* [CVE-2026-93262][MODERATE 7.0] md/raid5-ppl: fix use-after-free in ppl_do_flush() [ Upstream
@ 2026-09-24 23:54 21% AL-KERNEL
0 siblings, 0 replies; 1+ results
From: AL-KERNEL @ 2026-09-24 23:54 UTC (permalink / raw)
To: kernel-cve
CVE: CVE-2026-93262
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: md/raid5-ppl: fix use-after-free in ppl_do_flush() [ Upstream
Commit: cf01f9413565e86673baf927291a088b6975692b
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=cf01f9413565e86673baf927291a088b6975692b
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-93262
Analysis date: Thu, 24 Sep 2026 19:54:25 -0400
ActionableScore: 6
ActionableScore lower bound: 5
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-2026-93262 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-93262
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-2026-93262 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-93262 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-664;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 free in the RAID5 PPL flush path occurs because ppl_do_flush can continue iterating after ppl_io_unit_finished releases the io object. A later loop iteration performs an atomic operation on io pending_flushes after the object may have been freed. This can cause a kernel crash and local denial of service. The post free atomic write also creates a plausible memory corruption primitive if the freed allocation is reused, so confidentiality integrity and privilege escalation impact cannot be excluded without manual review. For the CVSS the PR:L rating is used because a local user who can generate writes and flushes to a filesystem or block device backed by the affected RAID volume may be able to participate in triggering the condition. Exploitation also requires a RAID5 PPL configuration and a suitable flush completion sequence, therefore AC:H is used. The issue is not directly network reachable. Remote influence is possible only indirectly when an external storage service exposes writes to the affected RAID volume.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 6) YES DISK INIT UAF KPANIC KPANIC INCREASED_TO_HIGH_BASED_ON_GUESSCVSS DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFKP NO NO checked
======================================================================
ACTIONABLESCORE ANALYSIS
======================================================================
ActionableScore=6
ActionableScoreLower=5
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=NO
NamespaceScopedPrivilege=NO
ImportantCandidate=YES
ManualReview=REQUIRED
AutoDowngradeAllowed=NO
1. ActionableScore
* Raw conservative score: 5
* Conservative score: 5
* Raw paranoid score: 6
* Paranoid score: 6
* Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**::
The real-UAF FLOOR 5 was checked. It applies because the object can be freed, the same `io` reference remains in the loop, and a later iteration performs another atomic lifetime-related completion-counter mutation. The conservative raw score already reaches 5, so the floor does not modify it.
2. Signal breakdown
Positive signals:
* +2 Real UAF / active lifetime violation. `ppl_io_unit_finished(io)` can free `io`, while the loop retains and subsequently accesses the same `io` pointer.
* +1 Concrete repeated release/refcount-style mutation. The first `atomic_dec_and_test(&io->pending_flushes)` reaches the terminal completion state and invokes the helper that may free `io`. The exact same local `io` reference survives, and a subsequent loop iteration performs another atomic decrement through potentially freed storage. This is a concrete post-lifetime write-like mutation, not merely a stale read.
* +1 Local unprivileged trigger. A local user able to write and flush data on a filesystem or block device backed by the affected RAID5 PPL array can participate in generating the relevant write and flush workload. The RAID5 PPL configuration is a deployment prerequisite, but does not necessarily require attacker host-admin privileges after the array is available.
* +1 Reliable kernel crash / strong DoS. Accessing `io->pending_flushes` after `io` is freed can cause a kernel fault, memory corruption detection failure, or later kernel instability.
Paranoid-only signal:
* +1 Post-free lifetime mutation leverage. The stale operation is an atomic modification of memory whose original `ppl_io_unit` lifetime may have ended. This is stronger than a stale dereference that immediately faults, but does not establish controlled reclaim, arbitrary write, refcount takeover, or privilege escalation.
Non-awarded material signals:
* Controlled reclaim or attacker-controlled replacement of the freed `ppl_io_unit` is not demonstrated.
* Arbitrary or directed kernel write is not demonstrated.
* Structured stale-object leverage is not separately awarded. The patch establishes only the stale `pending_flushes` mutation, not later consumption of attacker-controlled replacement fields, callbacks, pointers, offsets, or operations.
* Weak LPE concern is not separately awarded because that would double-count the same post-free atomic mutation.
* Confidentiality and integrity impact are not awarded. A post-free atomic decrement is a real corruption primitive, but no disclosure path or unauthorized modification of a protected object is established.
* Extra Availability impact is not awarded because it would double-count the same kernel crash / DoS consequence.
* Remote reachability is not established. Network storage exports may indirectly expose writes to the affected volume, but the vulnerable function is not itself a network protocol parser.
* No hard-race penalty is applied. This is an ordered loop-control lifetime error, not a timing race requiring an uncontrollable interleaving.
* No niche-subsystem penalty is applied. RAID5 PPL with write-back-cache support is a configuration prerequisite, but the supplied evidence does not establish that it is attacker-unreachable once deployed.
Repeated lifetime mutation assessment:
* First lifetime-ending operation: `ppl_io_unit_finished(io)` may free the object after the completion count reaches zero.
* Exact same reference survives: yes. The loop-local `io` pointer is not cleared and the loop continues.
* Second operation is concrete: yes. The next iteration executes `atomic_dec_and_test(&io->pending_flushes)`.
* Concrete repeated release/refcount signal: awarded as an equivalent post-lifetime completion-counter mutation.
* Paranoid repeated-refcount leverage: awarded. The reason is the concrete stale atomic write through storage after the original object lifetime may have ended, not generic theoretical heap reuse.
* FLOOR 5: applies, but does not change the conservative score because the raw conservative score is already 5.
3. Reachability analysis
The practical attacker model is a local user with permission to write to a filesystem or block device hosted on an already configured affected RAID5 PPL array. Normal writeback and flush activity can reach RAID5 PPL handling without requiring the user to administer the MD array directly.
The vulnerable array must already use RAID5 PPL and the relevant write-back-cache configuration. Creating or configuring the MD RAID array itself normally requires administrative action, but this is prior environment setup rather than a demonstrated per-attack host-admin requirement.
HostAdminRequired is set to NO because the vulnerable I/O path can be exercised by a below-host-admin user who is permitted to write data to the affected storage volume. This conclusion assumes the storage volume is exposed to such a user. If the only writable interface is restricted to host administrators, practical reachability would be lower, but that restriction is not established by the supplied patch.
NamespaceScopedPrivilege is set to NO. No network namespace, user namespace, or namespaced capability mechanism is relevant to the required write and flush path.
The issue is not directly network reachable. A remote client could only influence it indirectly if a separate storage service exports writable access to the affected RAID volume. That service exposure is outside the patched `raid5-ppl` logic.
4. Primitive and severity interpretation
The exact primitive is a real use-after-free in an active RAID5 PPL I/O object. The terminal decrement of `io->pending_flushes` can result in `ppl_io_unit_finished(io)` freeing the `ppl_io_unit`. The enclosing loop then continues and performs another `atomic_dec_and_test()` through the same stale `io` pointer.
This is not merely an ordinary stale cleanup dereference:
A. Ordinary stale cleanup dereference would read a stale field and immediately fault.
B. This case performs a second write-like completion/lifetime-counter mutation through storage whose original object lifetime may have ended.
The object is an active structured I/O state object, but no evidence shows controlled replacement of that object, attacker-controlled post-free contents, callback dispatch, pointer dereference from replacement data, or an attacker-directed write target. Therefore the issue remains CLASS_B rather than CLASS_A.
The demonstrated practical consequence is kernel crash or local denial of service. The speculative consequence is corruption of a subsequently reused allocation through the stale atomic decrement. That possibility requires manual examination of `ppl_io_unit` allocation, freeing, reuse behavior, and whether the `pending_flushes` offset can be meaningfully influenced after reuse.
ImportantCandidate is YES because the concrete second post-free mutation leaves meaningful exploitation leverage unresolved. AutoDowngradeAllowed is NO because automatically classifying the issue as ordinary crash-only cleanup would discard the concrete post-lifetime mutation primitive.
Likely operational severity is Actionable Moderate. It should be reviewed for Important classification if allocator reuse, reallocation control, or downstream corruption effects are established.
5. One-sentence report phrase
A use-after-free in the RAID5 PPL flush completion loop permits a second atomic update of `io->pending_flushes` after `ppl_io_unit_finished()` may free the same I/O object, causing local kernel DoS and requiring manual review for post-free corruption leverage.
6. Manual review recommendation
MANUAL CHECK REQUIRED
The patch establishes a real UAF with a concrete second atomic mutation through potentially freed object storage. Controlled reuse and privilege escalation are not proven, but the post-lifetime write-like operation should not be auto-closed as crash-only.
7. Call-site confidence
Call-site confidence: medium
======================================================================
UPSTREAM PATCH SUMMARY
======================================================================
Patch: md/raid5-ppl: fix use-after-free in ppl_do_flush() [ Upstream
Commit: cf01f9413565e86673baf927291a088b6975692b
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=cf01f9413565e86673baf927291a088b6975692b
Changed files:
drivers/md/raid5-ppl.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=cf01f9413565e86673baf927291a088b6975692b
======================================================================
DETAILED REPORT METHODOLOGY
======================================================================
The original Linux kernel CVE announcement for CVE-2026-93262 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-93262
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: 5
Paranoid ActionableScore: 6
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 [email protected] (and both
send reply to CVE record itself too and see "reply" button below for howto reply).
^ permalink raw reply [relevance 21%]
Results 1-1 of 1 | reverse | sort options + mbox downloads above
-- links below jump to the message on this page --
2026-09-24 23:54 21% [CVE-2026-93262][MODERATE 7.0] md/raid5-ppl: fix use-after-free in ppl_do_flush() [ Upstream AL-KERNEL
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox