[CVE-2026-98254][MODERATE 7.0] swiotlb: use the adjusted address for the highmem page lookup [ Upstream

[CVE-2026-98254][MODERATE 7.0] swiotlb: use the adjusted address for the highmem page lookup [ Upstream

[AL-KERNEL]

CVE: CVE-2026-98254
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: swiotlb: use the adjusted address for the highmem page lookup [ Upstream
Commit: 6e53b4d6afbde626255805d438cabb1cac482445
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=6e53b4d6afbde626255805d438cabb1cac482445
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98254
Analysis date: Sat, 10 Oct 2026 13:47:17 -0400
ActionableScore: 5
ActionableScore lower bound: 4
Actionable bucket: Actionable Moderate / Strong Important candidate requiring manual review
Manual review required: YES

Summary:
No one-sentence report phrase was found.

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

The original Linux kernel CVE announcement for CVE-2026-98254 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98254

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-2026-98254	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-98254 CHECK WITH IMPACT FROM ORIG NN MODERATE | AEGIS IMPACT MODERATE_KPANIC AEGIS_AS=1 LEGACY=MODERATE_KPANIC RESULT=SAME (legacy and Aegis reconciliation agree; DECISION: both MODERATE_KPANIC)	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-682;*CWE-787;CWE-125;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '7';DESCR 'A page calculation error in swiotlb_bounce can combine the page frame from the original DMA address with the offset from the adjusted address. When a partial DMA synchronization uses a nonzero tlb_offset and crosses a page boundary, DMA_FROM_DEVICE can overwrite the wrong highmem page while DMA_TO_DEVICE can expose data from a page outside the intended mapping to the device. This is a concrete wrong page read and write primitive with potential memory corruption rather than a warning only condition. For the CVSS the PR:L is selected because a local user or process must access an affected device interface and initiate driver operations that reach the partial DMA synchronization path. AC:H reflects the required SWIOTLB and highmem configuration, a suitable driver path, and a range adjustment that crosses a page boundary. The issue is not classified as directly network reachable because the generic SWIOTLB code does not establish a remote trigger without a specific network driver path. Impact is at least denial of service and limited confidentiality or integrity impact. In the worst defensible case the wrong page write may affect sensitive memory and could support privilege escalation, so manual review of reachable drivers and memory placement is required.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 5)	MAYBE	OOB HARDWARE LINUS TEST KPANIC MEMORY  KPANIC INCREASED_TO_HIGH_BASED_ON_GUESSCVSS DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFKP	-	-	checked

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

ActionableScore=5
ActionableScoreLower=4
PrimitiveClass=CLASS_B
RealUAF=NO
HostAdminRequired=UNKNOWN
NamespaceScopedPrivilege=UNKNOWN
ImportantCandidate=YES
ManualReview=REQUIRED
AutoDowngradeAllowed=NO
RequiresHostUID0=UNKNOWN
AuthorizationDecisionBypass=NO
ProtectedResourceImpact=NONE
B4PendingReview=NO
B4PendingConfirmed=NO

1. ActionableScore

* Raw conservative score: 4
* Conservative score: 4
* Raw paranoid score: 5
* Paranoid score: 5
* No floor or cap was applied.
* Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**::

2. Signal breakdown

Conservative signals:

* +1 Constrained kernel overwrite. `DMA_FROM_DEVICE` copies device data into the wrong adjacent highmem page when the adjusted address crosses a page boundary.
* +1 Confidentiality plausible. `DMA_TO_DEVICE` can copy data from a page outside the DMA mapping to the device, creating a concrete device-visible wrong-page read.
* +1 Integrity plausible. `DMA_FROM_DEVICE` modifies a page outside the authorized mapping, violating the mapping boundary even though no specific protected victim is identified.
* +1 Privileged kernel/device-management corruption path. The defect breaks DMA page ownership and address selection inside SWIOTLB.

Paranoid-only signal:

* +1 Strong DoS plausible. Copying an entire requested range into an adjacent active page can realistically corrupt state and cause a host fault, although the patch does not demonstrate a reliable panic.

Important non-awarded signals:

* No arbitrary or attacker-selected write target is demonstrated.
* Device data may influence the payload, but attacker control of that data is unresolved.
* No controlled reclaim, callback overwrite, credential modification, or completed privilege-escalation chain is established.
* No local-unprivileged bonus was awarded because no concrete driver call path or userspace interface is supplied.
* No remote bonus was awarded. SWIOTLB use by a network device alone would not prove direct remote reachability.
* No hardware, rarity, privilege, or timing penalty was applied because those restrictions are not positively established.
* B5-R does not apply. There is no retry-accumulated index or offset.
* B3-PENDING does not apply. The patch does not establish shared file-backed or cross-authority backing.
* B4-PENDING does not apply. No authorization decision is involved.

3. Reachability analysis

The vulnerable operation requires SWIOTLB, a highmem-relevant configuration, a partial `dma_sync_single_range_for_*()` operation with nonzero `tlb_offset`, and an adjustment crossing a page boundary. A suitable driver and device operation must reach this path.

The supplied evidence does not identify whether an ordinary user, delegated device service, guest, firmware, or host administrator controls the complete operation. Access through a userspace device API is plausible but not established. Host-admin privilege therefore remains unknown rather than assumed, and no privilege penalty or unprivileged-trigger bonus is applied.

No namespace-scoped complete trigger, prior privileged setup, direct network path, retry mechanism, or race is established. The defect is deterministic once the required address relationship occurs.

4. Primitive and severity interpretation

The exact primitive is a page-frame and page-offset mismatch. The old code derives `pfn` from the pre-adjustment `orig_addr`, then derives the offset from the post-adjustment address. Crossing a page boundary therefore produces a coherent-looking page and offset pair that identifies the wrong page.

This is not a UAF or stale-storage bug. It is a concrete, constrained wrong-page read/write in a DMA ownership path:

* `DMA_FROM_DEVICE` writes device data to the wrong page.
* `DMA_TO_DEVICE` exposes data from the wrong page to the device.
* A lowmem-to-highmem transition can also select the wrong access method.

The primitive is CLASS_B because real memory corruption and disclosure are established, but attacker control over the payload, destination, and complete triggering interface is unresolved. CLASS_A is not justified without a directed target, meaningful attacker-controlled payload, demonstrated protected victim, or completed exploitation chain.

ImportantCandidate is YES because automatic downgrade would discard a concrete write outside the intended DMA mapping and a concrete device-visible read outside that mapping. AutoDowngradeAllowed is NO for the same reason. Practical severity is Actionable Moderate pending driver-specific reachability and target-control analysis.

5. One-sentence report phrase

A page-frame and offset mismatch in swiotlb_bounce causes partial DMA synchronization across a page boundary to read or overwrite the wrong highmem page, creating a concrete but constrained DMA memory-corruption and disclosure primitive whose attacker reachability requires driver-specific review.

6. Manual review recommendation

MANUAL CHECK REQUIRED

Review affected driver call sites, userspace device access, control over partial synchronization ranges, device-data influence, adjacent-page ownership, and whether the wrong-page write can reach active kernel or protected user data.

7. Call-site confidence

Call-site confidence: medium

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

Patch: swiotlb: use the adjusted address for the highmem page lookup [ Upstream
Commit: 6e53b4d6afbde626255805d438cabb1cac482445
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=6e53b4d6afbde626255805d438cabb1cac482445

Changed files:
  kernel/dma/swiotlb.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=6e53b4d6afbde626255805d438cabb1cac482445

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

The original Linux kernel CVE announcement for CVE-2026-98254 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98254

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: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: 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).