[CVE-2026-63865][MODERATE REGULAR] bpf: Drop task_to_inode and inet_conn_established from lsm sleepable hooks [ Upstream

[AL-KERNEL <alkernel@kernelcve.org>]

CVE: CVE-2026-63865
Priority: MODERATE REGULAR
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: bpf: Drop task_to_inode and inet_conn_established from lsm sleepable hooks [ Upstream
Commit: 452a927cddcd67478d030e646f41cb904a93156f
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=452a927cddcd67478d030e646f41cb904a93156f
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-63865
Analysis date: Sun, 19 Jul 2026 11:20:52 -0400
ActionableScore: 3
ActionableScore lower bound: 0
Actionable bucket: Borderline Moderate. Manual review recommended, but not an Important candidate
Manual review required: YES

Summary:
A sleepable BPF LSM program could be attached to hooks that run under RCU or softirq context, allowing a privileged or delegated local BPF user to trigger sleeping-in-atomic failures and potential kernel DoS, with no evidence of memory corruption or privilege escalation.

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

The original Linux kernel CVE announcement for CVE-2026-63865 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-63865

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-2026-63865	MODERATE	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:N/I:N/A:H';CWE-667;CWE-703;**CWE-400;Other CVSS 'AV:L/AC:L/PR:H/UI:N/S:U/C:N/I:N/A:L';BEST CVSS score: '5.5';DESCR 'A BPF LSM sleepable program could be attached to hooks that are actually invoked from non sleepable contexts. bpf_lsm_task_to_inode() runs under rcu_read_lock() and bpf_lsm_inet_conn_established() runs from softirq context, so helpers or kfuncs that may sleep can execute where sleeping is not allowed. This can trigger sleeping in atomic context warnings, scheduling errors, or a kernel crash depending on configuration such as panic_on_warn. For the CVSS the PR:L is used for the paranoid score because some deployments may delegate BPF loading or run reduced capability service accounts that can attach BPF LSM programs, while typical systems require administrative capabilities. The issue is not directly network reachable because an attacker must first be able to load or attach the BPF LSM program locally, although the inet_conn_established hook can then be reached by network connection events. Impact is denial of service only and no concrete information leak or write primitive is evident from the affected code path.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline Moderate. Manual review recommended, but not an Important candidate (with actual score 2)  SKIP CVE-2026-63865 UNKNOWN SKIP No affected files built, so skip this CVE NO  - - unknown 	MAYBE	BPF LINUS  INCREASED_FROM_LOW_BASED_ON_REQUIREMANUALCHECK DECREASED_TO_MODERATEREG_BASED_ON_ACTIONABLESCORELESSTHAN5	-	-	checked

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

ActionableScore=3
ActionableScoreLower=0

## 1. ActionableScore

* Conservative score: 0
* Paranoid score: 3
* Final recommended bucket: **Borderline Moderate. Manual review recommended, but not an Important candidate**::

## 2. Signal breakdown

Conservative signals:

* Reliable kernel crash / strong DoS: +1. Sleepable BPF LSM programs can call helpers or kfuncs that may sleep, but the affected hooks run under `rcu_read_lock()` or softirq context.
* Broad/common subsystem exposure: +1. BPF and LSM are broadly present kernel facilities, although this specific sleepable BPF LSM path is not broadly reachable by ordinary users.
* Requires admin/root/CAP_* in typical deployments: -2. Loading and attaching BPF LSM programs normally requires privileged BPF capabilities or root-controlled configuration.

Conservative total: `+1 +1 -2 = 0`

Paranoid signals:

* Local delegated trigger: +1. Some environments may delegate BPF loading to a reduced-capability service account or container/root-like management process.
* Reliable kernel crash / strong DoS: +1. Sleeping in atomic context can lead to warnings, scheduling errors, or crash-like behavior depending on kernel configuration such as `panic_on_warn`.
* Broad/common subsystem exposure: +1. The affected mechanism is in BPF LSM hook registration, a security-sensitive and widely deployed kernel subsystem.

Paranoid total: `+1 +1 +1 = 3`

Not awarded:

* No remote/network-triggerable score. `inet_conn_established` may be reached by network events, but the attacker must first attach a local BPF LSM program.
* No memory corruption score. The patch shows invalid sleepable-hook classification, not UAF, OOB write, double free, refcount misuse, or type confusion.
* No LPE score. There is no concrete reclaim, overwrite, callback control, arbitrary write, or security-boundary bypass primitive.
* No high-control BPF API bonus. BPF privilege gating applies because unprivileged BPF is typically disabled and BPF LSM attach requires privileged capabilities.

## 3. Reachability analysis

The bug is triggerable by a local actor able to load and attach a sleepable BPF LSM program to the affected hooks. In typical product configurations this requires root, `CAP_BPF`, `CAP_SYS_ADMIN`, or equivalent privileged BPF setup, so the conservative model treats it as privileged local only.

Namespaces or containers only change the score if BPF loading is delegated to a reduced-capability service account or container-like root that is not equivalent to full host root. In that case the paranoid score is more appropriate.

The issue is not directly network reachable. Network activity can invoke `bpf_lsm_inet_conn_established()`, but only after a local privileged or delegated actor has installed the problematic BPF LSM program.

Call-site confidence: high. The patch directly removes `bpf_lsm_task_to_inode` and `bpf_lsm_inet_conn_established` from the sleepable LSM hook list, and the commit states the exact invalid contexts.

## 4. Severity interpretation

This behaves like an ordinary Moderate or Low-like privileged DoS in the conservative model. The realistic impact is invalid sleeping from non-sleepable kernel contexts, which can cause warnings or crashes depending on configuration.

The paranoid score reaches borderline manual-review territory because BPF LSM context bugs are security-sensitive and some deployments may delegate BPF capabilities to less-than-full-root actors. However, the patch does not support treating this as Important: there is no demonstrated memory corruption, privilege escalation, information leak, arbitrary write, or cross-boundary bypass.

## 5. One-sentence report phrase

A sleepable BPF LSM program could be attached to hooks that run under RCU or softirq context, allowing a privileged or delegated local BPF user to trigger sleeping-in-atomic failures and potential kernel DoS, with no evidence of memory corruption or privilege escalation.

## 6. Manual review recommendation

MANUAL CHECK RECOMMENDED

Reason: the conservative score is low because BPF LSM attachment is privileged in typical deployments, but manual review is still useful for product-specific BPF delegation, `panic_on_warn`, and container/service-account capability models.

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

Patch: bpf: Drop task_to_inode and inet_conn_established from lsm sleepable hooks [ Upstream
Commit: 452a927cddcd67478d030e646f41cb904a93156f
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=452a927cddcd67478d030e646f41cb904a93156f

Changed files:
  kernel/bpf/bpf_lsm.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=452a927cddcd67478d030e646f41cb904a93156f

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

The original Linux kernel CVE announcement for CVE-2026-63865 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-63865

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:N/I:N/A:L
  The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
  The Best / paranoid CVSS score: 5.5

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: 0
  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).