* [CVE-2026-98364][IMPORTANT] xfrm: hold net_device reference under RCU in bundle creation [ Upstream
@ 2026-10-08 10:30 21% AL-KERNEL
0 siblings, 0 replies; 1+ results
From: AL-KERNEL @ 2026-10-08 10:30 UTC (permalink / raw)
To: kernel-cve
CVE: CVE-2026-98364
Priority: IMPORTANT
AL-KERNEL base severity: IMPORTANT
KPANIC flag: YES
Patch: xfrm: hold net_device reference under RCU in bundle creation [ Upstream
Commit: 8d85d6bc9c46dc41cb2be0eac53f8ab068c3a807
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=8d85d6bc9c46dc41cb2be0eac53f8ab068c3a807
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98364
Analysis date: Thu, 08 Oct 2026 06:30:04 -0400
ActionableScore: 5
ActionableScore lower bound: 4
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-2026-98364 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98364
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-2026-98364 IMPORTANT https://www.kernelcve.org/list/?q=CVE-2026-98364 CHECK WITH IMPACT FROM ORIG NN MODERATE | AEGIS IMPACT MODERATE_KPANIC AEGIS_AS=4 LEGACY=IMPORTANT RESULT=DIFF (legacy and Aegis reconciliation differ; DECISION: legacy IMPORTANT vs Aegis MODERATE_KPANIC) 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:L/I:L/A:H';BEST CVSS score: '7';DESCR 'A use after free in XFRM bundle creation occurs because xfrm_bundle_create and xfrm_create_dummy_bundle can retain dst dev without holding a net_device reference. A concurrent RTM_DELLINK operation can replace and free the old net_device before xfrm_fill_dst or xfrm6_fill_dst uses the stale pointer. The observed access reaches netdev_hold on the freed object and can cause a kernel crash. For the CVSS the PR:L because reliable triggering requires local control of interface removal and XFRM related traffic or route lookup, typically through CAP_NET_ADMIN in a network namespace. This is not directly network reachable because a remote packet alone does not provide control over RTM_DELLINK. The base impact is local denial of service with possible limited confidentiality or integrity impact from the confirmed use after free. The paranoid score uses C:H I:H A:H because the stale object is accessed through reference handling and the memory corruption risk should be reviewed manually.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 5) MAYBE READ LOCK DANGER UAF HARDWARE IPV6 KPANIC PACKET KPANIC INCREASED_TO_HIGH_BASED_ON_GUESSCVSS - - checked
======================================================================
ACTIONABLESCORE ANALYSIS
======================================================================
ActionableScore=5
ActionableScoreLower=4
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=NO
NamespaceScopedPrivilege=YES
ImportantCandidate=YES
ManualReview=REQUIRED
AutoDowngradeAllowed=NO
1. ActionableScore
* Raw conservative score: 4
* Conservative score: 4
* Raw paranoid score: 5
* Paranoid score: 5
* Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**::
The conservative score counts the confirmed active-object UAF, the broken RCU lifetime relationship, and realistic kernel crash impact. The paranoid score additionally counts attacker-influenced repeatable lifetime manipulation where an actor with CAP_NET_ADMIN in the relevant network namespace can drive RTM_DELLINK while causing XFRM bundle lookups.
2. Signal breakdown
* +2 Real UAF / active lifetime violation. The patch directly establishes that dst->dev can be replaced and the previous net_device freed while xfrm_bundle_create or xfrm_create_dummy_bundle retains and later uses the stale pointer.
* +1 Concrete RCU referenced-object lifetime mismatch. The dst object remains usable while its referenced net_device can be released before readers complete. The fix extends the RCU read-side critical section until xfrm_fill_dst obtains the required device references.
* +1 Reliable kernel crash / strong DoS. KASAN reports slab-use-after-free in xfrm6_fill_dst during netdev_hold. A stale net_device dereference in networking output processing can crash the kernel.
* +1 Paranoid only. Attacker-influenced active-object lifetime manipulation. A capable actor can repeatedly cause interface removal through RTM_DELLINK and initiate IPv6 UDP or XFRM route lookup activity in the same network namespace. This provides a plausible repeatable race-driving mechanism, but the patch does not demonstrate controlled post-free object replacement.
Important non-awarded signals:
* Controlled reclaim or controlled replacement of the freed net_device is not demonstrated.
* Arbitrary kernel read or directed kernel write is not demonstrated.
* Structured stale-object exploitation leverage is not separately awarded. The stale operation reaches netdev_hold, but no attacker-controlled replacement object or attacker-shaped post-free field consumption is established.
* Weak LPE concern is not awarded solely because the bug is a UAF.
* Concrete repeated release/refcount operation is not established. The stale operation is a netdev_hold reference acquisition attempt, not a visible first release followed by a second put or release on the exact same dangling reference.
* No paranoid repeated-refcount bonus is awarded because the dedicated repeated release/refcount mechanics are absent.
* No extra Availability point is awarded because the observed kernel crash is already counted.
* Confidentiality and Integrity consequence points are not awarded. The stale reference may lead to memory corruption, but no disclosure, unauthorized modification, controlled reuse, or privilege escalation is demonstrated.
* Direct remote reachability is not established. A remote packet alone does not perform RTM_DELLINK or control the required network-device lifetime transition.
* No hard-race negative is applied. The affected operations can plausibly be repeated by a capable actor, and the patch does not establish a one-shot or externally uncontrollable timing condition.
* No host-admin penalty is applied. The relevant RTM_DELLINK and XFRM operations are network-namespace scoped and do not inherently require host root.
3. Reachability analysis
The evaluated attacker is a local process with CAP_NET_ADMIN in the affected network namespace. The required sequence is to manage a network device, trigger its deletion through RTM_DELLINK, and concurrently cause XFRM bundle creation through an IPv6 route or UDP send path.
RTM_DELLINK authorization is evaluated against the target network namespace user namespace. XFRM policy and route processing are also per-network-namespace operations. Therefore the complete vulnerable sequence can in principle be exercised with namespace-scoped CAP_NET_ADMIN and does not inherently require init_user_ns privileges or host root.
Ordinary unprivileged reachability is not awarded because availability of unprivileged user namespaces and delegated network administration is distribution and deployment dependent. However, a container root, delegated network-management service, or process with CAP_NET_ADMIN in a non-host network namespace is below host-admin-equivalent privilege.
No special hardware, firmware condition, boot option, or rare device is indicated. XFRM configuration is required, but the vulnerable lookup path is reached from normal IPv6 UDP output processing once relevant policy and routing state exist.
4. Primitive and severity interpretation
The exact primitive is a concrete RCU lifetime violation involving an active structured net_device object. A reader retains dst->dev without a device reference while a concurrent RTM_DELLINK path can replace the pointer and release the old device. The later stale use enters xfrm_fill_dst and xfrm6_fill_dst, where netdev_hold accesses the freed object.
This is not fixed stale DMA storage or a shallow cleanup-only dereference. The stale net_device is used during active XFRM bundle construction and reference handling. Nevertheless, controlled reuse, replacement with attacker-controlled contents, callback dispatch, arbitrary read/write, credential manipulation, and privilege escalation are not established.
The issue is CLASS_B because it is a real active-object UAF with an RCU lifetime mismatch. It is not CLASS_A because the supplied evidence does not establish controlled reclaim, refcount takeover, directed corruption, arbitrary write, or a demonstrated security-boundary consequence.
The demonstrated practical consequence is local kernel crash or denial of service. The speculative consequence is post-free reference handling against reused storage, but this remains unresolved rather than demonstrated exploitation leverage. ImportantCandidate is YES because this is an active structured-object UAF with a concrete RCU lifetime failure and namespace-reachable lifetime manipulation. AutoDowngradeAllowed is NO because automatically reducing it to ordinary crash-only behavior would discard unresolved active-object memory-corruption risk.
Likely operational severity is Actionable Moderate, with Important-level manual review warranted for environments that permit untrusted tenants, containers, or service accounts to obtain CAP_NET_ADMIN in network namespaces with XFRM configuration.
5. One-sentence report phrase
A local attacker with network namespace CAP_NET_ADMIN may race RTM_DELLINK against XFRM bundle creation and trigger a use after free of net_device in xfrm6_fill_dst, causing kernel denial of service and requiring manual review for post-free reference handling risk.
6. Manual review recommendation
MANUAL CHECK REQUIRED
The patch establishes a concrete active-object RCU UAF in which a freed net_device reaches netdev_hold during XFRM processing. Controlled reclaim is not proven, but the lifetime violation is stronger than an ordinary invalid-pointer crash and can be driven through namespace-scoped network administration.
7. Call-site confidence
Call-site confidence: medium
======================================================================
UPSTREAM PATCH SUMMARY
======================================================================
Patch: xfrm: hold net_device reference under RCU in bundle creation [ Upstream
Commit: 8d85d6bc9c46dc41cb2be0eac53f8ab068c3a807
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=8d85d6bc9c46dc41cb2be0eac53f8ab068c3a807
Changed files:
net/ipv6/xfrm6_policy.c
net/xfrm/xfrm_policy.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=8d85d6bc9c46dc41cb2be0eac53f8ab068c3a807
======================================================================
DETAILED REPORT METHODOLOGY
======================================================================
The original Linux kernel CVE announcement for CVE-2026-98364 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98364
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:L/I:L/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: 4
Paranoid ActionableScore: 5
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 [email protected] (and both
send reply to CVE record itself too and see "reply" button below for howto reply).
^ permalink raw reply [relevance 21%]
Results 1-1 of 1 | reverse | sort options + mbox downloads above
-- links below jump to the message on this page --
2026-10-08 10:30 21% [CVE-2026-98364][IMPORTANT] xfrm: hold net_device reference under RCU in bundle creation [ Upstream AL-KERNEL
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox