[CVE-2026-45853][MODERATE 7.0] drm/amdgpu: Use kvfree instead of kfree in amdgpu_gmc_get_nps_memranges() [ Upstream

[CVE-2026-45853][MODERATE 7.0] drm/amdgpu: Use kvfree instead of kfree in amdgpu_gmc_get_nps_memranges() [ Upstream

[AL-KERNEL]

CVE: CVE-2026-45853
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: drm/amdgpu: Use kvfree instead of kfree in amdgpu_gmc_get_nps_memranges() [ Upstream
Commit: 16e7e7ad8cdc6b4c4af7f31e262f1494c1b2a55e
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=16e7e7ad8cdc6b4c4af7f31e262f1494c1b2a55e
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-45853
Analysis date: Sat, 01 Aug 2026 11:47:44 -0400
ActionableScore: 7
ActionableScore lower bound: 4
Actionable bucket: Strong Important candidate / Actionable Moderate at minimum
Manual review required: YES

Summary:
A mismatched free in the AMDGPU NPS discovery path can release vmalloc-backed memory allocated by kvcalloc() through kfree(), leading to kernel memory corruption or crash, with realistic exploitation depending on AMDGPU hardware, discovery data, and local ability to trigger driver initialization or reset paths.

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

The original Linux kernel CVE announcement for CVE-2026-45853 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-45853

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: YES

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

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

CVE-2026-45853	MODERATE	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:L/UI:N/S:U/C:H/I:H/A:H';CWE-762;CWE-763;CWE-119;Other CVSS 'AV:L/AC:H/PR:H/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '7';DESCR 'A mismatched free in the amdgpu GMC discovery path can release memory allocated by kvcalloc with kfree instead of kvfree. Since kvcalloc may return vmalloc backed memory for large allocations, this can corrupt kernel memory when the NPS memory range data is released. The realistic trigger appears local and hardware dependent, likely involving AMDGPU initialization, reset, reprobe, or a large discovery table rather than packet based network input. For the CVSS the PR:L is used in the paranoid score because local users or reduced capability services may have access to GPU related interfaces in some deployments, although reliable triggering of this exact path may still require driver or hardware specific conditions. Impact is at least local denial of service via kernel crash and in worst case may allow broader confidentiality or integrity impact due to kernel memory corruption.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Strong Important candidate / Actionable Moderate at minimum (with actual score 6) 	MAYBE	DANGER SIMPLEFIX HARDWARE TEST  KPANIC INCREASED_TO_MODERATE7_BASED_ON_GUESSCVSS	-	-	checked

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

ActionableScore=7
ActionableScoreLower=4

## 1. ActionableScore

* Conservative score: 4
* Paranoid score: 7
* Final recommended bucket: **Strong Important candidate / Actionable Moderate at minimum**::

## 2. Signal breakdown

Conservative signals:

* Generic memory corruption: +2. `ranges` is allocated by `kvcalloc()`, which may return vmalloc-backed memory, but the old code released it with `kfree()`.
* Real lifetime/freeing corruption: +1. This is an allocator/freeing mismatch, not only a bookkeeping bug.
* Firmware-mediated external influence: +1. The NPS range data comes from the AMDGPU discovery/NPS path, which is device/firmware mediated rather than ordinary local user input.
* Reliable kernel crash / strong DoS: +1. `kfree()` on vmalloc memory is a concrete invalid-free class and can corrupt allocator state or crash the kernel.
* Privileged kernel/device-management memory corruption path: +1. The bug is in GPU driver initialization/discovery style code running in a trusted kernel device-management context.
* Requires admin/root/CAP_* in typical deployments: -2. Reliable triggering likely needs driver load, reprobe, reset, specific hardware discovery data, or privileged GPU management.

Conservative total: 4.

Paranoid additions/changes:

* Local reduced-privilege trigger plausibility: +1. Some systems expose GPU reset/recovery or GPU-management paths to local services or less-than-full-root users.
* Weak LPE concern: +1. The primitive is kernel memory corruption, but there is no demonstrated reclaim, object replacement, callback control, or arbitrary write.
* Do not double-count C/I separately. The weak LPE concern already captures the possible confidentiality/integrity escalation concern.
* Hard/special trigger conditions remain a limiting factor, but not enough to auto-close because the patch explicitly fixes a memory-corruption-class allocator mismatch.

Paranoid total: 7.

## 3. Reachability analysis

The realistic trigger appears local and hardware dependent. It likely requires an AMDGPU device using the affected NPS discovery path, and the allocation must be large enough for `kvcalloc()` to use vmalloc-backed memory. In typical deployments, a normal user cannot directly feed arbitrary NPS discovery tables or freely reprobe the driver, so the conservative privilege model is admin or privileged device-management access.

Namespaces and containers usually do not make this path directly reachable unless GPU devices and reset or management interfaces are delegated into the container. A reduced-capability service account or container root with GPU device access may lower the practical privilege boundary, which is why the paranoid score treats PR:L-style reachability as possible. This is not network reachable.

Call-site confidence: medium. The commit identifies the allocator source and the wrong free site, but the full caller graph and all ways to reach `amdgpu_gmc_get_nps_memranges()` are not included in the provided patch.

## 4. Severity interpretation

This is more than an ordinary Moderate correctness issue because the patch explicitly fixes an invalid freeing mismatch that can cause kernel memory corruption. The realistic exploitation evidence is limited: there is no shown attacker-controlled payload, no demonstrated reclaim, no arbitrary write, and no public exploit path in the provided data. Therefore the conservative result is borderline manual-review territory.

The paranoid interpretation is stronger because allocator misuse inside a GPU driver discovery path can be security-relevant, especially if local users or reduced-privilege services can trigger reset/reinit paths. It should not be auto-closed as a harmless DoS without checking reachability on the target product.

## 5. One-sentence report phrase

A mismatched free in the AMDGPU NPS discovery path can release vmalloc-backed memory allocated by kvcalloc() through kfree(), leading to kernel memory corruption or crash, with realistic exploitation depending on AMDGPU hardware, discovery data, and local ability to trigger driver initialization or reset paths.

## 6. Manual review recommendation

MANUAL CHECK REQUIRED

Reason: the bug is a concrete allocator/free mismatch with explicit memory corruption impact, but practical reachability depends on AMDGPU hardware, NPS discovery behavior, allocation size, and whether local or reduced-privilege users can trigger the affected path.

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

Patch: drm/amdgpu: Use kvfree instead of kfree in amdgpu_gmc_get_nps_memranges() [ Upstream
Commit: 16e7e7ad8cdc6b4c4af7f31e262f1494c1b2a55e
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=16e7e7ad8cdc6b4c4af7f31e262f1494c1b2a55e

Changed files:
  drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.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=16e7e7ad8cdc6b4c4af7f31e262f1494c1b2a55e

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

The original Linux kernel CVE announcement for CVE-2026-45853 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-45853

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:L/I:L/A:H
  The Best / paranoid CVSS vector: AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H
  The Best / paranoid CVSS score: 7

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: 7

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).