[CVE-2026-93809][MODERATE REGULAR] drm/amdgpu: flush pending RCU callbacks on module unload [ Upstream

[CVE-2026-93809][MODERATE REGULAR] drm/amdgpu: flush pending RCU callbacks on module unload [ Upstream

[AL-KERNEL]

CVE: CVE-2026-93809
Priority: MODERATE REGULAR
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: drm/amdgpu: flush pending RCU callbacks on module unload [ Upstream
Commit: 5063ae84c45299a100bf1d6ed11945f688063649
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=5063ae84c45299a100bf1d6ed11945f688063649
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-93809
Analysis date: Thu, 24 Sep 2026 20:40:35 -0400
ActionableScore: 1
ActionableScore lower bound: unknown
Actionable bucket: Ordinary Moderate
Manual review required: NO

Summary:
No one-sentence report phrase was found.

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

The original Linux kernel CVE announcement for CVE-2026-93809 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-93809

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: NO

A detailed explanation of the methodology and priority rules is included
at the end of this message.

======================================================================
AL-KERNEL CLASSIFICATION RESULT
======================================================================

CVE-2026-93809	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-93809 CHECK WITH IMPACT FROM ORIG NN MODERATE	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:H/PR:H/UI:N/S:U/C:L/I:L/A:H';*CWE-416;*CWE-672;CWE-667;Other CVSS 'AV:L/AC:H/PR:H/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '5.2';DESCR 'The amdgpu module can be unloaded while callbacks previously queued through call_rcu() are still pending. After the module text is freed, rcu_do_batch may invoke a stale callback address and attempt to execute code from unmapped or reused module memory, resulting in an Oops or kernel panic. For the CVSS the PR:H value is used because reliable triggering normally requires CAP_SYS_MODULE or equivalent authority to unload the amdgpu module. AC:H is used because exploitation depends on a pending RCU callback surviving until the module text has been released. The issue is not directly network reachable. The demonstrated impact is local denial of service, while limited confidentiality or integrity impact is retained only in the paranoid score because reuse of freed module memory is theoretically possible but no controlled execution primitive is established.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Ordinary Moderate (with actual score 1)  SKIP CVE-2026-93809 UNKNOWN SKIP No affected files built, so skip this CVE NO  - - unknown 	MAYBE	DANGER INIT UAF KERNEL_PANIC_PLUS_UAF HARDWARE  DECREASED_TO_MODERATE_REGULAR_FROM_MODERATE7_BASED_ON_GUESSCVSS_AND_CIANNH	-	-	checked

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

ActionableScore=1
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=YES
NamespaceScopedPrivilege=NO
ImportantCandidate=NO
ManualReview=RECOMMENDED
AutoDowngradeAllowed=YES

1. ActionableScore

* Raw conservative score: 1
* Conservative score: 1
* Raw paranoid score: 1
* Paranoid score: 1
* No floor or cap was applied.
* Final recommended bucket: **Ordinary Moderate**::

2. Signal breakdown

* +2 Real UAF or lifetime violation. A queued RCU callback remains callable after its implementing module text has been freed.
* +1 Reliable kernel crash. The supplied trace demonstrates an instruction-fetch page fault from `rcu_do_batch()` at an unmapped module address.
* -2 Proven host-admin-equivalent privilege. Unloading a kernel module requires host-level `CAP_SYS_MODULE`, and the operation affects global module state.
* No race penalty. Although callback timing matters, the evidence does not establish that the ordering is unusually narrow or difficult to reproduce.
* No controlled-reclaim or callback-control bonus. The callback target is a previously registered fixed module address, not an attacker-selected function pointer.
* No structured stale-object leverage. The stale operation is execution at a retired code address during module teardown, with an immediate fault as the demonstrated result.
* No C/I or LPE signal. Reuse of the unloaded module address range is theoretical, and no attacker-controlled replacement, arbitrary write, disclosure, or useful control-flow primitive is established.
* No extra Availability point. It would double-count the same Oops or kernel panic.
* No remote-reachability signal.
* No rare-hardware penalty. AMD GPU hardware is not inherently rare, and the principal restriction is privileged module unload.
* First release capable of ending an object lifetime: not applicable as a repeated object release. Module text is retired by module unload.
* Exact reference surviving into a second release: no.
* Second lifetime or refcount mutation: no.
* Concrete repeated release/refcount +1: not awarded.
* Paranoid repeated-refcount +1: not awarded.
* FLOOR 5: not applied.

3. Reachability analysis

The complete trigger requires the amdgpu module to have an outstanding `call_rcu()` callback and then to be unloaded before that callback executes. Module unload is a global kernel-management operation requiring host-level `CAP_SYS_MODULE`. It is not available through an attacker-created user namespace.

An ordinary logged-in user, DRM client, compositor, DRM master, or delegated display service cannot normally unload the module merely through DRM or KMS access. A host administrator can unbind devices, unload the module, and arrange the relevant teardown sequence.

The system must use amdgpu as a loadable module and have a pending callback at unload. No direct network trigger is established. The timing is asynchronous, but the supplied evidence does not prove a genuinely hard or externally uncontrollable race.

4. Primitive and severity interpretation

The exact primitive is late invocation of a fixed RCU callback after the module text containing that callback has been freed or unmapped. This is a real post-lifetime use and therefore a constrained Class B lifetime violation.

The stale storage is effectively a fixed callback target into retired executable module memory. It is not an attacker-controlled active structured object. The demonstrated consequence is an invalid instruction fetch followed by an Oops or kernel panic.

There is no concrete repeated release or refcount mutation. There is also no evidence of controlled module-address reuse, attacker-shaped replacement code, arbitrary kernel read/write, disclosure, or privilege escalation. Reuse of the virtual address by another module is a theoretical consequence rather than a patch-supported exploitation primitive, and an actor able to load malicious kernel modules already has host-admin-equivalent authority.

ImportantCandidate is NO because the established path is a host-admin-triggered teardown crash without meaningful unresolved below-admin exploitation leverage. AutoDowngradeAllowed is YES because this is a constrained Class B case with fixed stale code, crash-only demonstrated behavior, no repeated lifetime mutation, and no security-boundary bypass.

5. One-sentence report phrase

A pending amdgpu RCU callback can execute after module text is freed during privileged module unload, causing a stale instruction fetch and kernel crash without an established exploitation primitive beyond denial of service.

6. Manual review recommendation

MANUAL CHECK RECOMMENDED

Confirm that every relevant amdgpu callback is module-owned and that no product-specific automatic unload path can be induced below host-admin privilege, but the supplied evidence supports a constrained privileged crash rather than an Important-grade primitive.

7. Call-site confidence

Call-site confidence: medium

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

Patch: drm/amdgpu: flush pending RCU callbacks on module unload [ Upstream
Commit: 5063ae84c45299a100bf1d6ed11945f688063649
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=5063ae84c45299a100bf1d6ed11945f688063649

Changed files:
  drivers/gpu/drm/amd/amdgpu/amdgpu_drv.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=5063ae84c45299a100bf1d6ed11945f688063649

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

The original Linux kernel CVE announcement for CVE-2026-93809 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-93809

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:N/I:N/A:H
  The Best / paranoid CVSS vector: AV:L/AC:H/PR:H/UI:N/S:U/C:L/I:L/A:H
  The Best / paranoid CVSS score: 5.2

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: 1

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).