[CVE-2026-89924][LOW] KVM: s390: Fix old_data leak in guest debug error path

[AL-KERNEL <alkernel@kernelcve.org>]

CVE: CVE-2026-89924
Priority: LOW
AL-KERNEL base severity: LOW
KPANIC flag: NO
Patch: KVM: s390: Fix old_data leak in guest debug error path
Commit: 124c81ee610e1fbdd93d4399f88d9e28ba97a941
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=124c81ee610e1fbdd93d4399f88d9e28ba97a941
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89924
Analysis date: Wed, 16 Sep 2026 12:38:14 -0400
ActionableScore: 2
ActionableScore lower bound: unknown
Actionable bucket: Ordinary Moderate
Manual review required: NO

Summary:
KVM s390 guest debug can leak per-watchpoint old_data buffers on a failed KVM_SET_GUEST_DEBUG request, allowing a local process with KVM access to repeatedly consume host kernel memory and cause resource-exhaustion DoS.

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

The original Linux kernel CVE announcement for CVE-2026-89924 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89924

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-2026-89924	LOW	https://www.kernelcve.org/list/?q=CVE-2026-89924 CHECK WITH IMPACT FROM ORIG NN LOW	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-401;*CWE-772;**CWE-400;Other CVSS 'AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L';BEST CVSS score: '5.5';DESCR 'KVM s390 guest debug can leak old_data buffers when KVM_SET_GUEST_DEBUG imports several watchpoints and a later watchpoint import fails. The already imported watchpoints keep allocated backup buffers, but the error path frees only the arrays and not those old_data allocations, so repeated failed requests can consume host kernel memory. For the CVSS the PR:L is used because a local process needs access to the KVM ioctl interface and to create or control a VM, but full host root privileges are not necessarily required when /dev/kvm access is delegated. The issue is not network reachable and there is no direct guest to host remote vector. Impact is denial of service through resource exhaustion, with no demonstrated information disclosure or privilege escalation primitive because the leaked memory is not returned to the attacker.';NO MANUAL CHECK; ,and ActionableScore result is Ordinary Moderate (with actual score 2)	YES	VIRT SIMPLEFIX LEAK ERRORPATH HARDWARE LINUS  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=2
PrimitiveClass=NONE
RealUAF=NO
HostAdminRequired=NO
NamespaceScopedPrivilege=NO
ImportantCandidate=NO
ManualReview=NO
AutoDowngradeAllowed=YES

## 1. ActionableScore

* Raw conservative score: 2
* Conservative score: 2
* Raw paranoid score: 2
* Paranoid score: 2
* Final recommended bucket: **Ordinary Moderate**::

No floor or cap changed the score.

## 2. Signal breakdown

Positive signals:

* +1 Concrete persistent kernel resource/reference leak. `__import_wp_info()` allocates `old_data` buffers, and the old error path freed `bp_data` and `wp_info` but not already allocated `wp_info[nr_wp].old_data`.
* +1 Repeatable resource exhaustion potential. The commit states that up to `MAX_BP_COUNT - 1` buffers of up to `MAX_WP_SIZE` may be leaked per failed `KVM_SET_GUEST_DEBUG` request, and the request can be repeated.

Negative or non-awarded signals:

* No UAF awarded. The bug is a missing cleanup of allocated buffers, not stale use after free.
* No concrete repeated release/refcount mutation. The patch adds missing `kfree()` calls, but there is no second put/free on an already freed object.
* No controlled reclaim, arbitrary write, OOB write, callback control, or type confusion demonstrated.
* No confidentiality or integrity signal. The leaked `old_data` remains kernel memory and is not returned to the attacker.
* No reliable immediate kernel crash signal. The likely consequence is gradual memory/resource exhaustion rather than a direct panic.
* No remote reachability. Triggering requires local KVM ioctl access.
* No KVM guest-to-host exploitation leverage awarded. The relevant actor is the local VMM or process using `/dev/kvm`, not guest-only code, and no guest-to-host corruption primitive is established.
* No manual-review preservation signal. This is a pure resource leak without memory corruption, ownership violation, or authorization bypass.

## 3. Reachability analysis

The trigger is a local KVM s390 guest-debug ioctl path, specifically a `KVM_SET_GUEST_DEBUG` request containing multiple watchpoints where one later watchpoint import fails after earlier `old_data` allocations succeeded.

The attacker model is a local VMM or local process with access to `/dev/kvm` and control over a VM or vCPU debug configuration. Full host root is not inherently required when KVM access is delegated through normal `/dev/kvm` permissions or a `kvm` group, so `HostAdminRequired=NO`.

This is not namespace-scoped. User or network namespaces do not appear to provide the complete vulnerable operation.

There is no network trigger and no guest-only trigger established. A malicious guest by itself is not shown to issue the host-side `KVM_SET_GUEST_DEBUG` ioctl. The complete trigger is controlled by host userspace with KVM access.

The issue is architecture-specific to s390 KVM guest debugging, but this is not scored as rare hardware because the patch is for a normal supported KVM architecture path. No race or timing requirement is involved.

## 4. Primitive and severity interpretation

The exact primitive is a persistent kernel memory leak in an error path. Successfully imported watchpoints allocate `old_data`, then a later failure exits through an error label that previously skipped freeing those per-watchpoint buffers.

This is not active structured stale-object use and not fixed stale storage UAF. It is a resource cleanup omission.

There is no concrete repeated release/refcount operation. The first operation is allocation, not a release that can end lifetime. The same reference does not survive into a second lifetime-mutating operation. The patch only adds the missing release of allocated buffers.

PrimitiveClass is `NONE` because no meaningful corruption, UAF, ownership violation, authorization bypass, or semantic reinterpretation primitive is established.

The demonstrated consequence is host kernel memory/resource leakage. The plausible operational consequence is local DoS through repeated failed requests causing memory pressure or exhaustion.

Speculative consequences such as privilege escalation or information disclosure are not supported by the patch mechanics.

`ImportantCandidate=NO` because this is a pure repeatable leak with no memory corruption or security-boundary primitive. `AutoDowngradeAllowed=YES` because no unresolved Important-grade primitive would be lost by treating it as ordinary Moderate/Low-like triage.

## 5. One-sentence report phrase

KVM s390 guest debug can leak per-watchpoint old_data buffers on a failed KVM_SET_GUEST_DEBUG request, allowing a local process with KVM access to repeatedly consume host kernel memory and cause resource-exhaustion DoS.

## 6. Manual review recommendation

NO MANUAL CHECK NEEDED

Reason: this is a concrete but straightforward memory leak with no UAF, no repeated refcount mutation, no C/I primitive, no authorization bypass, and no remote or guest-only trigger.

## 7. Call-site confidence

Call-site confidence: high

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

Patch: KVM: s390: Fix old_data leak in guest debug error path
Commit: 124c81ee610e1fbdd93d4399f88d9e28ba97a941
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=124c81ee610e1fbdd93d4399f88d9e28ba97a941

Commit description:

__import_wp_info() allocates a per-watchpoint old_data buffer to back up
the original guest memory contents. If a later watchpoint of the same
KVM_SET_GUEST_DEBUG request fails to import, kvm_s390_import_bp_data()
jumps to the error label, which frees the wp_info array but not the
old_data buffers of the entries that were imported successfully. Up to
MAX_BP_COUNT - 1 buffers of up to MAX_WP_SIZE bytes are leaked per failed
request, and the request can be repeated.

Create error handling for cleaning up all created old_data memory
areas.

Fixes: 27291e2 ("KVM: s390: hardware support for guest debugging")
Cc: stable@vger.kernel.org
Signed-off-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Reviewed-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260805110455.7200-6-borntraeger@linux.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed files:
  arch/s390/kvm/guestdbg.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=124c81ee610e1fbdd93d4399f88d9e28ba97a941

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

The original Linux kernel CVE announcement for CVE-2026-89924 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89924

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: 2

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).