* [CVE-2026-98033][MODERATE 7.0] bpf: Preserve inner map identity in callback frames [ Upstream
@ 2026-09-26 18:29 21% AL-KERNEL
0 siblings, 0 replies; 1+ results
From: AL-KERNEL @ 2026-09-26 18:29 UTC (permalink / raw)
To: kernel-cve
CVE: CVE-2026-98033
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: bpf: Preserve inner map identity in callback frames [ Upstream
Commit: 66ff2eff093d2a0838f39c713eeb590bc0946c2a
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=66ff2eff093d2a0838f39c713eeb590bc0946c2a
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98033
Analysis date: Sat, 26 Sep 2026 14:29:00 -0400
ActionableScore: 4
ActionableScore lower bound: unknown
Actionable bucket: Borderline / manual review recommended
Manual review required: YES
Summary:
A BPF verifier map identity loss can bind a timer to the wrong inner map without retaining a reference, allowing a later timer callback to dereference a freed kernel map and cause local denial of service.
======================================================================
ABOUT THIS REPORT
======================================================================
The original Linux kernel CVE announcement for CVE-2026-98033 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98033
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-2026-98033 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-98033 CHECK WITH IMPACT FROM ORIG NN LOW 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-284;*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 BPF callback verification can occur because callback frame construction restores map_ptr but loses map_uid which distinguishes inner maps sharing the same template. A local process can use callbacks involving bpf_for_each_map_elem and BPF timers to make a timer retain a pointer to a different inner map without a reference. Releasing that map can leave a dangling pointer which is later dereferenced by a timer callback in kernel context. For the CVSS the PR:L is selected because an attacker needs local code execution and the ability to load a suitable BPF program. The issue is not directly network reachable. Impact is at least local denial of service through a kernel use-after-free. The paranoid assessment allows confidentiality and integrity impact because the patch describes a stale kernel map pointer dereferenced after free. Reliable exploitation is complex and depends on BPF permissions and the affected map and timer features.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline / manual review recommended (with actual score 4) YES BPF UAF IMPROVEONLY LINUS KPANIC INCREASED_FROM_LOW_BASED_ON_REQUIREMANUALCHECK YES NO checked
======================================================================
ACTIONABLESCORE ANALYSIS
======================================================================
ActionableScore=4
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=UNKNOWN
NamespaceScopedPrivilege=UNKNOWN
ImportantCandidate=NO
ManualReview=RECOMMENDED
AutoDowngradeAllowed=YES
## 1. ActionableScore
* Raw conservative score: 4
* Conservative score: 4
* Raw paranoid score: 4
* Paranoid score: 4
* Final recommended bucket: **Borderline / manual review recommended**::
No floor or cap changed the score.
## 2. Signal breakdown
* +2 Real UAF / active lifetime violation. The verifier can accept an invalid pairing between a timer stored in one inner map and a different inner map. The timer retains the first map pointer without a reference. After that map is freed, a later timer callback dereferences the stale `struct bpf_map` pointer.
* +1 Reliable kernel crash / strong DoS. Dereferencing a freed kernel map object from an asynchronous timer callback can cause a kernel fault, Oops, or panic.
* +1 Attacker influenced active-object lifetime manipulation. The BPF program controls creation and use of the relevant inner maps, timer initialization, callback construction, and the lifecycle transition where the incorrectly referenced map is freed while the timer remains associated with another map.
Not awarded:
* +0 Controlled reclaim or controlled object replacement. The patch establishes a dangling map pointer but does not demonstrate that the attacker can reclaim the freed address with attacker-shaped data.
* +0 Arbitrary or directed kernel write. No write primitive is shown.
* +0 Structured stale-object exploitation leverage. The supplied evidence establishes a later stale dereference, but does not show consumption of attacker-controlled post-free map fields, callbacks, operations, offsets, or replacement-object contents.
* +0 Strong or weak LPE leverage. A BPF verifier bypass resulting in UAF deserves review, but the patch does not establish credential manipulation, callback hijacking, refcount takeover, arbitrary read/write, or a practical privilege-escalation chain.
* +0 Remote reachability. This is a local BPF program loading and execution path.
* +0 Local unprivileged trigger. The complete BPF load, map creation, timer setup, and callback sequence is not shown to work for an ordinary unprivileged user.
* +0 extra Availability impact. The potential kernel crash is already counted once.
* +0 C/I impact. Confidentiality or integrity effects are not established merely from the possibility that freed map storage could later be reused.
There is no concrete repeated release/refcount mutation on the same dangling reference. The relevant lifetime defect is a missing reference held by the timer, followed by a later stale dereference. Therefore the dedicated repeated release/refcount +1 and the paranoid repeated-refcount +1 do not apply.
## 3. Reachability analysis
The evaluated attacker is a local process able to load a suitable BPF program and use BPF map-in-map, BPF timers, callback frames, and `bpf_for_each_map_elem()` or equivalent callback paths.
The patch does not establish whether the complete trigger sequence is available to an ordinary unprivileged user. Modern Linux deployments commonly restrict BPF loading through `kernel.unprivileged_bpf_disabled`, `CAP_BPF`, `CAP_SYS_ADMIN`, LSM policy, verifier policy, and map-type-specific restrictions. Therefore `HostAdminRequired=UNKNOWN` is appropriate. The evidence does not prove host-admin-equivalent privilege is necessary, but also does not prove below-host-admin reachability.
`NamespaceScopedPrivilege=UNKNOWN` is also appropriate. The patch does not show that every required stage, including program loading, map creation, timer helper access, and callback execution, can be performed using only attacker-controlled namespace-scoped capabilities. A network namespace or container alone would not establish that condition.
No special hardware, firmware, boot option, or direct network exposure is indicated. The issue is not a narrow timing race. The vulnerable sequence is primarily a verifier state-confusion and object-lifetime sequence, although successful triggering requires construction of a sufficiently specific BPF program and compatible inner maps.
## 4. Primitive and severity interpretation
The exact primitive is an active structured-object use-after-free. A verifier bookkeeping error loses `map_uid` while retaining `map_ptr` in callback register state. This permits different inner maps sharing the same template to be treated as identical. A timer can then retain a non-referenced pointer to the wrong map, and freeing that map leaves a stale pointer later consumed by timer callback processing.
This is not fixed stale DMA storage or a shallow destructor-only dereference. The stale reference is retained in an active timer object and is later consumed asynchronously in callback handling. However, the available patch context does not establish post-free object replacement, attacker-controlled reclaimed contents, function-pointer or ops-table consumption, a write through the stale pointer, disclosure, or a concrete LPE chain.
The primitive is therefore `CLASS_B`, not `CLASS_A`. The demonstrated practical consequence is local kernel DoS. Privilege escalation is a theoretical possibility for some UAF classes, but it is not supported strongly enough here to receive an exploitation-leverage bonus or an Important-preserving classification.
`ImportantCandidate=NO` is selected because the supplied evidence does not establish meaningful reuse, post-free structured-field use, repeated stale refcount mutation, authorization bypass, or another attacker-useful primitive beyond the UAF and crash surface. `AutoDowngradeAllowed=YES` is selected because this is currently a privilege-gated BPF UAF with demonstrated crash potential but no established controlled reclaim or stronger exploitation mechanism. It should still receive a targeted manual BPF-permission and callback-lifetime review before operational closure.
Likely operational severity is borderline Moderate. It may become Actionable Moderate or an Important candidate if further source analysis establishes unprivileged BPF reachability, controlled post-free map reuse, dereference of attacker-controlled map operations or fields, or a concrete privilege-escalation path.
## 5. One-sentence report phrase
A BPF verifier map identity loss can bind a timer to the wrong inner map without retaining a reference, allowing a later timer callback to dereference a freed kernel map and cause local denial of service.
## 6. Manual review recommendation
MANUAL CHECK RECOMMENDED
Review the complete BPF helper permission model, affected map types, unprivileged BPF policy, and the timer callback dereference path to determine whether a below-host-admin attacker can control post-free map reuse or map operations.
## 7. Call-site confidence
Call-site confidence: medium
======================================================================
UPSTREAM PATCH SUMMARY
======================================================================
Patch: bpf: Preserve inner map identity in callback frames [ Upstream
Commit: 66ff2eff093d2a0838f39c713eeb590bc0946c2a
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=66ff2eff093d2a0838f39c713eeb590bc0946c2a
Changed files:
kernel/bpf/verifier.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=66ff2eff093d2a0838f39c713eeb590bc0946c2a
======================================================================
DETAILED REPORT METHODOLOGY
======================================================================
The original Linux kernel CVE announcement for CVE-2026-98033 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98033
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: 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: 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 [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 --
2026-09-26 18:29 21% [CVE-2026-98033][MODERATE 7.0] bpf: Preserve inner map identity in callback frames [ 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