[CVE-2026-63918][MODERATE 7.0] l2tp: use refcount_inc_not_zero in l2tp_session_get_by_ifname

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

CVE: CVE-2026-63918
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: l2tp: use refcount_inc_not_zero in l2tp_session_get_by_ifname
Commit: ee80455feffb9cb62b5b58715cabeff495e666b2
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=ee80455feffb9cb62b5b58715cabeff495e666b2
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-63918
Analysis date: Sun, 19 Jul 2026 19:16:15 -0400
ActionableScore: 5
ActionableScore lower bound: 4
Actionable bucket: Strong Important candidate / Actionable Moderate at minimum
Manual review required: YES

Summary:
A race in l2tp_session_get_by_ifname() can increment a zero refcount and return a session object that is later freed by kfree_rcu(), creating a local race-dependent use-after-free with at least kernel crash impact and possible privilege escalation concern.

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

The original Linux kernel CVE announcement for CVE-2026-63918 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-63918

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 7.0
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-2026-63918	MODERATE	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-911;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '7';DESCR 'A race in l2tp_session_get_by_ifname() can return a session pointer after its refcount has already reached zero. The getter used refcount_inc() instead of refcount_inc_not_zero(), so a concurrent teardown path can queue kfree_rcu() while the lookup still returns the stale session to the caller. This creates a use-after-free candidate after the RCU grace period and can cause at least a local kernel crash. For the CVSS the PR:L value is used because practical triggering requires local code or a process able to exercise the L2TP lookup path, and in some deployments this may be reachable by reduced capability root, container root, or a service account with delegated network control rather than full host administration. The issue is not directly network reachable as a packet-only attack. Impact is at least local denial of service and in the paranoid score may include confidentiality and integrity impact due to kernel use-after-free risk.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Strong Important candidate / Actionable Moderate at minimum (with actual score 5)  SKIP CVE-2026-63918 UNKNOWN SKIP The Fixes patch not applied yet, so unlikely that actual: abe7a1a7d0b69e63b1bca5f9531023a52336784f YES  NO NO unknown 	MAYBE	LOCK RACE UAF HARDWARE LINUS KPANIC 	-	-	checked

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

ActionableScore=5
ActionableScoreLower=4

## 1. ActionableScore

* Conservative score: 4
* Paranoid score: 5
* Final recommended bucket: **Strong Important candidate / Actionable Moderate at minimum**::

## 2. Signal breakdown

Triggered signals:

* Local unprivileged or reduced-privilege local trigger: +1
  Practical triggering appears local. L2TP control paths often require CAP_NET_ADMIN, but this may be reachable by reduced-capability root, container root, service accounts, or user namespace plus net namespace configurations.

* Memory corruption, weak or race-dependent UAF primitive: +1
  The patch fixes refcount_inc() on an object whose refcount may already be zero. The returned session can become stale after kfree_rcu(), producing a slab-use-after-free. No controlled reclaim or write primitive is shown.

* Real lifetime corruption: +1
  This is a concrete refcount and RCU lifetime bug involving refcount_dec_and_test(), l2tp_session_free(), and kfree_rcu().

* Reliable kernel crash / strong DoS: +1
  The commit explicitly describes refcount_t warning and later slab-use-after-free if the returned session is dereferenced after the grace period.

* Weak LPE concern in paranoid score: +1
  The bug involves refcount misuse and UAF in a network kernel object. Privilege escalation is not demonstrated, but it is plausible enough for manual review.

Subtracted signals:

* Hard race / special timing required: -1
  The issue depends on a race between lookup and session teardown. The commit says the window is real on PREEMPT_RT, while stock PREEMPT kernels reduce practical exposure.

No remote/network-triggerable score was added. This is not a packet-only remote attack.

## 3. Reachability analysis

The bug is triggered through the L2TP session lookup by interface name while a matching session is being torn down. A local actor must be able to exercise the L2TP control or lookup path and race it against session destruction.

In typical deployments, creating and deleting L2TP sessions is controlled by CAP_NET_ADMIN or privileged network management components. However, this should not be treated as full host-root only in all cases, because container root, delegated network administration, or user namespace plus network namespace setups may lower the practical privilege boundary.

The affected path is not a common packet receive path and is not directly reachable by unauthenticated network traffic. Exploitability depends on L2TP availability, namespace policy, PREEMPT_RT or equivalent preemption behavior, and the ability to force session lookup and teardown timing.

Call-site confidence: medium. The patch and commit describe the lifetime path clearly, but full caller behavior is not included.

## 4. Severity interpretation

This behaves as an actionable Moderate at minimum, with Important-candidate handling justified for manual review. The realistic impact is local DoS through kernel crash or UAF-triggered fault. The theoretical risk is higher because this is a real lifetime bug returning a stale kernel object pointer after refcount zero and kfree_rcu scheduling.

The conservative score is capped at 4 because this is a race-dependent UAF without demonstrated attacker-controlled reclaim, type confusion, callback control, write-after-free, or arbitrary write. The paranoid score is 5 because refcount misuse plus returned stale session pointer creates enough LPE concern to avoid auto-closure.

## 5. One-sentence report phrase

A race in l2tp_session_get_by_ifname() can increment a zero refcount and return a session object that is later freed by kfree_rcu(), creating a local race-dependent use-after-free with at least kernel crash impact and possible privilege escalation concern.

## 6. Manual review recommendation

MANUAL CHECK REQUIRED

Reason: this is a real refcount/RCU lifetime bug and UAF candidate in kernel networking code. Even though exploitation is race-dependent and no controlled reclaim primitive is shown, the stale returned object and refcount-on-zero behavior make it unsuitable for automatic closure.

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

Patch: l2tp: use refcount_inc_not_zero in l2tp_session_get_by_ifname
Commit: ee80455feffb9cb62b5b58715cabeff495e666b2
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=ee80455feffb9cb62b5b58715cabeff495e666b2

Commit description:

A reader in l2tp_session_get_by_ifname() can return a pointer to a
session whose refcount has reached zero. The getter takes its
reference with plain refcount_inc(), but every other session getter
in the same file (l2tp_v2_session_get, l2tp_v3_session_get, and the
corresponding _get_next variants) uses refcount_inc_not_zero()
because the IDR/RCU lookup can race with refcount_dec_and_test() ->
l2tp_session_free() -> kfree_rcu(). The ifname getter is the only
outlier; the inconsistency was raised on-list after 979c017
("l2tp: use list_del_rcu in l2tp_session_unhash").

A reader inside rcu_read_lock_bh() that matches session->ifname can
be preempted between the strcmp() and the refcount_inc(). If the
last reference drops on another CPU in that window, the reader's
refcount_inc() runs on a counter that has reached zero. refcount_t
catches the addition-on-zero, prints "refcount_t: addition on 0;
use-after-free", saturates the counter, and returns the saturated
pointer to the caller. Session memory is held live by the in-flight
RCU read section, but the kfree_rcu() callback queued from
l2tp_session_free() will free it once the grace period closes; a
caller that dereferences the returned session past that point hits
a slab-use-after-free. On PREEMPT_RT local_bh_disable() is a per-CPU
sleeping lock and the preemption window is real; on stock PREEMPT
kernels local_bh_disable() is a preempt_count increment that closes
the cross-CPU race in practice (see below).

Use refcount_inc_not_zero() and continue the list walk on failure,
matching the other session getters in the file. The ifname getter
is the only session getter in net/l2tp/ that still uses the bare
refcount_inc() pattern; this change restores file-internal
consistency. The success path is unchanged.

Fixes: abe7a1a ("l2tp: improve tunnel/session refcount helpers")
Cc: stable@vger.kernel.org
Signed-off-by: Michael Bommarito <michael.bommarito@gmail.com>
Reviewed-by: James Chapman <jchapman@katalix.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20260523023423.2568972-1-michael.bommarito@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed files:
  net/l2tp/l2tp_core.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=ee80455feffb9cb62b5b58715cabeff495e666b2

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

The original Linux kernel CVE announcement for CVE-2026-63918 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-63918

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: MODERATE
KPANIC detected for this report: YES
Published priority for this report (same as in Subject): MODERATE 7.0

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).