[CVE-2026-90391][LOW] lib/test_hmm: fail dmirror_fault() when the mirrored mm is gone [ Upstream

[CVE-2026-90391][LOW] lib/test_hmm: fail dmirror_fault() when the mirrored mm is gone [ Upstream

[AL-KERNEL]

CVE: CVE-2026-90391
Priority: LOW
AL-KERNEL base severity: LOW
KPANIC flag: NO
Patch: lib/test_hmm: fail dmirror_fault() when the mirrored mm is gone [ Upstream
Commit: 50606ced303782f8715e2d0b98ca374ea498fa29
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=50606ced303782f8715e2d0b98ca374ea498fa29
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90391
Analysis date: Thu, 17 Sep 2026 20:35:39 -0400
ActionableScore: 1
ActionableScore lower bound: unknown
Actionable bucket: Ordinary Moderate
Manual review required: NO

Summary:
`dmirror_fault in the HMM selftest driver could incorrectly report success after the mirrored mm exited, causing dmirror_read or dmirror_write to retry forever and creating a local denial of service through a stuck ioctl path.`

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

The original Linux kernel CVE announcement for CVE-2026-90391 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90391

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: LOW
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-2026-90391	LOW	https://www.kernelcve.org/list/?q=CVE-2026-90391 CHECK WITH IMPACT FROM ORIG NN NONE	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H';*CWE-835;CWE-754;CWE-703;Other CVSS 'AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L';BEST CVSS score: '5.5';DESCR 'dmirror_fault in the HMM selftest driver could return success when mmget_not_zero failed because the mirrored mm had already exited. This makes dmirror_read or dmirror_write retry the same address without progress after ENOENT and can create an endless local retry loop. For the CVSS the PR:L is used for the paranoid interpretation because a local user with access to the test_hmm device interface could trigger the ioctl path without full administrative privileges. The issue is not network reachable and does not provide a demonstrated memory corruption primitive. Impact is denial of service through CPU consumption or a stuck ioctl path. Practical exposure is low because lib/test_hmm is a test driver and is usually not enabled or exposed on production systems.';NO MANUAL CHECK; ,and ActionableScore result is Ordinary Moderate (with actual score 1)	YES	WRITE OOB SIMPLEFIX SKIP HARDWARE TEST  INCREASED_FROM_LOW_BASED_ON_HIGHCVSSSCORE DECREASED_TO_MODERATEREG_BASED_ON_ACTIONABLESCORELESSTHAN4 DECREASED_TO_LOW_FROM_MODERATE_BASED_ON_ACTIONABLESCORELESSTHAN3	NO	NO	checked

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

ActionableScore=1
PrimitiveClass=NONE
RealUAF=NO
HostAdminRequired=UNKNOWN
NamespaceScopedPrivilege=NO
ImportantCandidate=NO
ManualReview=NO
AutoDowngradeAllowed=YES

## 1. ActionableScore

* Raw conservative score: 1
* Conservative score: 1
* Raw paranoid score: 1
* Paranoid score: 1
* Final recommended bucket: **Ordinary Moderate**::

No floor or cap was applied.

## 2. Signal breakdown

Positive signals:

* +1 Limited local DoS / stuck ioctl path: `dmirror_fault()` returned success when `mmget_not_zero()` failed, causing `dmirror_read()` or `dmirror_write()` retry loops to repeat the same address without progress.

Negative or non-awarded signals:

* +0 Real UAF not awarded: `mmget_not_zero()` fails before taking a reference to the exiting `mm`, and the patch does not show post-lifetime dereference of `mm`.
* +0 Memory corruption not awarded: no OOB access, stale object use, double free, refcount corruption, arbitrary write, or controlled kernel object corruption is established.
* +0 Reliable kernel crash not awarded: demonstrated consequence is an endless retry loop / stuck ioctl, not a kernel panic.
* +0 Confidentiality and integrity not awarded: no disclosure path or unauthorized modification primitive is shown.
* +0 Remote reachability not awarded: this is a local ioctl path in a test HMM driver.
* +0 Local unprivileged trigger not awarded as a scoring bonus: a local user with access to the `test_hmm` device may trigger it, but the supplied evidence does not fully establish ordinary unprivileged access in production.
* +0 Rare subsystem penalty not applied numerically: `lib/test_hmm` is a selftest driver and practical exposure is low, but the low score already reflects the weak security impact.

Repeated-release / refcount check:

* First release can end lifetime: no.
* Exact same reference survives: no concrete dangling reference is established.
* Second lifetime/refcount mutation: no.
* Concrete repeated release/refcount +1 awarded: no.
* Paranoid repeated-refcount leverage +1 awarded: no.
* FLOOR 5 applied: no.

## 3. Reachability analysis

The vulnerable path is local and goes through the HMM selftest `dmirror_read()` / `dmirror_write()` ioctl-style interface. A trigger requires access to the `test_hmm` driver interface and a mirrored process whose `mm` has already exited.

Host-admin-equivalent privilege is not positively established by the patch itself, but below-host-admin reachability is also not fully established because `lib/test_hmm` is a test driver and device exposure depends on local configuration and permissions. Therefore `HostAdminRequired=UNKNOWN`.

No namespace-scoped capability path is involved or established. No network trigger exists. No special race is required beyond the lifecycle condition that the mirrored `mm` is already gone. Practical deployment exposure is low because this is a selftest driver, not a normal production data path.

## 4. Primitive and severity interpretation

The exact primitive is an incorrect success return from `dmirror_fault()` when the mirrored `mm` is no longer faultable. This causes callers to retry the same missing device page table entry indefinitely.

This is a logic / error-handling bug, not a real UAF. The failed `mmget_not_zero()` means the code does not acquire a live reference, and the fix changes the return value from success to `-EFAULT` rather than adding lifetime synchronization or preventing stale dereference.

PrimitiveClass is `NONE` because no meaningful corruption, ownership, authorization, lifetime, or security-boundary primitive is established. The demonstrated consequence is local denial of service by CPU consumption or a stuck ioctl path. Speculative privilege escalation is not supported.

ImportantCandidate is `NO` because there is no memory corruption, no security-boundary bypass, no repeated lifetime/refcount mutation, and no unresolved exploitation primitive beyond the retry loop. AutoDowngradeAllowed is `YES` because this can be handled as an ordinary low/moderate DoS issue.

## 5. One-sentence report phrase

`dmirror_fault in the HMM selftest driver could incorrectly report success after the mirrored mm exited, causing dmirror_read or dmirror_write to retry forever and creating a local denial of service through a stuck ioctl path.`

## 6. Manual review recommendation

NO MANUAL CHECK NEEDED

Reason: the patch supports a DoS-only error-handling loop in a test driver, with no UAF, no memory corruption, no privilege escalation path, and no network exposure.

## 7. Call-site confidence

Call-site confidence: high

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

Patch: lib/test_hmm: fail dmirror_fault() when the mirrored mm is gone [ Upstream
Commit: 50606ced303782f8715e2d0b98ca374ea498fa29
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=50606ced303782f8715e2d0b98ca374ea498fa29

Changed files:
  lib/test_hmm.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=50606ced303782f8715e2d0b98ca374ea498fa29

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

The original Linux kernel CVE announcement for CVE-2026-90391 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90391

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:L/PR:L/UI:N/S:U/C:N/I:N/A:L
  The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
  The Best / paranoid CVSS score: 5.5

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: 1

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: LOW
KPANIC detected for this report: NO
Published priority for this report (same as in Subject): LOW

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).