[CVE-2026-98041][MODERATE 7.0] bpf: Don't predict JMP32 pointer vs zero comparisons [ Upstream

[CVE-2026-98041][MODERATE 7.0] bpf: Don't predict JMP32 pointer vs zero comparisons [ Upstream

[AL-KERNEL]

CVE: CVE-2026-98041
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: bpf: Don't predict JMP32 pointer vs zero comparisons [ Upstream
Commit: d16f0eafcaa62e677a2b5b0dabc1f8d31dfe3d4d
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=d16f0eafcaa62e677a2b5b0dabc1f8d31dfe3d4d
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98041
Analysis date: Fri, 25 Sep 2026 14:12:45 -0400
ActionableScore: 5
ActionableScore lower bound: unknown
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-2026-98041 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98041

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-2026-98041	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-98041 CHECK WITH IMPACT FROM ORIG NN LOW	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H';*CWE-20;CWE-125;*CWE-787;CWE-670;Other CVSS 'AV:L/AC:L/PR:H/UI:N/S:U/C:H/I:H/A:H';BEST CVSS score: '7.8';DESCR 'A BPF verifier logic flaw occurs because is_branch_taken does not distinguish 32 bit JMP32 pointer comparisons from full width pointer comparisons. A valid pointer can have low 32 bits equal to zero, but the verifier can incorrectly predict the branch outcome and validate pointer arithmetic using a register value that differs from the value used at runtime. This can permit an unsafe kernel memory access outside the verifier approved pointer range. The demonstrated pattern performs an invalid kernel memory read, and the same path validation weakness can plausibly be used with memory write operations. Impact may include kernel crash, kernel memory disclosure, memory corruption, and local privilege escalation. For the CVSS the PR:L value is used for the paranoid score because BPF access can be available to an unprivileged user through enabled unprivileged BPF or delegated BPF capabilities. The issue is local and is not directly network reachable. YES REQUIRES MANUAL CHECK';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 5)	YES	BPF SIMPLEFIX LINUS  KPANIC INCREASED_FROM_LOW_BASED_ON_REQUIREMANUALCHECK	NO	NO	checked

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

ActionableScore=5
PrimitiveClass=CLASS_A
RealUAF=NO
HostAdminRequired=UNKNOWN
NamespaceScopedPrivilege=UNKNOWN
ImportantCandidate=YES
ManualReview=REQUIRED
AutoDowngradeAllowed=NO

1. ActionableScore

* Raw conservative score: 4
* Conservative score: 5
* Raw paranoid score: 5
* Paranoid score: 5
* Applied conservative floor: 4 -> 5 because a verifier bypass can permit an attacker-controlled unsafe kernel pointer access and must not be automatically closed as an ordinary crash-only issue.
* Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**::

2. Signal breakdown

* +2 Strong corruption primitive. The verifier can prove that the register used as a pointer offset is zero on the predicted branch while runtime execution reaches the other branch with an attacker-selected nonzero value. Pointer arithmetic is then accepted under an invalid verifier state. The demonstrated load establishes an unsafe pointer-access primitive. An equivalent memory-store instruction would use the same verifier-approved pointer state, so an attacker-influenced out-of-bounds kernel write is a technically supported concern.
* +1 Integrity plausible. A verifier-approved store through the incorrectly validated pointer can modify kernel memory outside the intended map-value bounds. Arbitrary target selection and a complete privilege-escalation chain are not established.
* +1 Reliable kernel crash or strong DoS. The supplied program performs a dereference through a pointer modified by a large invalid runtime offset and can fault in kernel context.
* +1 Paranoid confidentiality plausible. The supplied program demonstrates an invalid kernel load. BPF programs can normally transfer scalar results through permitted BPF data paths, but a complete read-back construction is not included in the patch. This point is used only in the paranoid interpretation.
* No Local unprivileged trigger bonus. The supplied evidence does not establish that an ordinary user can load and execute the complete BPF program in the evaluated product.
* No Local unprivileged high-control BPF API bonus. BPF is a high-control API, but ordinary unprivileged reachability is not established.
* No remote-reachability bonus. This is a local BPF verifier and program-loading path.
* No Strong LPE plausibility bonus. The unsafe pointer primitive is already counted as the strong corruption primitive. No controlled credential overwrite, controlled object replacement, callback control, or demonstrated local privilege escalation is supplied.
* No Practical exploit-chain bonus. The patch establishes the verifier bypass and unsafe access mechanism, but not a complete chain to disclosure, arbitrary kernel read or write, or privilege escalation.
* No privilege penalty. Host-admin-equivalent privilege is not positively established. Conversely, below-host-admin reachability is also not positively established.
* No race penalty. This is deterministic verifier path analysis, not a timing race.

3. Reachability analysis

The attacker model is a local process that can load and execute an eBPF program accepted by the kernel verifier. The supplied program requires a map-value pointer whose lower 32 bits are zero at runtime and relies on the verifier incorrectly treating a JMP32 comparison as a full-width pointer-versus-zero comparison.

On many current distributions, BPF program loading is restricted by CAP_BPF, CAP_SYS_ADMIN, unprivileged BPF policy, LSM policy, or other privileged configuration. However, the supplied evidence does not establish that the full load, map creation, execution, and result-observation sequence requires host-admin-equivalent privilege in every affected deployment. Therefore HostAdminRequired is UNKNOWN.

Namespace-scoped reachability is also unresolved. The patch does not establish that an attacker-created user or network namespace can perform the complete BPF load and execution sequence. NamespaceScopedPrivilege is UNKNOWN. No network, hardware, firmware, or race prerequisite is shown.

4. Primitive and severity interpretation

The exact defect is an unsound eBPF verifier branch prediction. For a JMP32 comparison such as `if w1 != 0`, the verifier incorrectly applies full-width pointer reasoning and concludes that the branch is always taken because a valid pointer is non-NULL. A valid kernel pointer may nevertheless have zero in its low 32 bits. Runtime control flow can therefore differ from the verifier-approved path.

The verifier may validate `r1 += r6` while treating `r6` as zero, whereas runtime execution retains an attacker-selected nonzero `r6`. The subsequent dereference then accesses memory outside the verified map-value range. This is not a UAF and does not involve fixed stale storage or cleanup behavior.

The demonstrated patch context directly supports an invalid kernel read and kernel crash. A store through the same incorrectly validated pointer state is a strong technically connected corruption concern, making this CLASS_A rather than crash-only CLASS_C. Arbitrary kernel read or write, controlled target placement, information disclosure, local privilege escalation, and root compromise are not directly demonstrated.

ImportantCandidate is YES because the verifier is the primary kernel security boundary preventing BPF programs from constructing unsafe kernel pointers. AutoDowngradeAllowed is NO because closing this as a normal BPF crash would discard a concrete verifier-bypass and memory-corruption concern that needs manual validation against product BPF policy and reachable program types.

Likely operational severity is Actionable Moderate at minimum. It becomes a stronger Important candidate where unprivileged BPF is enabled, BPF capabilities are delegated to less-trusted workloads, or the affected verifier path can be reached by a tenant or service account.

5. One-sentence report phrase

A BPF verifier flaw can mispredict JMP32 pointer comparisons and approve pointer arithmetic with a runtime attacker-controlled offset, potentially causing kernel memory corruption or a kernel crash.

6. Manual review recommendation

MANUAL CHECK REQUIRED

The patch establishes a verifier security-boundary bypass with an attacker-influenced unsafe kernel pointer-access mechanism, but product-specific BPF privilege policy and the practical read or write primitive require confirmation.

7. Call-site confidence

Call-site confidence: medium

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

Patch: bpf: Don't predict JMP32 pointer vs zero comparisons [ Upstream
Commit: d16f0eafcaa62e677a2b5b0dabc1f8d31dfe3d4d
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=d16f0eafcaa62e677a2b5b0dabc1f8d31dfe3d4d

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=d16f0eafcaa62e677a2b5b0dabc1f8d31dfe3d4d

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

The original Linux kernel CVE announcement for CVE-2026-98041 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98041

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:H/I:H/A:H
  The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
  The Best / paranoid CVSS score: 7.8

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: 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).