* [CVE-2026-98076][MODERATE REGULAR] tracing/probes: Fix use-after-free on field name/type of events with multiple probes [ Upstream
@ 2026-09-25 16:41 21% AL-KERNEL
0 siblings, 0 replies; 1+ results
From: AL-KERNEL @ 2026-09-25 16:41 UTC (permalink / raw)
To: kernel-cve
CVE: CVE-2026-98076
Priority: MODERATE REGULAR
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: tracing/probes: Fix use-after-free on field name/type of events with multiple probes [ Upstream
Commit: 68ae584169c7a41fc9bc4677c1d368983cf44b21
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=68ae584169c7a41fc9bc4677c1d368983cf44b21
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98076
Analysis date: Fri, 25 Sep 2026 12:41:41 -0400
ActionableScore: 4
ActionableScore lower bound: unknown
Actionable bucket: Borderline / manual review recommended
Manual review required: YES
Summary:
No one-sentence report phrase was found.
======================================================================
ABOUT THIS REPORT
======================================================================
The original Linux kernel CVE announcement for CVE-2026-98076 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98076
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-2026-98076 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-98076 CHECK WITH IMPACT FROM ORIG NN MODERATE Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:H';*CWE-416;CWE-825;*CWE-664;Other CVSS 'AV:L/AC:L/PR:H/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '6.6';DESCR 'A deterministic use-after-free exists in probe based tracing events with multiple probes because the event field list keeps name and type pointers owned by the first probe. Deleting only that probe frees the argument strings while a sibling probe keeps the shared event and its fields registered. A later field lookup such as applying an event filter passes the dangling field name to strcmp and can cause an invalid read, kernel Oops, or crash. This is not a timing race because deletion and field lookup are serialized by event_mutex and the documented sequence is reliably reproducible. For the CVSS the PR:L value is used for the paranoid score because some systems may delegate dynamic tracing control to a restricted service account or container administrator, although normal tracefs probe management usually requires administrative privileges. The issue is local and is not network reachable. Impact is at least local denial of service and the confirmed primitive is a read through freed memory. Limited confidentiality or integrity impact is retained only in the paranoid score because slab reuse can change the referenced string contents, but the available code does not demonstrate an arbitrary write or a reliable privilege escalation primitive.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline / manual review recommended (with actual score 4) MAYBE TRACE NOMEMCHK RACE UAF LINUS DECREASED_TO_MODERATEREG_BASED_ON_FALSEPOSCHECKOFKP - - checked
======================================================================
ACTIONABLESCORE ANALYSIS
======================================================================
ActionableScore=4
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=UNKNOWN
NamespaceScopedPrivilege=NO
ImportantCandidate=NO
ManualReview=RECOMMENDED
AutoDowngradeAllowed=YES
1. ActionableScore
* Raw conservative score: 3.
* Conservative score: 4. REAL UAF FLOOR 4 changes 3 to 4 because a freed heap string is actually consumed and host-admin-equivalent privilege is not positively established.
* Raw paranoid score: 4.
* Paranoid score: 4. No additional floor or cap applies.
* Final recommended bucket: **Borderline / manual review recommended**::
2. Signal breakdown
Conservative signals:
* +2 Real UAF. Removing the first probe frees `parg->name` and possibly `parg->fmt`, while the surviving shared event retains pointers to those allocations.
* +1 Reliable kernel fault or DoS. A later field lookup passes the dangling name to `strcmp`, with a reproducible KASAN slab-use-after-free and plausible Oops or crash.
* +0 Race or timing. This is a deterministic ordered lifetime error under `event_mutex`, not a race.
* +0 Privilege penalty. Dynamic-probe management is commonly restricted, but the supplied evidence does not positively establish that host-admin-equivalent privilege is mandatory for every complete trigger.
* +0 Extra availability. The same UAF-induced fault is not counted twice.
Paranoid signal:
* +1 Attacker-influenced repeatable lifetime transition. An actor with write access to the relevant tracefs controls creation of probes A and B, deletion of A, and the subsequent filter lookup. This replaces the conservative crash-only interpretation for raw arithmetic rather than stacking another crash point.
Material non-awarded signals:
* Controlled reclaim or attacker-selected replacement at the stale address is not demonstrated.
* Arbitrary or directed kernel write is not demonstrated.
* Structured stale-object leverage is not awarded. The stale allocations contain field name and type strings, and the demonstrated use is a read by `strcmp`, not consumption of callbacks, pointers, credentials, or ownership fields.
* Weak LPE concern is not awarded merely because slab reuse is theoretically possible.
* Confidentiality is not awarded because no stale bytes are copied to userspace or another observable output.
* Integrity is not awarded because altered field matching is not shown to modify protected state or cross a security boundary.
* Remote reachability is not established.
* No concrete repeated release, put, free, or refcount mutation occurs through the dangling reference.
* The first free ends the string allocation lifetime, but the later operation is an invalid read rather than a second lifetime mutation.
* Concrete repeated release or refcount +1 is therefore not awarded.
* Paranoid repeated-refcount leverage is withheld.
* FLOOR 5 is not applied because there is no second lifetime mutation, controlled reclaim, security-sensitive structured processing, or other meaningful exploitation surface beyond the stale read.
3. Reachability analysis
The complete trigger requires control of dynamic probe definitions and event filters through tracefs. An authorized actor can create two probes sharing an event, remove the probe owning the field strings, and then initiate field lookup by writing a filter. The sequence is reliable and does not require a timing window.
Ordinary unprivileged reachability is not established. Dynamic tracing files are normally protected by tracefs permissions and system policy, but the supplied material does not prove an unconditional host capability check covering every operation. Access may also be delegated through file permissions or an already opened descriptor. Therefore `HostAdminRequired=UNKNOWN`, and no speculative host-admin penalty is applied.
Dynamic kprobe and event definitions are global kernel tracing facilities rather than resources scoped to an attacker-created user or network namespace. The complete operation is not made namespace-scoped merely by running inside a container, so `NamespaceScopedPrivilege=NO`.
Tracefs must be available and the attacker must have write access to the relevant control files. No special hardware, network exposure, boot mode, or narrow race is required.
4. Primitive and severity interpretation
The exact primitive is a deterministic heap use-after-free read. The event stores pointers to separately allocated field name and type strings. Removing their owning probe frees those strings, while a sibling probe keeps the event and dangling field pointers alive. A later lookup dereferences the stale name through `strcmp`.
This is active event metadata rather than fixed DMA storage, but it is not a rich active structured object whose pointers, callbacks, ownership, or refcount fields are consumed after free. The demonstrated stale operation is a string read used for field matching.
There is no concrete repeated release or refcount operation. The first operation frees the strings, but the second operation does not mutate their lifetime metadata. It only reads the stale allocation. This is therefore an ordinary stale dereference rather than a second lifetime mutation through a dangling object.
The issue is `CLASS_B` because an actual post-lifetime kernel access is established. It is not `CLASS_A` because controlled reclaim, directed corruption, arbitrary read or write, disclosure, callback control, and privilege escalation are not established.
The demonstrated consequence is an invalid read with possible kernel Oops or crash. Slab reuse could alter the bytes observed by field lookup, but no security-useful outcome from that behavior is shown. `ImportantCandidate=NO` because the available evidence supports a constrained UAF rather than an Important-grade exploitation primitive. `AutoDowngradeAllowed=YES` because no meaningful unresolved reuse, lifetime-mutation, authorization, or boundary-bypass primitive remains beyond the crash-capable stale read.
Likely operational severity is Moderate, with manual confirmation recommended because it is a real and reliably reproducible kernel UAF.
5. One-sentence report phrase
A reliably reproducible tracing UAF leaves shared event field names and types pointing to freed probe strings, allowing a later filter lookup to perform a stale heap read and potentially crash the kernel.
6. Manual review recommendation
MANUAL CHECK RECOMMENDED
The defect is a confirmed deterministic UAF, but the demonstrated primitive is limited to a stale string read and no Important-grade reuse, write, disclosure, or privilege-escalation mechanism is established.
7. Call-site confidence
Call-site confidence: high
======================================================================
UPSTREAM PATCH SUMMARY
======================================================================
Patch: tracing/probes: Fix use-after-free on field name/type of events with multiple probes [ Upstream
Commit: 68ae584169c7a41fc9bc4677c1d368983cf44b21
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=68ae584169c7a41fc9bc4677c1d368983cf44b21
Changed files:
kernel/trace/trace_probe.c
kernel/trace/trace_probe.h
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=68ae584169c7a41fc9bc4677c1d368983cf44b21
======================================================================
DETAILED REPORT METHODOLOGY
======================================================================
The original Linux kernel CVE announcement for CVE-2026-98076 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98076
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:L/PR:H/UI:N/S:U/C:N/I:N/A:H
The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:H
The Best / paranoid CVSS score: 6.6
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: unknown
Paranoid ActionableScore: 4
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 [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 --
2026-09-25 16:41 21% [CVE-2026-98076][MODERATE REGULAR] tracing/probes: Fix use-after-free on field name/type of events with multiple probes [ 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