reclaim not demonstrated.
* Arbitrary write not demonstrated.
* Confidentiality and integrity points not awarded merely because UAF could theoretically corrupt reused memory.
* Structured stale-object leverage not awarded: `struct nfs4_client` is active and structured, but no attacker-shaped replacement or security-useful stale field consumption is established.
* Weak LPE not awarded solely from UAF class.
* Concrete repeated release/refcount signal not awarded: the fix adds `cl_rpc_users` pinning, but the vulnerable path does not show a first release followed by a second concrete lifetime/refcount mutation on the exact same dangling reference.
* Extra Availability not awarded because it would double-count the same kernel crash.
* Remote reachability not awarded: ordinary NFS packet traffic alone is not shown to drive the complete vulnerable operation.

Repeated-release/refcount check:

* First release can end lifetime: client expiry/teardown can free the client.
* Exact same reference survives: `clp` can remain in local use while lock is dropped.
* Second operation mutates lifetime/refcount: no, the visible stale operations are dereference/use, not a second `put` or refcount decrement.
* Dedicated +1 Concrete repeated release/refcount: not awarded.
* Paranoid repeated-refcount leverage +1: not awarded.
* FLOOR 5: not applied.

## 3. Reachability analysis

The visible trigger path is local control-plane export removal: `exportfs -u` reaches this path through `NFSD_CMD_UNLOCK_EXPORT`. The commit text calls this an administrator action, so practical triggering appears to require export-management privilege on the NFS server.

However, the supplied snippet does not fully resolve the exact authorization boundary for `NFSD_CMD_UNLOCK_EXPORT`, whether it is strictly host-admin-equivalent in all deployments, or whether a reduced service account can perform the operation. Therefore `HostAdminRequired=UNKNOWN` is used rather than assuming either ordinary local reachability or guaranteed host-root-only reachability.

A remote NFS client can contribute the racing client expiry condition, but the complete trigger is not shown to be reachable by remote NFS traffic alone. No namespace-scoped complete trigger chain is established. No special hardware requirement is present. The race is real and depends on export revocation overlapping client expiry.

## 4. Primitive and severity interpretation

The exact primitive is a race-based use-after-free of `struct nfs4_client`. The object is an active structured NFSD client object, not a fixed stale DMA buffer. The stale use occurs during export state revocation and includes use inside `revoke_one_stid()` plus a later read of `clp->cl_minorversion`.

This is `PrimitiveClass=CLASS_B`: a real kernel lifetime violation exists, but no controlled reclaim, arbitrary R/W, callback control, credential manipulation, or demonstrated privilege escalation is established.

Demonstrated consequence is kernel crash / DoS. Speculative consequence is broader memory corruption if freed client memory is reused in an exploitable way, but that is not established by the patch. Because the complete trigger appears tied to export-management control and no meaningful exploitation primitive beyond stale dereference is shown, this does not rise to an Important candidate from the supplied evidence alone.

`ImportantCandidate=NO` because the available evidence supports a real UAF but not a strong unresolved exploitation primitive such as controlled reclaim, repeated lifetime mutation, or security-boundary bypass. `AutoDowngradeAllowed=YES` because this appears to be a constrained control-plane UAF with demonstrated DoS impact and no established leverage beyond crash, though manual review is still recommended due to kernel UAF semantics in NFSD state handling.

## 5. One-sentence report phrase

NFSD export state revocation can race client expiry because `nfsd4_revoke_export_states()` drops `client_lock` without pinning the `nfs4_client`, allowing a freed client object to be used during `revoke_one_stid()` or a later `cl_minorversion` read and causing a kernel DoS.

## 6. Manual review recommendation

MANUAL CHECK RECOMMENDED

Reason: this is a real race-based UAF in NFSD client state handling, but the supplied evidence mainly supports crash/DoS and does not establish controlled reclaim, repeated refcount mutation, direct remote triggering, or privilege escalation.

## 7. Call-site confidence

Call-site confidence: high

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

Patch: NFSD: Prevent client use-after-free during export state revocation [ Upstream
Commit: c05ae58fe06ce2ac34312f649e71c3596bdde358
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=c05ae58fe06ce2ac34312f649e71c3596bdde358

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=c05ae58fe06ce2ac34312f649e71c3596bdde358

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

The original Linux kernel CVE announcement for CVE-2026-90038 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90038

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:H/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:H/I:H/A:H
  The Best / paranoid CVSS score: 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).