[CVE-2026-90313][MODERATE 7.0] bpf, cgroup: Fix invalid storage access after __cgroup_bpf_attach failed [ Upstream

[CVE-2026-90313][MODERATE 7.0] bpf, cgroup: Fix invalid storage access after __cgroup_bpf_attach failed [ Upstream

[AL-KERNEL]

CVE: CVE-2026-90313
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: bpf, cgroup: Fix invalid storage access after __cgroup_bpf_attach failed [ Upstream
Commit: e26db0d636e4c24a6b16683ea95cf5077c64d74b
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=e26db0d636e4c24a6b16683ea95cf5077c64d74b
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90313
Analysis date: Fri, 18 Sep 2026 00:42:37 -0400
ActionableScore: 5
ActionableScore lower bound: unknown
Actionable bucket: Actionable Moderate / Strong Important candidate requiring manual review
Manual review required: YES

Summary:
A failed cgroup BPF `BPF_F_REPLACE` rollback can leave an active program with dangling local storage pointers, allowing later `bpf_get_local_storage()` execution to trigger invalid kernel memory access and requiring manual review for possible impact beyond DoS.

======================================================================
ABOUT THIS REPORT
======================================================================

The original Linux kernel CVE announcement for CVE-2026-90313 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90313

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-2026-90313	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-90313 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-672;CWE-703;Other CVSS 'AV:L/AC:H/PR:H/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '7';DESCR 'An invalid storage access in cgroup BPF can occur after a failed BPF_F_REPLACE operation in multi attach mode because pl storage is updated before attach completion and is not restored on the error path. The still active old program can keep a dangling cgroup storage pointer after the newly allocated storage has been freed, and a later bpf_get_local_storage call may dereference or use that stale storage. For the CVSS the PR:L is used for the paranoid score because reliable triggering requires local ability to attach or replace cgroup BPF programs, but this may be delegated to a service account or container root rather than full host root. The issue is not network reachable and must be triggered through local BPF and cgroup control plane operations. Impact is at least local denial of service via invalid memory access and in the worst case may allow confidentiality or integrity impact due to a use after free style dangling pointer in kernel memory.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 5)	YES	BPF IMPROVEONLY  KPANIC INCREASED_FROM_LOW_BASED_ON_REQUIREMANUALCHECK	NO	NO	checked

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

ActionableScore=5
PrimitiveClass=CLASS_B
RealUAF=YES
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
* Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**::

Applied conservative floor: 4 -> 5. Reason: real active-object lifetime violation with non-cleanup stale use through `bpf_get_local_storage()`, and host-admin-equivalent reachability is not positively established.

## 2. Signal breakdown

Positive signals:

* +2 Real UAF / lifetime violation. `pl->storage` is overwritten with newly allocated storage before attach completion. On failure, `new_storage` is freed, but `pl->storage` is not restored, so still-active `prog1` may keep dangling storage pointers.
* +1 Reliable kernel crash / strong DoS. The commit states that later execution of `prog1` and `bpf_get_local_storage()` can lead to invalid memory access.
* +1 Structured stale-object exploitation surface. The stale pointer is not only used in final cleanup. It remains attached to an active cgroup BPF program and is later consumed through a BPF helper during program execution.
* +1 Paranoid attacker-influenced lifetime manipulation. An actor able to perform cgroup BPF multi-attach replacement controls the replace sequence and later attachment/execution flow. The failed attach path is an ordered rollback bug, not merely a narrow timing race.

Non-awarded or withheld signals:

* Controlled reclaim not demonstrated.
* Arbitrary kernel read/write not demonstrated.
* Confidentiality and integrity are not awarded as independent ActionableScore points, because the supplied evidence establishes dangling storage access but not a concrete disclosure or unauthorized write primitive.
* Remote reachability not established.
* Concrete repeated release/refcount signal not awarded. The patch shows dangling storage use after free, but not a second `put` / `free` / refcount mutation on the exact same dangling reference.
* Weak LPE not separately awarded, because the meaningful concern is already represented by the active structured stale-object lifetime primitive.
* Extra availability not awarded, because the invalid access / crash is already counted once.
* BPF privileged-control-plane cap does not reduce below 5 here, because the stale use is active program execution rather than cleanup-only crash behavior.

Repeated-release checklist:

* First release can end lifetime: yes, `bpf_cgroup_storages_free(new_storage)` can free the newly allocated storage.
* Exact same reference survives: yes, `pl->storage` can still point to the freed storage.
* Second operation is a real lifetime/refcount mutation: no, later evidence is invalid access through `bpf_get_local_storage()`, not a concrete second lifetime mutation.
* Concrete repeated release/refcount +1 awarded: no.
* Paranoid repeated-refcount leverage +1 awarded: no.
* FLOOR 5 applied: yes, due to active non-cleanup stale use and unresolved exploitation leverage.

## 3. Reachability analysis

The trigger requires local ability to load and attach or replace cgroup BPF programs in multi-attach mode using `BPF_F_REPLACE`. This is a BPF and cgroup control-plane operation, not a network packet path.

For BPF/XDP privilege scope:

* BPF program loading privilege is required, but the exact evaluated product policy is not supplied.
* cgroup BPF attach or replace privilege is required.
* The prompt hints that this may be delegated to a service account or container root, but complete below-host-admin reachability is not proven.
* `HostAdminRequired=UNKNOWN` is therefore used rather than assuming either host root or ordinary user reachability.
* `NamespaceScopedPrivilege=UNKNOWN` is used because the supplied evidence does not prove that all required stages, including BPF load, attach, replace, failure path, and later execution, are fully namespace-scoped.
* No prior public network exposure is involved.
* No special hardware requirement is visible.
* This is not primarily a race. The bug is an error-path rollback/lifetime mismatch after a partial failure.

## 4. Primitive and severity interpretation

The exact primitive is a dangling cgroup BPF storage pointer after failed replacement rollback. The active old program remains installed, but its `pl->storage` may refer to storage that was allocated for the failed replacement and then freed.

This is an active structured object lifetime issue rather than fixed stale storage. The stale use occurs during later BPF program execution via `bpf_get_local_storage()`, not only during cleanup or destruction. That gives more unresolved exploitation surface than a simple cleanup-only NULL dereference.

However, the patch does not establish controlled reclaim, attacker-shaped replacement contents after free, arbitrary read/write, callback hijack, credential manipulation, or demonstrated LPE. Therefore this remains `PrimitiveClass=CLASS_B`, not `CLASS_A`.

Demonstrated consequence is local kernel invalid memory access and DoS. Speculative consequence is possible stronger impact if the dangling storage can be reused or interpreted in an attacker-useful way, but that is not proven by the supplied patch.

`ImportantCandidate=YES` because this is a real active-object UAF-style lifetime bug in BPF/cgroup state with unresolved exploitation leverage. `AutoDowngradeAllowed=NO` because automatically closing it as ordinary Moderate would discard a concrete dangling active storage primitive that deserves manual review.

Likely operational severity: actionable Moderate at minimum, with Important-class review warranted depending on actual BPF privilege policy and exploitability of the stale storage access.

## 5. One-sentence report phrase

A failed cgroup BPF `BPF_F_REPLACE` rollback can leave an active program with dangling local storage pointers, allowing later `bpf_get_local_storage()` execution to trigger invalid kernel memory access and requiring manual review for possible impact beyond DoS.

## 6. Manual review recommendation

MANUAL CHECK REQUIRED

Reason: real dangling active BPF storage is established, later non-cleanup use is documented, and exploitation leverage beyond crash is unresolved.

## 7. Call-site confidence

Call-site confidence: high

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

Patch: bpf, cgroup: Fix invalid storage access after __cgroup_bpf_attach failed [ Upstream
Commit: e26db0d636e4c24a6b16683ea95cf5077c64d74b
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=e26db0d636e4c24a6b16683ea95cf5077c64d74b

Changed files:
  kernel/bpf/cgroup.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=e26db0d636e4c24a6b16683ea95cf5077c64d74b

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

The original Linux kernel CVE announcement for CVE-2026-90313 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90313

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: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: 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).