[CVE-2026-93288][IMPORTANT] netfilter: nfnetlink_log: wait for rcu grace period before freeing pernet state [ Upstream

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

CVE: CVE-2026-93288
Priority: IMPORTANT
AL-KERNEL base severity: IMPORTANT
KPANIC flag: YES
Patch: netfilter: nfnetlink_log: wait for rcu grace period before freeing pernet state [ Upstream
Commit: f4461654374576e9d5d0245fd534c46ad8509051
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=f4461654374576e9d5d0245fd534c46ad8509051
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-93288
Analysis date: Thu, 24 Sep 2026 18:50:50 -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-2026-93288 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-93288

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-2026-93288	IMPORTANT	https://www.kernelcve.org/list/?q=CVE-2026-93288 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-664;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '7';DESCR 'A use-after-free race exists in nfnetlink_log during network namespace teardown. nf_log_unset can clear the logger pointer and the per network namespace log state can then be freed while a concurrent packet is still executing nf_log_packet under RCU protection with a previously obtained logger reference. This can cause a kernel crash and denial of service. For the CVSS the PR:L rating is used because reliable triggering requires local control of a network namespace and NFLOG configuration, including the ability to race namespace teardown with matching packet processing. The issue is not directly reachable by a remote sender alone because the sender cannot normally configure the required logging rule or destroy the target namespace. The base impact is denial of service. The paranoid impact considers the kernel use-after-free capable of causing confidentiality and integrity impact if freed pernet memory is reused in a controllable way. YES REQUIRES MANUAL CHECK.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 6)	YES	REMOTE DISK INIT SIMPLEFIX NETFILTER LOG KPANIC PACKET  KPANIC INCREASED_TO_HIGH_BASED_ON_GUESSCVSS	NO	NO	checked

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

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

1. ActionableScore

* Raw conservative score: 4
* Conservative score: 5
* Raw paranoid score: 5
* Paranoid score: 6
* Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**::

The conservative raw score is raised from 4 to 5 by the active structured RCU referenced-object lifetime mismatch calibration. The paranoid score is raised from 5 to 6 because the stale per-network namespace logging state remains reachable from active packet processing after its lifetime may have ended, while the exact post-free field consumption and reuse behavior are not visible in the supplied patch.

2. Signal breakdown

Positive signals:

* +2 Real UAF / active lifetime violation. The commit explicitly states that `ops_free_list()` can free nfnetlink_log pernet state while concurrent packets are still executing `nf_log_packet()` under `rcu_read_lock()`.
* +1 Concrete RCU referenced-object lifetime mismatch. The logger remains observable by legitimate RCU readers while the logger callback can still access pernet logging state that is freed before the relevant RCU grace period.
* +1 Kernel crash / DoS plausibility. A packet-processing callback can dereference or traverse freed pernet logging state in kernel context. The demonstrated consequence is at least a kernel fault or denial of service.

Conservative raw arithmetic: 2 + 1 + 1 = 4.

Conservative floor:

* Active structured RCU lifetime calibration: 4 to 5. `struct nfnl_log_net` is active per-network namespace packet logging state, not fixed stale DMA storage. Host-admin-equivalent privilege is not positively established by the supplied evidence.

Paranoid interpretation:

* +1 RCU active structured stale-object uncertainty. The stale state is consumed from the live packet logging path after a real lifetime violation. The patch does not show controlled reclaim, but the exact post-free access depth and fields used by the logger callback are unresolved.

Paranoid raw arithmetic: 2 + 1 + 1 + 1 = 5.

Paranoid calibration:

* Active structured RCU lifetime mismatch: 5 to 6. This is a bounded escalation for unresolved active post-free packet processing, not an assumption of arbitrary write or privilege escalation.

Important non-awarded signals:

* Controlled reclaim or attacker-controlled replacement is not demonstrated.
* Arbitrary kernel read or write is not demonstrated.
* Confidentiality or integrity impact is not awarded. A UAF alone does not establish disclosure or unauthorized modification.
* Remote reachability is not established. A remote sender may generate matching packets, but cannot normally configure NFLOG state and initiate target namespace teardown alone.
* Local unprivileged trigger is not awarded because the complete privilege chain for NFLOG configuration, namespace control, and teardown is not shown.
* Structured stale-object leverage is not separately awarded. The supplied diff does not establish attacker-controlled post-free contents, replacement, callbacks, or meaningful post-reuse field control.
* No concrete repeated release or refcount mutation on the same dangling reference is shown.
* No hard-race penalty is applied. The patch establishes a race, but does not prove that its timing is one-shot, externally uncontrollable, or materially unreliable.
* No subsystem rarity penalty is applied. Netfilter packet logging is optional, but optional configuration alone is not a supported negative scoring signal.
* No extra Availability point is awarded because the likely crash and DoS consequence is already counted once.

3. Reachability analysis

The immediate stale use occurs when a packet reaches the netfilter logging path while a network namespace is being torn down. The attacker model requires influence over both a matching packet path and the lifetime transition that removes the namespace-specific nfnetlink_log state.

NFLOG configuration normally requires network-administration authority for the relevant network namespace. However, the supplied patch does not show the exact netlink capability checks, whether they are evaluated against the target `net->user_ns`, or whether an attacker-created user and network namespace can complete the entire sequence without host-admin-equivalent privilege.

Therefore:

* HostAdminRequired is UNKNOWN. There is no proof that host root or init-user-namespace `CAP_NET_ADMIN` is required for every deployment.
* NamespaceScopedPrivilege is UNKNOWN. The bug involves per-network namespace state, but the supplied context does not prove that NFLOG configuration, packet triggering, and namespace teardown are all available to an attacker through namespace-scoped capabilities.
* Remote-only reachability is not established. A remote sender can potentially provide traffic to match a pre-existing NFLOG rule, but cannot independently arrange the necessary logging configuration and target namespace destruction.
* Prior privileged setup may be required if NFLOG rules or module support must be configured by another actor. The patch does not resolve this deployment-specific prerequisite.
* Race repeatability is unresolved. Namespace creation and teardown can generally be repeated, but the patch alone does not establish that a below-host-admin attacker can reliably configure and repeatedly drive the complete NFLOG logging and teardown sequence.

4. Primitive and severity interpretation

The exact primitive is a real RCU lifetime violation and use-after-free involving per-network namespace nfnetlink_log state. The logger pointer is removed without waiting for RCU readers that may already be executing `nf_log_packet()`. The pernet storage can then be freed while the packet logging callback still has a valid execution path to that state.

This is an active structured-object UAF, not fixed stale storage. The freed object contains active logging configuration and instance-table state used by packet logging. The stale use is associated with namespace teardown, but it is not merely an immediate destructor dereference because concurrent packet processing continues through the logger path.

No concrete repeated `put`, `release`, `free`, or refcount mutation through the same dangling reference is established. Therefore, the repeated-release/refcount signal and the repeated-release floor are not applicable.

The vulnerability is CLASS_B. It is stronger than a NULL dereference or warning-only condition because the patch directly establishes post-lifetime consumption of freed kernel state by concurrent RCU readers. It is not CLASS_A because the supplied evidence does not establish controlled reclaim, object substitution, controlled post-free contents, arbitrary read/write, callback control, information disclosure, privilege escalation, or a policy-boundary bypass.

The demonstrated practical consequence is kernel crash or denial of service. A stronger consequence remains speculative, but active packet-processing access to freed structured pernet state leaves meaningful exploitation analysis unresolved. This supports ImportantCandidate=YES and ManualReview=REQUIRED. AutoDowngradeAllowed=NO because automatically closing the issue as ordinary crash-only behavior would discard a concrete active-object RCU UAF whose post-free processing depth is not fully established by the provided context.

Likely operational severity is Actionable Moderate at minimum. The bug should be manually reviewed for actual `nfulnl_logger` callback behavior, allocation reuse characteristics of the pernet storage, capability scope for NFLOG configuration, and whether unprivileged user namespaces can complete the full trigger sequence.

5. One-sentence report phrase

A race during nfnetlink_log network namespace teardown can free active pernet logging state while concurrent RCU protected packet logging still accesses it, causing at least kernel denial of service and requiring manual review for post-free exploitation leverage.

6. Manual review recommendation

MANUAL CHECK REQUIRED

A real active-object RCU use-after-free is explicitly established, while the supplied patch does not expose enough callback and capability-check context to determine whether the stale packet-processing path is limited to crash-only behavior or can consume meaningful reused structured state.

7. Call-site confidence

Call-site confidence: medium

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

Patch: netfilter: nfnetlink_log: wait for rcu grace period before freeing pernet state [ Upstream
Commit: f4461654374576e9d5d0245fd534c46ad8509051
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=f4461654374576e9d5d0245fd534c46ad8509051

Changed files:
  net/netfilter/nf_log_syslog.c
  net/netfilter/nfnetlink_log.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=f4461654374576e9d5d0245fd534c46ad8509051

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

The original Linux kernel CVE announcement for CVE-2026-93288 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-93288

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: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: 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).