[CVE-2026-53143][IMPORTANT] drm/amdkfd: Fix buffer overflow in SDMA queue checkpoint/restore on GFX11

[CVE-2026-53143][IMPORTANT] drm/amdkfd: Fix buffer overflow in SDMA queue checkpoint/restore on GFX11

[AL-KERNEL]

CVE: CVE-2026-53143
Priority: IMPORTANT
AL-KERNEL base severity: IMPORTANT
KPANIC flag: YES
Patch: drm/amdkfd: Fix buffer overflow in SDMA queue checkpoint/restore on GFX11
Commit: 16dad1fb0d783a4008de30e32d0038c393de05b1
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=16dad1fb0d783a4008de30e32d0038c393de05b1
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-53143
Analysis date: Thu, 25 Jun 2026 17:42:11 -0400
ActionableScore: 8
ActionableScore lower bound: 7
Actionable bucket: Strong Important candidate / Actionable Moderate at minimum
Manual review required: YES

Summary:
A GFX11 AMD KFD SDMA queue checkpoint/restore bug copies compute MQD sized data for smaller SDMA MQDs, leaking adjacent GTT memory during checkpoint and corrupting adjacent GTT memory during restore, with local PR:L reachability through GPU device access and plausible DoS or privilege escalation impact.

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

The original Linux kernel CVE announcement for CVE-2026-53143 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-53143

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: IMPORTANT
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-2026-53143	IMPORTANT	CHECK	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H';*CWE-787;CWE-125;CWE-120;BEST CVSS score: '7.8';DESCR 'A buffer overflow in the AMD KFD GFX11 SDMA MQD checkpoint and restore path occurs because SDMA queues were assigned compute MQD handlers that copy 2048 bytes while the SDMA MQD object is only 512 bytes. During checkpoint, a local user with KFD queue access can read 1536 bytes past the SDMA MQD buffer and copy adjacent GTT memory to userspace. During restore, the same size mismatch can write 1536 bytes past the SDMA MQD buffer and corrupt adjacent GTT memory such as ring buffers or neighboring MQDs. For the CVSS the PR:L is used because triggering requires local code execution and access to the AMD KFD or render device, but not necessarily full administrator privileges. The issue is not network reachable. Impact includes information disclosure, memory corruption, denial of service, and a plausible privilege escalation path due to attacker influenced out of bounds writes in GPU queue metadata.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Strong Important candidate / Actionable Moderate at minimum (with actual score 8)  SKIP CVE-2026-53143  SKIP No affected files built, so skip this CVE NO  - - unknown 	MAYBE	OOB LEAK HARDWARE  KPANIC INCREASED_TO_MODERATE_FROM_LOW_BASED_ON_GUESSCVSS KPANIC INCREASED_TO_HIGH_BASED_ON_ACTIONABLESCOREHIGHEREQTHAN7	-	-	checked

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

ActionableScore=8
ActionableScoreLower=7

## 1. ActionableScore

* Conservative score: 7
* Paranoid score: 8
* Final recommended bucket: **Strong Important candidate / Actionable Moderate at minimum**::

## 2. Signal breakdown

* Local unprivileged trigger: +1
  A local user with access to AMD KFD or render device nodes can plausibly reach KFD queue checkpoint or restore paths.

* Memory corruption, Strong corruption primitive: +2
  Restore copies 2048 bytes into a 512 byte SDMA MQD object, causing a 1536 byte OOB write into adjacent GTT memory.

* Confidentiality impact plausible: +1
  Checkpoint copies 2048 bytes from a 512 byte SDMA MQD object and leaks 1536 bytes of adjacent GTT memory to userspace.

* Integrity impact plausible: +1
  Restore corrupts adjacent GTT memory, potentially including ring buffers or neighboring MQDs.

* Reliable kernel or device DoS concern: +1
  Corrupting queue metadata or ring related GTT memory can realistically crash or hang the GPU/KFD path, and possibly destabilize the host graphics/compute stack.

* Privilege escalation plausible: +1 conservative, +2 paranoid
  Conservative scoring treats this as a weak but real LPE concern because the corrupted target is adjacent GPU/KFD metadata rather than a demonstrated kernel heap object takeover. Paranoid scoring treats the attacker influenced OOB write into queue metadata as a plausible stronger escalation primitive.

* Privileged kernel/device-management memory corruption path: +1
  The corruption occurs in a trusted GPU device-management path involving KFD MQD state and GTT memory.

* Hardware/path specificity: -1
  The issue is specific to GFX11/Navi3x SDMA queues and checkpoint/restore behavior, not a generic always-on kernel path.

## 3. Reachability analysis

The bug is local, not network reachable. The attacker needs local code execution and access to AMD KFD/render device interfaces, which is commonly granted to users in GPU-enabled desktop, compute, or containerized GPU workloads, but is not universal.

Namespaces and containers can matter: if a container or reduced-privilege service is given GPU device access, this should not be treated as full host-root privilege. No CAP_SYS_ADMIN assumption is needed for the practical risk model if the relevant device nodes and KFD operations are exposed.

The path is hardware and workload dependent: it requires affected GFX11 AMDGPU/KFD support and SDMA queue checkpoint/restore activity. That reduces broad exposure, but the primitive is explicit OOB read plus OOB write once the path is reachable.

## 4. Severity interpretation

This behaves more like an Important-class kernel/device vulnerability candidate than an ordinary Moderate. The patch shows a concrete size mismatch causing both information disclosure and memory corruption, not merely a crash or warning.

Realistic exploitation evidence is not enough to claim guaranteed privilege escalation, because target placement in adjacent GTT memory and downstream effects need manual validation. However, the OOB write is large, attacker influenced through restore input, and affects trusted GPU queue metadata, so auto-closing as a routine Moderate would be unsafe.

## 5. One-sentence report phrase

A GFX11 AMD KFD SDMA queue checkpoint/restore bug copies compute MQD sized data for smaller SDMA MQDs, leaking adjacent GTT memory during checkpoint and corrupting adjacent GTT memory during restore, with local PR:L reachability through GPU device access and plausible DoS or privilege escalation impact.

## 6. Manual review recommendation

MANUAL CHECK REQUIRED

Explicit OOB read and OOB write are present, the write affects trusted GPU/KFD metadata, and the impact may extend beyond DoS into information disclosure or privilege escalation depending on adjacent object layout and restore controllability.

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

Patch: drm/amdkfd: Fix buffer overflow in SDMA queue checkpoint/restore on GFX11
Commit: 16dad1fb0d783a4008de30e32d0038c393de05b1
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=16dad1fb0d783a4008de30e32d0038c393de05b1

Commit description:

The v11 MQD manager incorrectly assigned the CP-compute variants of
checkpoint_mqd/restore_mqd for KFD_MQD_TYPE_SDMA queues. These functions
use sizeof(struct v11_compute_mqd) (2048 bytes) instead of sizeof(struct
v11_sdma_mqd) (512 bytes), causing a 1536-byte overflow.

During CRIU checkpoint of an SDMA queue on Navi3x:
- checkpoint_mqd() reads 2048 bytes from a 512-byte SDMA MQD buffer,
  leaking 1536 bytes of adjacent GTT memory to userspace

During CRIU restore:
- restore_mqd() writes 2048 bytes into a 512-byte SDMA MQD buffer,
  corrupting 1536 bytes of adjacent GTT memory (often the ring buffer
  or neighboring MQDs)

This is a copy-paste regression unique to v11. All other ASIC backends
(cik, vi, v9, v10, v12) correctly use the SDMA-specific variants.

Add checkpoint_mqd_sdma() and restore_mqd_sdma() functions that properly
handle the smaller v11_sdma_mqd structure, matching the pattern used in
other MQD managers.

Fixes: cc009e6 ("drm/amdkfd: Add KFD support for soc21 v3")
Assisted-by: Claude:Sonnet 4-5
Signed-off-by: Andrew Martin <andrew.martin@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 6fa41db )
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed files:
  drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.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=16dad1fb0d783a4008de30e32d0038c393de05b1

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

The original Linux kernel CVE announcement for CVE-2026-53143 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-53143

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: Not available
  The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
  The Best / paranoid CVSS score: 7.8

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: 7
  Paranoid ActionableScore: 8

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: IMPORTANT
KPANIC detected for this report: YES
Published priority for this report (same as in Subject): IMPORTANT

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).