[CVE-2026-89799][MODERATE REGULAR] bpf: Disable preemption in bpf_get_stackid [ Upstream

[CVE-2026-89799][MODERATE REGULAR] bpf: Disable preemption in bpf_get_stackid [ Upstream

[AL-KERNEL]

CVE: CVE-2026-89799
Priority: MODERATE REGULAR
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: bpf: Disable preemption in bpf_get_stackid [ Upstream
Commit: ccf481d73bce6e851cec364fe831c86fe66877cf
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=ccf481d73bce6e851cec364fe831c86fe66877cf
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89799
Analysis date: Wed, 16 Sep 2026 08:10:51 -0400
ActionableScore: 3
ActionableScore lower bound: 2
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-2026-89799 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89799

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-2026-89799	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-89799 CHECK WITH IMPACT FROM ORIG NN NONE	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:L';CWE-362;CWE-667;*CWE-664;Other CVSS 'AV:L/AC:H/PR:H/UI:N/S:U/C:L/I:L/A:N';BEST CVSS score: '4.5';DESCR 'bpf_get_stackid can access the perf callchain trace entries buffer after get_perf_callchain without keeping preemption disabled for the full lifetime of that access. This creates a race on per CPU callchain storage where a task may be preempted or migrated while stackid_fastpath or stackid_new_bucket still reads the returned trace buffer. For the CVSS the PR:L is used in the paranoid score because some deployments may allow a non root local user or a constrained service account to load and run BPF programs that use the affected helper. The issue is local and is not network reachable. Impact is limited confidentiality or integrity exposure through corrupted or cross context stack trace data, with possible limited availability impact in the paranoid case. No strong UAF, OOB write, or arbitrary write primitive is visible from this patch, so privilege escalation is not claimed without further evidence.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline / manual review recommended (with actual score 3)	YES	BPF NOMEMCHK SIMPLEFIX LINUS  INCREASED_FROM_LOW_BASED_ON_REQUIREMANUALCHECK DECREASED_TO_MODERATEREG_BASED_ON_ACTIONABLESCORELESSTHAN4	NO	NO	checked

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

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

1. ActionableScore

* Raw conservative score: 2
* Conservative score: 2. No floor or cap changed the score.
* Raw paranoid score: 3
* Paranoid score: 3. No floor or cap changed the score.
* Final recommended bucket: **Borderline / manual review recommended**::

2. Signal breakdown

Conservative signals:

* +1 Weak / indirect corruption candidate. `bpf_get_stackid()` used the trace entries buffer returned by `get_perf_callchain()` without keeping preemption disabled for the entire lifetime of the access. This can allow use of the wrong or overwritten per CPU callchain buffer.
* +1 Confidentiality plausible. The callchain entries may be copied into a BPF stack trace map, so cross context or stale stack trace data may become visible to the map reader, but only limited disclosure is supported.

Paranoid additional signal:

* +1 Broad practical review surface. BPF stack trace maps are a realistic observability feature on systems that allow BPF tracing, and the affected helper stores derived stack data into persistent map state.

Non-awarded signals:

* Real UAF not awarded. The patch shows a preemption / per CPU buffer lifetime invariant violation, not object free followed by stale post-free use.
* CLASS_A not awarded. No controlled reclaim, arbitrary write, directed kernel write, callback control, or demonstrated LPE is established.
* Strong or weak LPE not awarded. The bug class is not enough by itself, and no exploitation primitive beyond wrong or stale stack trace buffer consumption is shown.
* Reliable kernel crash not awarded. The supplied text does not establish a panic, Oops, or fatal fault.
* Integrity not awarded conservatively. Wrong stack map contents are possible, but unauthorized modification of protected kernel state is not established.
* Remote reachability not awarded. This is a local BPF helper path, not packet-triggered or network-triggered.
* Race penalty not applied. The issue is a race / preemption invariant bug, but the evidence does not establish a narrow one-shot timing window requiring an extra negative point.
* Extra availability not awarded. No distinct availability consequence beyond speculative instability is shown.

Repeated-release/refcount checklist:

* First release can end lifetime: no such release is shown.
* Exact same reference survives: not applicable.
* Second operation mutates lifetime/refcount: no.
* Concrete repeated release/refcount +1 awarded: no.
* Paranoid repeated-refcount leverage +1 awarded: no.
* FLOOR 5 applied: no.

3. Reachability analysis

The vulnerable path is reached through a BPF program using `bpf_get_stackid()` with a stack trace map. The complete trigger requires the ability to load and run a BPF program that can call this helper and access the relevant map.

Host-admin-equivalent privilege is not positively established from the supplied evidence, but below-host-admin reachability is also not fully established. Modern BPF deployments often gate loading or attaching BPF programs behind `CAP_BPF`, `CAP_PERFMON`, `CAP_SYS_ADMIN`, `CAP_NET_ADMIN`, or sysctl-controlled policy. Therefore `HostAdminRequired=UNKNOWN`.

Namespace-scoped privilege is also unresolved. The patch does not establish that all required stages, including BPF program load, attach or execution context, map creation, and helper use, can be completed inside attacker-created namespaces.

No direct network trigger is established. No special hardware requirement is established. The issue is a preemption-safety bug rather than a proven hard race. The attacker influence is mainly selection and execution of the BPF helper path, not confirmed control over kernel object lifetime or reclaim.

4. Primitive and severity interpretation

The exact primitive is a preemption / per CPU buffer synchronization violation. `get_perf_callchain()` returns a trace entries buffer that must be consumed while preemption remains disabled. Without that protection, `stackid_fastpath()` and `stackid_new_bucket()` may read inconsistent or cross-context callchain data.

This is not a real UAF and not a concrete repeated release/refcount bug. The affected storage is closer to fixed per CPU temporary callchain storage than to an attacker-reclaimable active structured object. The demonstrated issue is wrong lifetime protection for temporary per CPU data, not heap object lifetime corruption.

Primitive class is `CLASS_C`: demonstrated behavior is primarily invalid state / race behavior with limited plausible disclosure through stack trace data. The practical consequence is likely incorrect or cross-context stack trace map entries. Privilege escalation is speculative and unsupported by the patch.

`ImportantCandidate=NO` because there is no strong memory corruption primitive, no controlled reuse, no arbitrary write, no security-boundary bypass, and no demonstrated LPE. `AutoDowngradeAllowed=YES` because the unresolved primitive does not currently preserve Important-class severity, though manual review is still useful due to the BPF helper context.

Likely operational severity is ordinary to borderline Moderate. It should not be treated as a strong Important candidate without additional evidence showing controlled corruption, kernel address disclosure of higher value, or a below-host-admin exploit path.

5. One-sentence report phrase

A preemption race in `bpf_get_stackid()` can cause the helper to consume an unsafe per CPU perf callchain buffer after `get_perf_callchain()`, potentially producing stale or cross context stack trace data in a BPF stack map.

6. Manual review recommendation

MANUAL CHECK RECOMMENDED

Reason: this is a BPF helper race involving perf callchain data copied into persistent map state, but the supplied evidence does not establish UAF, arbitrary write, reliable crash, or privilege escalation.

7. Call-site confidence

Call-site confidence: medium

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

Patch: bpf: Disable preemption in bpf_get_stackid [ Upstream
Commit: ccf481d73bce6e851cec364fe831c86fe66877cf
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=ccf481d73bce6e851cec364fe831c86fe66877cf

Changed files:
  kernel/bpf/stackmap.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=ccf481d73bce6e851cec364fe831c86fe66877cf

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

The original Linux kernel CVE announcement for CVE-2026-89799 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89799

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:L/I:L/A:N
  The Best / paranoid CVSS vector: AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:L
  The Best / paranoid CVSS score: 4.5

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).