* [CVE-2026-90286][MODERATE REGULAR] drm/amdgpu/gfx6: Use PFP on the compute queues too [ Upstream
@ 2026-09-18 13:20 21% AL-KERNEL
0 siblings, 0 replies; 1+ results
From: AL-KERNEL @ 2026-09-18 13:20 UTC (permalink / raw)
To: kernel-cve
CVE: CVE-2026-90286
Priority: MODERATE REGULAR
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: drm/amdgpu/gfx6: Use PFP on the compute queues too [ Upstream
Commit: f37211b9c01433f0bbb4709d25df7a0257cf915b
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=f37211b9c01433f0bbb4709d25df7a0257cf915b
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90286
Analysis date: Fri, 18 Sep 2026 09:20:51 -0400
ActionableScore: 2
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-2026-90286 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90286
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-2026-90286 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-90286 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:L/I:L/A:H';CWE-662;CWE-362;CWE-691;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '5.8';DESCR 'An execution ordering flaw affects AMD GFX6 compute queues because they use the same CP path and PFP engine as the graphics queue, while the affected code treated PFP synchronization as graphics only. The PFP can begin processing a later submission while the ME is still completing the previous submission, and a compute queue can continue after a VM flush without the required PFP to ME synchronization. This can cause a GPU hang, timeout, reset, or incorrect use of stale GPU VM translations. For the CVSS the PR:L is selected because a local user or process with access to the DRM render node can normally submit compute workloads without full administrative privileges. AC:H is selected because exploitation depends on GFX6 hardware and a timing sensitive sequence involving compute submissions and VM flush operations. The issue is not directly network reachable. The established impact is local denial of service, while limited confidentiality or integrity impact is conservatively possible if stale GPU mappings weaken memory isolation.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Ordinary Moderate (with actual score 2) SKIP CVE-2026-90286 UNKNOWN SKIP No affected files built, so skip this CVE NO - - unknown MAYBE INIT HARDWARE PACKET INCREASED_FROM_LOW_BASED_ON_REQUIREMANUALCHECK DECREASED_TO_MODERATEREG_BASED_ON_ACTIONABLESCORELESSTHAN4 - - checked
======================================================================
ACTIONABLESCORE ANALYSIS
======================================================================
ActionableScore=2
PrimitiveClass=CLASS_C
RealUAF=NO
HostAdminRequired=NO
NamespaceScopedPrivilege=NO
ImportantCandidate=NO
ManualReview=RECOMMENDED
AutoDowngradeAllowed=YES
1. ActionableScore
* Raw conservative score: 2
* Conservative score: 2
* Raw paranoid score: 2
* Paranoid score: 2
* No floor or cap was applied.
* Final recommended bucket: **Ordinary Moderate**::
2. Signal breakdown
Positive signals:
* +1 Local unprivileged trigger. A user with access to the DRM render node can submit GFX6 compute workloads without host-administrator privileges.
* +1 Active translation synchronization. The fix restores PFP and ME ordering after a GPU VM TLB flush, showing that the old translation state could remain operationally relevant while a later submission began execution.
Negative signals:
* No negative points. GFX6-specific hardware is required, but hardware specificity alone does not justify a rare-hardware penalty.
* No race penalty. The defect is an execution-ordering failure, but the evidence does not establish a narrow or uncontrollable one-shot timing window.
Material non-awarded signals:
* No real UAF, OOB access, controlled reclaim, arbitrary write, or CPU kernel-memory corruption is established.
* Confidentiality and integrity points are not awarded. Stale GPU translation use is technically conceivable, but no protected-memory read or unauthorized write path is demonstrated.
* Strong DoS is not awarded. A GPU hang, timeout, or reset is plausible, but a reliable host kernel crash or host-wide failure is not established.
* Active semantic-domain transition is not awarded. The patch corrects ordering within the same GPU VM domain rather than reinterpreting a live mapping as a different mapping type.
* No secondary index or reverse-mapping staleness is shown.
* No additional availability point is awarded for the same possible GPU disruption.
3. Reachability analysis
The evaluated attacker is an ordinary local process with authorized access to the AMDGPU render node. Such a process can submit compute command streams and cause normal GPU VM flush operations without DRM master, compositor, KMS-control, or host-administrator privileges.
The path is not network reachable. It is controlled through device-file access rather than a namespace-scoped capability, so user or network namespace creation does not independently provide access. No separate privileged configuration is indicated beyond an affected GFX6 GPU, the AMDGPU driver, and render-node exposure. The defect applies specifically to GFX6 compute queues. The supplied evidence does not establish that exploitation requires a difficult or one-shot race, and submissions can normally be repeated.
4. Primitive and severity interpretation
The exact primitive is incorrect hardware command ordering between the PFP and ME engines. Compute queues used ME-oriented register writes and synchronization even though GFX6 compute and graphics rings share the PFP-based CP path. The missing PFP synchronization after a VM flush could let a later submission begin before the preceding work and translation flush were fully ordered.
This is neither active structured UAF nor fixed stale storage. There is no release, free, stale CPU object, or repeated refcount mutation. The behavior is therefore CLASS_C invalid-state and availability behavior.
The demonstrated mechanism supports incorrect GPU execution ordering. A GPU hang, timeout, reset, or stale translation use is plausible. Cross-process GPU-memory access, host-memory corruption, privilege escalation, and arbitrary read or write remain speculative and are not scored.
ImportantCandidate is NO because no independently useful corruption or security-boundary primitive is established. AutoDowngradeAllowed is YES because the evidence supports an ordinary synchronization defect with possible device-local disruption, while meaningful exploitation leverage beyond that disruption is unresolved but not concretely established.
5. One-sentence report phrase
An AMD GFX6 compute-queue synchronization defect can allow the PFP to start a later submission before prior ME work or a GPU VM flush is fully ordered, potentially causing incorrect GPU execution, timeout, reset, or device-local denial of service.
6. Manual review recommendation
MANUAL CHECK RECOMMENDED
Review should confirm whether stale GPU VM translations can cross process or VMID isolation boundaries on GFX6 hardware, since such evidence would materially change the primitive and score.
7. Call-site confidence
Call-site confidence: medium
======================================================================
UPSTREAM PATCH SUMMARY
======================================================================
Patch: drm/amdgpu/gfx6: Use PFP on the compute queues too [ Upstream
Commit: f37211b9c01433f0bbb4709d25df7a0257cf915b
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=f37211b9c01433f0bbb4709d25df7a0257cf915b
Changed files:
drivers/gpu/drm/amd/amdgpu/gfx_v6_0.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=f37211b9c01433f0bbb4709d25df7a0257cf915b
======================================================================
DETAILED REPORT METHODOLOGY
======================================================================
The original Linux kernel CVE announcement for CVE-2026-90286 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90286
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:N/I:N/A:H
The Best / paranoid CVSS vector: AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:H
The Best / paranoid CVSS score: 5.8
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: 2
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 [email protected] (and both
send reply to CVE record itself too and see "reply" button below for howto reply).
^ permalink raw reply [relevance 21%]
Results 1-1 of 1 | reverse | sort options + mbox downloads above
-- links below jump to the message on this page --
2026-09-18 13:20 21% [CVE-2026-90286][MODERATE REGULAR] drm/amdgpu/gfx6: Use PFP on the compute queues too [ Upstream AL-KERNEL
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox