[CVE-2026-89805][MODERATE REGULAR] drm/pagemap: Fix folio allocation fallback and use-after-put

[CVE-2026-89805][MODERATE REGULAR] drm/pagemap: Fix folio allocation fallback and use-after-put

[AL-KERNEL]

CVE: CVE-2026-89805
Priority: MODERATE REGULAR
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: drm/pagemap: Fix folio allocation fallback and use-after-put
Commit: 10e92e18e3dab081789d868878f2ecce6166948a
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=10e92e18e3dab081789d868878f2ecce6166948a
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89805
Analysis date: Wed, 16 Sep 2026 10:37:17 -0400
ActionableScore: 4
ActionableScore lower bound: unknown
Actionable bucket: Borderline / manual review recommended
Manual review required: YES

Summary:
No one-sentence report phrase was found.

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

The original Linux kernel CVE announcement for CVE-2026-89805 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89805

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: YES

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

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

CVE-2026-89805	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-89805 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-416;*CWE-672;CWE-703;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '7';DESCR 'A use-after-put can occur in drm_pagemap_migrate_populate_ram_pfn because the error cleanup path called put_page(page) before reading the folio order from the same page. If put_page drops the last reference, later folio_order(page_folio(page)) can observe released or reused page state and may drive incorrect cleanup of the migration range. For the CVSS the PR:L is used because a local user or process with access to the affected DRM device can potentially reach GPU memory migration paths, while full administrator privileges are not clearly required by the patch context. The issue is not network reachable and requires local interaction with DRM or GPU memory migration behavior under conditions such as higher order folios and memory pressure. Impact is at least a local denial of service via kernel crash or failed cleanup, and in the paranoid case may allow confidentiality or integrity impact due to the use-after-put primitive in memory migration code. YES REQUIRES MANUAL CHECK';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline / manual review recommended (with actual score 4)	YES	READ ERRORPATH UAF HARDWARE TEST MEMORY  DECREASED_TO_MODERATEREG_BASED_ON_FALSEPOSCHECKOFKP	YES	NO	checked

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

ActionableScore=4
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=UNKNOWN
NamespaceScopedPrivilege=NO
ImportantCandidate=NO
ManualReview=RECOMMENDED
AutoDowngradeAllowed=YES

1. ActionableScore

* Raw conservative score: 3
* Conservative score: 4
* Raw paranoid score: 3
* Paranoid score: 4
* Final recommended bucket: **Borderline / manual review recommended**::

Applied conservative floor: 3 -> 4 due to a real use-after-put with stale reference use and no positively established host-admin requirement.
Applied paranoid floor: 3 -> 4 for the same reason. No Floor 5 was applied.

2. Signal breakdown

Positive signals:

* +2 Real UAF / lifetime violation: the patch explicitly states that `folio_order(page_folio(page))` was computed after `put_page(page)`, and `put_page()` may drop the last reference.
* +1 Reliable kernel crash / strong DoS: use-after-put of page or folio metadata in the cleanup path can plausibly fault or corrupt cleanup state enough to crash or destabilize the kernel.

Negative or non-awarded signals:

* +1 Concrete repeated release/refcount was not awarded. The stale operation is a metadata read through `page_folio()` / `folio_order()`, not a second `put_page()` or other lifetime/refcount mutation on the same dangling reference.
* Floor 5 was not applied. There is no concrete second lifetime mutation, controlled reclaim, callback use, arbitrary write, or established below-host-admin high-control exploitation surface.
* Structured stale-object leverage was not separately awarded. The stale value can affect cleanup iteration, but attacker-controlled replacement or security-useful post-free field consumption is not established.
* Weak LPE was not awarded. The evidence supports UAF and DoS risk, but not a concrete privilege-escalation primitive.
* C/I impact was not awarded. The paranoid CVSS hint with C:H/I:H is not enough by itself for ActionableScore C/I bonuses.
* Extra availability was not awarded. The same crash or failed cleanup is already counted once.
* Remote reachability was not awarded. This is a local DRM/GPU memory migration path, not a network path.
* Hardware rarity penalty was not applied. GPU/DRM involvement alone does not prove rare or physical-only hardware.
* Host-admin penalty was not applied. Host-admin-equivalent privilege is not positively established from the supplied context.

Repeated release/refcount check:

* First release can end lifetime: yes, `put_page(page)` can drop the last reference.
* Exact same reference survives: yes, `page` remains used after `put_page(page)`.
* Second operation mutates lifetime/refcount: no, the second operation reads folio metadata.
* +1 Concrete repeated release/refcount awarded: no.
* Paranoid repeated-refcount leverage awarded: no.
* Floor 5 applied: no.

3. Reachability analysis

The vulnerable code is in DRM pagemap migration from device memory to RAM, including eviction and migration-to-RAM paths. A realistic trigger likely requires local interaction with an affected DRM/GPU device and conditions that enter `drm_pagemap_migrate_populate_ram_pfn()`, especially higher-order folios and an allocation failure or cleanup path.

Below-host-admin reachability is plausible for systems where user processes can access the relevant DRM render or GPU memory-management interface. However, the supplied patch does not show the complete userspace entry point or the exact device access policy. Therefore `HostAdminRequired=UNKNOWN`, not `NO`.

No namespace-scoped capability chain is established. There is no evidence that the complete operation is available through attacker-created namespaces. `NamespaceScopedPrivilege=NO` for this analysis.

Prior setup may include an affected GPU driver using DRM pagemap device memory migration. This is a configuration and hardware dependency, but not enough to apply a rare-hardware penalty. The issue is not network reachable. It is not primarily a race. It is an ordered error-cleanup sequence involving memory pressure or allocation failure.

For DRM roles: ordinary logged-in user or compositor-like access may be relevant in real deployments, but the complete vulnerable operation is not proven from the patch alone. Host administrator access is also not proven as required.

4. Primitive and severity interpretation

The exact primitive is a use-after-put of a `struct page` / folio reference in the `free_pages` error path. The code releases the page first and then reads folio metadata from the same reference to compute the order used for cleanup iteration.

This is an active kernel memory-management object rather than a fixed async buffer. However, the demonstrated stale use is in error cleanup and consists of metadata consumption, not a second lifetime mutation, callback dispatch, controlled object replacement, arbitrary read/write, or demonstrated privilege escalation.

PrimitiveClass is CLASS_B because a real kernel lifetime violation exists. It is not CLASS_A because controlled reclaim, directed corruption, refcount takeover, callback control, or demonstrated LPE is not established. It is not CLASS_C because the patch directly establishes an actual post-lifetime object use, not merely an invalid state or warning.

Demonstrated consequence: local kernel crash or failed cleanup.
Speculative consequence: stronger memory-management corruption if the released page metadata is reused before the stale read, but the supplied evidence does not establish attacker-controlled reuse or a security-useful primitive.

ImportantCandidate=NO because this looks like a constrained cleanup-path UAF with DoS as the established impact and no concrete repeated refcount mutation or strong exploitation leverage. ManualReview=RECOMMENDED because it is still a real UAF in page migration code and should not be dismissed as harmless from the one-line fix alone. AutoDowngradeAllowed=YES because no unresolved Important-grade primitive is established under the provided evidence.

Likely operational severity: borderline Moderate. It is worth manual triage, especially for products exposing affected DRM/GPU migration paths to unprivileged users, but the supplied evidence does not justify Strong Important classification.

5. One-sentence report phrase

A use-after-put in drm_pagemap_migrate_populate_ram_pfn can read folio metadata after put_page may have released the last reference, causing a local kernel crash or incorrect cleanup during DRM GPU memory migration error handling.

6. Manual review recommendation

MANUAL CHECK RECOMMENDED

Reason: the patch explicitly fixes a real use-after-put in memory migration code, but the established stale operation is cleanup-path metadata reading and no controlled reclaim, repeated refcount mutation, arbitrary write, or LPE path is shown.

7. Call-site confidence

Call-site confidence: medium

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

Patch: drm/pagemap: Fix folio allocation fallback and use-after-put
Commit: 10e92e18e3dab081789d868878f2ecce6166948a
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=10e92e18e3dab081789d868878f2ecce6166948a

Commit description:

drm_pagemap_migrate_populate_ram_pfn() had two issues when populating
RAM PFNs with higher-order folios:

1. The higher-order vma_alloc_folio()/folio_alloc() calls did not pass
   __GFP_NOWARN, so a THP allocation failure under memory pressure
   would spam the kernel log, and there was no fallback path despite a
   TODO comment stating one was needed. Add __GFP_NOWARN to the
   higher-order allocation and, on failure, fall back to order-0
   allocations for the entire range originally covered by the failed
   higher-order allocation, leaving MIGRATE_PFN_COMPOUND unset for
   those PFNs.

2. In the free_pages error path, order was computed via
   folio_order(page_folio(page)) *after* put_page(page) had already
   dropped the reference, resulting in a use-after-free/put when that
   was the last reference on the page. Compute order before releasing
   the page.

Introducing the fallback in 1. also requires the source page array
handed to ->copy_to_ram() to be built differently. Both callers only
populated the entry at the head of each source folio, relying on the
copy callback to derive the rest of the folio from the order recorded
in the matching drm_pagemap_addr. Once the destination has been demoted
to order-0 folios the drm_pagemap_addr entries are per-page, so a source
page is needed for every one of them; leaving them NULL makes the copy
callback stop after the first page and the remainder of the range is
never copied.

The source folio is only split later, by migrate_vma_pages() /
migrate_device_pages(), so its order cannot be used to detect the
demotion - test the destination for MIGRATE_PFN_COMPOUND instead. Factor
the array population out into drm_pagemap_migrate_populate_src_pages()
and use it from both drm_pagemap_evict_to_ram() and
__drm_pagemap_migrate_to_ram().

Fixes: ddeda61 ("drm/pagemap: Allocate folios when possible")
Cc: stable@vger.kernel.org
Assisted-by: GitHub_Copilot:claude-opus-5
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Link: https://patch.msgid.link/20260902063504.3024362-2-matthew.brost@intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed files:
  drivers/gpu/drm/drm_pagemap.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=10e92e18e3dab081789d868878f2ecce6166948a

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

The original Linux kernel CVE announcement for CVE-2026-89805 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89805

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: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: unknown
  Paranoid ActionableScore: 4

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).