[CVE-2026-80907][MODERATE 7.0] drm/amdgpu: Fix UVD dpb min size calculation for H264

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

CVE: CVE-2026-80907
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: drm/amdgpu: Fix UVD dpb min size calculation for H264
Commit: fa96c24485942e277483cc70d9551d9e0111d7c5
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=fa96c24485942e277483cc70d9551d9e0111d7c5
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-80907
Analysis date: Fri, 04 Sep 2026 18:33:42 -0400
ActionableScore: 5
ActionableScore lower bound: 4
Actionable bucket: Actionable Moderate
Manual review required: NO

Summary:
No one-sentence report phrase was found.

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

The original Linux kernel CVE announcement for CVE-2026-80907 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-80907

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: NO

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

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

CVE-2026-80907	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-80907 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:L/I:L/A:H';CWE-131;*CWE-20;*CWE-787;Other CVSS 'AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '6.6';DESCR 'A DPB size validation issue in the AMDGPU UVD H.264 decode path can occur because the driver used a level derived maximum instead of the actual reference count provided in the decode message. A local user with access to the DRM render device may submit a crafted UVD decode message where the effective number of reference pictures does not match the size validated by the kernel. For the CVSS the PR:L is used because triggering requires local code execution with access to the GPU render or video decode interface, not full administrative privileges. The issue is not network reachable and is triggered through local GPU command submission rather than packet traffic. Impact is at least denial of service through GPU fault, reset, or decoder failure. In the worst still defensible case, an undersized DPB may allow limited confidentiality or integrity impact inside GPU accessible memory, so this should be manually reviewed.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate (with actual score 5)  SKIP CVE-2026-80907 UNKNOWN SKIP No affected files built, so skip this CVE NO  - - unknown 	MAYBE	SIMPLEFIX HARDWARE  KPANIC INCREASED_FROM_LOW_BASED_ON_REQUIREMANUALCHECK	-	-	checked

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

ActionableScore=5
ActionableScoreLower=4
PrimitiveClass=CLASS_B
RealUAF=NO
HostAdminRequired=NO
NamespaceScopedPrivilege=NO
ImportantCandidate=NO
ManualReview=RECOMMENDED
AutoDowngradeAllowed=NO

1. ActionableScore

* Raw conservative score: 4
* Conservative score: 4
* Raw paranoid score: 5
* Paranoid score: 5
* Applied conservative floor/cap: none
* Applied paranoid floor/cap: none
* Final recommended bucket: **Actionable Moderate**::

2. Signal breakdown

Conservative signals:

* +1 Local below-host-admin trigger. The vulnerable path is a DRM AMDGPU UVD video decode command parser path, normally reachable by a local user or graphics process with access to the DRM render or video device, not by host-admin-only code.
* +1 Memory layout invariant restoration. The patch changes DPB minimum size validation from a level-derived estimate to the actual H264 reference count from the decode message.
* +1 Weak or constrained corruption candidate. A too-small DPB may be accepted for a user-controlled decode message and then consumed by GPU firmware or hardware, creating a plausible constrained GPU-accessible out-of-bounds memory access concern.
* +1 Availability impact. The most directly supported practical impact is GPU decode failure, GPU fault, reset, or local denial of service.

Paranoid additional signal:

* +1 Local high-control kernel data-plane API. The attacker can influence decode message fields such as the reference count and buffer sizing through command submission, which are directly relevant to the validation defect.

Non-awarded signals:

* RealUAF not awarded. No use-after-free or stale lifetime operation is established.
* Concrete repeated release/refcount not awarded. No first release, surviving dangling reference, or second lifetime mutation exists.
* Strong exploitation leverage not awarded. Controlled reclaim, directed kernel write, callback control, refcount takeover, or credential impact is not established.
* Confidentiality and integrity points not awarded as direct signals. Limited C/I is plausible only as a paranoid concern inside GPU-accessible memory, not as a demonstrated protected kernel memory disclosure or overwrite.
* Reliable kernel crash not awarded. The patch does not establish host-wide panic as the primary outcome.
* Remote reachability not awarded. The trigger is local GPU command submission, not packet traffic.
* Hardware rarity penalty not applied. AMDGPU/UVD is hardware-specific but not rare or physical-only in the scoring sense.
* Host-admin penalty not applied. The supplied path does not positively establish a host-admin-equivalent requirement.

3. Reachability analysis

The evaluated attacker is a local user or local process with access to the AMDGPU DRM render or video decode interface. This is below host-admin in common DRM render-node deployments, although product policy may restrict device access to a graphics or video group.

The issue is not network reachable. It is triggered through local GPU command submission. Namespace-scoped privilege is not relevant to the complete trigger chain.

Prior host setup is limited to the presence of an AMDGPU device and enabled UVD decode support. No special boot option, rare hardware mode, or host administrator action beyond normal device availability is established. No timing race is involved.

For DRM role separation, the likely actors are an ordinary logged-in user with render-node access, a media process, compositor-adjacent graphics service, or another delegated graphics user. Full host administrator privileges are not positively required by the patch context.

4. Primitive and severity interpretation

The exact primitive is an incorrect DPB size validation calculation in the AMDGPU UVD H264 and H264 Perf decode paths. The old code derived the number of DPB buffers from the H264 level and frame size, while the fixed code uses the actual reference count from msg[61] and rejects values above 17.

This is not a lifetime bug and not a UAF. It is a validation and size-invariant bug affecting a buffer size check before GPU firmware or hardware consumes the decode parameters.

PrimitiveClass is CLASS_B because the patch supports a plausible constrained GPU-accessible memory access issue beyond a purely harmless logic bug, but it does not establish a strong kernel corruption primitive. It is not CLASS_A because there is no demonstrated arbitrary write, controlled object replacement, protected file corruption, callback hijack, credential modification, or privilege escalation.

The demonstrated practical consequence is local denial of service through GPU decoder failure, GPU fault, or reset. The speculative consequence is limited confidentiality or integrity impact inside GPU-accessible memory if an undersized DPB causes out-of-bounds access, but this remains unresolved and is not enough to classify it as Important.

ImportantCandidate is NO because the primitive is constrained and does not show strong exploitation leverage. AutoDowngradeAllowed is NO because the issue is still an actionable user-controlled GPU command validation defect with unresolved memory-access implications, so automatic closure as ordinary low-risk DoS would be unsafe.

Likely operational severity is Actionable Moderate.

5. One-sentence report phrase

A local user with access to the AMDGPU UVD decode interface may submit an H264 decode message whose actual reference count is not correctly reflected in the DPB size validation, potentially causing GPU fault or constrained GPU memory access and requiring manual security review.

6. Manual review recommendation

MANUAL CHECK RECOMMENDED

Manual review is recommended because the bug affects user-controlled GPU command validation and buffer sizing. The most likely impact is DoS, but the undersized DPB condition creates enough constrained memory-access uncertainty to avoid automatic downgrade.

7. Call-site confidence

Call-site confidence: medium

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

Patch: drm/amdgpu: Fix UVD dpb min size calculation for H264
Commit: fa96c24485942e277483cc70d9551d9e0111d7c5
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=fa96c24485942e277483cc70d9551d9e0111d7c5

Commit description:

This should use actual number of references from the decode
message, instead of maximum derived from level.

Signed-off-by: David Rosca <david.rosca@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Reviewed-by: Ruijing Dong <ruijing.dong@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 64b525e )
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed files:
  drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.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=fa96c24485942e277483cc70d9551d9e0111d7c5

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

The original Linux kernel CVE announcement for CVE-2026-80907 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-80907

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:L/UI:N/S:U/C:N/I:N/A:H
  The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:H
  The Best / paranoid CVSS score: 6.6

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: 4
  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).