[CVE-2026-90358][MODERATE 7.0] bpf, x86: Fix trampoline stack size for 128-bit arguments [ Upstream

[CVE-2026-90358][MODERATE 7.0] bpf, x86: Fix trampoline stack size for 128-bit arguments [ Upstream

[AL-KERNEL]

CVE: CVE-2026-90358
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: bpf, x86: Fix trampoline stack size for 128-bit arguments [ Upstream
Commit: 4aed2346fa9b88d458eeb8c01123f3c07585a675
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=4aed2346fa9b88d458eeb8c01123f3c07585a675
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90358
Analysis date: Fri, 18 Sep 2026 06:20:00 -0400
ActionableScore: 6
ActionableScore lower bound: 5
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-2026-90358 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90358

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-2026-90358	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-90358 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-131;*CWE-787;CWE-121;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '7';DESCR 'An x86 BPF trampoline stack overwrite can occur when a 128-bit scalar argument such as __int128 reaches the trampoline with arg_size 16. The previous register save area calculation only added extra slots for struct arguments, so save_args can write past the allocated save area and corrupt adjacent kernel stack slots. For the CVSS the PR:L is used because a local user or service with BPF trampoline loading privileges can trigger the vulnerable path, while ordinary remote packet traffic cannot trigger it directly. The issue is not network reachable. Impact is at least local denial of service via kernel crash and in the paranoid case may include confidentiality and integrity impact because this is a stack memory corruption primitive in generated kernel trampoline code.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 6)	YES	INIT BPF SIMPLEFIX HARDWARE LINUS  KPANIC INCREASED_FROM_LOW_BASED_ON_REQUIREMANUALCHECK KPANIC INCREASED_TO_HIGH_BASED_ON_ACTIONABLESCOREHIGHERTHAN5 DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFKP	YES	NO	checked

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

ActionableScore=6
ActionableScoreLower=5
PrimitiveClass=CLASS_A
RealUAF=NO
HostAdminRequired=UNKNOWN
NamespaceScopedPrivilege=UNKNOWN
ImportantCandidate=YES
ManualReview=REQUIRED
AutoDowngradeAllowed=NO

1. ActionableScore

* Raw conservative score: 5
* Conservative score: 5
* Raw paranoid score: 6
* Paranoid score: 6
* Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**::

No floor or cap changed the conservative score. No floor or cap changed the paranoid score. The paranoid score is emitted as ActionableScore because it is higher than the conservative score.

2. Signal breakdown

Conservative positive signals:

* +2 Strong corruption primitive: the patch directly states that the x86 BPF trampoline register save area is under-allocated and save_args() overwrites adjacent kernel stack slots.
* +2 Strong exploitation leverage: this is not merely an invalid read or WARN. It is a kernel stack overwrite in generated trampoline code, affecting adjacent active stack state.
* +1 Reliable kernel crash / strong DoS: corrupting adjacent kernel stack slots can realistically crash the kernel even if a full exploit chain is not shown.
* +1 Memory layout invariant restoration: the fix restores consistency between arg_size, register count, and stack save-area allocation.

Conservative negative signals:

* -1 Privilege-gated BPF control-plane reachability: triggering requires BPF trampoline loading or attachment privileges. The supplied evidence does not prove ordinary unprivileged reachability, but it also does not positively establish host-admin-equivalent privilege.

Paranoid additional signal:

* +1 Unresolved active stack corruption leverage: overwritten data is derived from saved argument register state and the target is adjacent trampoline stack storage, so meaningful integrity impact beyond crash remains technically plausible and requires manual review.

Important non-awarded signals:

* RealUAF not awarded: the bug is an under-allocation / stack overwrite, not a use-after-free.
* Controlled reclaim not demonstrated.
* Arbitrary kernel read/write not demonstrated.
* Remote reachability not established.
* Local ordinary unprivileged trigger not established.
* Namespace-scoped complete trigger not established.
* Confidentiality impact not separately awarded because no disclosure path is shown.
* Integrity impact is not separately counted as an extra point in the conservative score because the stack overwrite is already counted as strong corruption and exploitation leverage.
* Extra availability point not awarded because the crash/DoS consequence is already counted once.
* BPF object complexity is not used as a standalone scoring reason.

3. Reachability analysis

The reachable interface is the BPF trampoline generation path on x86. A trigger requires a function prototype containing a 128-bit scalar argument such as __int128 so that btf_distill_func_proto() produces arg_size == 16 and the trampoline register accounting is wrong.

The attacker is best modeled as a local user or local service able to load or attach BPF trampoline programs. Modern Linux systems commonly gate this behind CAP_BPF, CAP_PERFMON, CAP_SYS_ADMIN, or related privileged BPF policy. The supplied evidence does not prove that a normal unprivileged user can complete the full load, JIT, attach, and trigger sequence.

HostAdminRequired is UNKNOWN because host-admin-equivalent privilege is not positively established from the supplied patch, but below-host-admin reachability is also not fully established. NamespaceScopedPrivilege is UNKNOWN because the evidence does not show that all BPF load, trampoline attach, and trigger operations are available inside attacker-created namespaces.

No network trigger is established. Ordinary remote packet traffic cannot directly reach this path. No special physical hardware requirement is established beyond x86 and enabled BPF trampoline/JIT functionality. This is not a timing race.

4. Primitive and severity interpretation

The exact primitive is a kernel stack overwrite caused by under-allocation of the BPF trampoline register save area. The old code only added extra register slots for arguments marked as struct arguments, but a 128-bit scalar can also require two register slots. As a result, save_args() may write beyond the allocated save area and corrupt adjacent stack slots.

This is active kernel stack corruption, not fixed stale storage and not cleanup-only behavior. There is no concrete repeated release/refcount operation. There is no first lifetime operation that may free an object, and no second lifetime mutation through a dangling reference.

PrimitiveClass is CLASS_A because a concrete kernel overwrite primitive is directly established. The demonstrated minimum consequence is local kernel crash / DoS. The unresolved worst case is corruption of active stack state in generated kernel trampoline code, which may be security-relevant even though controlled code execution, arbitrary write, or privilege escalation is not demonstrated.

ImportantCandidate is YES because the primitive is direct kernel stack memory corruption and the exploitation leverage cannot be safely dismissed from the patch alone. AutoDowngradeAllowed is NO because automatically treating this as ordinary Moderate would discard a concrete kernel corruption primitive with unresolved impact.

Likely operational severity is actionable Moderate at minimum, with Strong Important candidacy requiring manual kernel/BPF exploitation review.

5. One-sentence report phrase

An x86 BPF trampoline stack overwrite can occur for 128-bit scalar arguments because the save area is under-allocated, allowing save_args() to corrupt adjacent kernel stack slots and potentially causing at least a local kernel crash with unresolved integrity impact.

6. Manual review recommendation

MANUAL CHECK REQUIRED

Manual review is required because the patch establishes a concrete kernel stack overwrite in generated BPF trampoline code. The issue is not proven to be ordinary unprivileged or remotely reachable, but the corruption primitive is strong enough that it should not be automatically downgraded or closed as crash-only.

7. Call-site confidence

Call-site confidence: medium

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

Patch: bpf, x86: Fix trampoline stack size for 128-bit arguments [ Upstream
Commit: 4aed2346fa9b88d458eeb8c01123f3c07585a675
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=4aed2346fa9b88d458eeb8c01123f3c07585a675

Changed files:
  arch/x86/net/bpf_jit_comp.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=4aed2346fa9b88d458eeb8c01123f3c07585a675

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

The original Linux kernel CVE announcement for CVE-2026-90358 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90358

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: 5
  Paranoid ActionableScore: 6

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).