[CVE-2026-99183][LOW] accel/rocket: search every core slot when looking up a scheduler

[CVE-2026-99183][LOW] accel/rocket: search every core slot when looking up a scheduler

[AL-KERNEL]

CVE: CVE-2026-99183
Priority: LOW
AL-KERNEL base severity: LOW
KPANIC flag: NO
Patch: accel/rocket: search every core slot when looking up a scheduler
Commit: c137756469d5bcc79e97beb406ae4d14b9585463
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=c137756469d5bcc79e97beb406ae4d14b9585463
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-99183
Analysis date: Fri, 30 Oct 2026 12:41:37 -0400
ActionableScore: 0
ActionableScore lower bound: unknown
Actionable bucket: Ordinary Moderate
Manual review required: NO

Summary:
Host-administrative removal of a middle Rocket NPU core can hide a still-running core from scheduler lookup and cause unchecked NULL dereferences in job and timeout workqueues.

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

The original Linux kernel CVE announcement for CVE-2026-99183 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-99183

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-2026-99183	LOW	https://www.kernelcve.org/list/?q=CVE-2026-99183 CHECK WITH IMPACT FROM ORIG NN LOW | AEGIS IMPACT MODERATE_KPANIC AEGIS_AS=1 LEGACY=LOW RESULT=DIFF TEMP_DISTANCE_MERGE_SKIP=AEGIS_NO_POSITIVE_SEVERITY_EVIDENCE (final Legacy and Aegis differ; FINAL: LOW, Aegis: MODERATE_KPANIC)	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:L/PR:H/UI:N/S:U/C:N/I:N/A:H';**CWE-476;CWE-682;*CWE-664;BEST CVSS score: '4.4';DESCR 'A scheduler lookup logic error occurs because rocket_remove decreases num_cores without compacting the rdev cores array, while sched_to_core searches only slots below num_cores. Unbinding a middle core can therefore hide a still bound trailing core and make sched_to_core return NULL, after which rocket_job_run or rocket_job_timedout dereferences the result and can crash the kernel. For the CVSS the PR:H value is selected because reliably creating the faulty state normally requires permission to unbind a platform device, which is an administrative operation even when ordinary clients can submit NPU jobs. The demonstrated primitive is a NULL pointer dereference with host availability impact. There is no supported UAF, OOB access, information disclosure, or writable memory corruption primitive. Automated impact decision path. The original SimpleNet classifier selected MODERATE_KPANIC. The primary LLM estimated paranoid CVSS 4.4 and the production ActionableScore returned 0 with recommendation Ordinary Moderate. The production rules then evaluated CVSS thresholds and ActionableScore thresholds. Recorded rule transitions were CVSS and initial triage stage MODERATE_KPANIC to MODERATE because the CVSS score, vector or bug class threshold rule was applied. ActionableScore thresholds stage MODERATE to LOW because the ActionableScore threshold rule was applied. final reconciliation stage LOW to LOW because KP specialist did not establish a qualifying panic and the cross model evidence reconciliation rule was applied. The legacy KP specialist has no review artifact for this run. The legacy FD or DH downgrade specialist has no review artifact for this run. Exact triggered rule markers were DECREASED_TO_MODERATE_REGULAR_FROM_MODERATE7_BASED_ON_GUESSCVSS and DECREASED_TO_LOW_FROM_MODERATE_BASED_ON_ACTIONABLESCORELESSTHAN3 and KPANIC_UNSUPPORTED_REMOVED and V64_AEGIS_UNSUPPORTED_PROMOTION_BLOCKED. The independent weaker Aegis shadow evaluated CVSS 2.2 and ActionableScore 1 with primitive class CLASS_C and manual review REQUIRED. Its independently reconciled impact was MODERATE_KPANIC. The models disagreed but the aggregation rules retained the production result. The final reported impact is LOW.';NO MANUAL CHECK; ,and ActionableScore result is Ordinary Moderate (with actual score 0)	MAYBE	KASAN NULLPTR LEAK HARDWARE LINUS MEMORY DECREASED_TO_MODERATE_REGULAR_FROM_MODERATE7_BASED_ON_GUESSCVSS DECREASED_TO_LOW_FROM_MODERATE_BASED_ON_ACTIONABLESCORELESSTHAN3 KPANIC_UNSUPPORTED_REMOVED V64_AEGIS_UNSUPPORTED_PROMOTION_BLOCKED	-	-	checked

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

ActionableScore=0
PrimitiveClass=CLASS_C
RealUAF=NO
HostAdminRequired=YES
NamespaceScopedPrivilege=NO
ImportantCandidate=NO
ManualReview=NO
AutoDowngradeAllowed=YES
RequiresHostUID0=UNKNOWN
AuthorizationDecisionBypass=NO
ProtectedResourceImpact=NONE
B4PendingReview=NO
B4PendingConfirmed=NO

1. ActionableScore

* Raw base score: -1.
* Conservative score: 0, applying the score-range lower bound.
* Raw paranoid score: -1.
* Paranoid score: 0, applying the same lower bound.
* Final recommended bucket: **Ordinary Moderate**::

2. Signal breakdown

* +1 Reliable kernel crash/strong DoS: the demonstrated scheduler lookup returns NULL, followed by unchecked dereferences in `rocket_job_run()` and `rocket_job_timedout()`.
* -2 Host-administrator-equivalent prerequisite: creating the faulty state requires unbinding a platform-device core; the supplied evidence identifies this as an administrative host operation.
* No UAF, OOB access/write, stale allocation, refcount mutation, attacker-controlled overwrite, or authorization bypass is shown.
* No extra availability point: the two faults are manifestations of the same lookup defect. The separate unsignaled-fence behavior remains explicitly unresolved and is not fixed by this patch.
* No rare-hardware or race penalty: RK3588-specific reproduction does not establish rare physical access, and no narrow timing race is required.
* NEW52 B5-R does not apply: there is no retry-accumulated heap OOB write or later consumption of corrupted structured state.
* NEW53/B3-PENDING and NEW54/B4-PENDING do not apply.

3. Reachability analysis

The complete demonstrated sequence requires a multi-core Rocket NPU, unbinding a non-final core, and jobs executing on a surviving higher-index core. Ordinary clients can submit jobs, but they are not shown able to perform the decisive unbind operation. No namespace-scoped capability path or delegated below-host-admin trigger is evidenced. Numeric UID 0 is not specifically established, although effective host-administrator power is.

4. Primitive and severity reading

The primitive is a deterministic NULL-pointer dereference caused by searching only `num_cores` entries after a non-compacting removal. It is Class C crash/DoS behavior, not active-object lifetime corruption. No useful non-crash corruption or trust-boundary primitive remains. Important candidacy is therefore unsupported, and automatic Moderate downgrade is safe.

5. One-sentence report phrase

Host-administrative removal of a middle Rocket NPU core can hide a still-running core from scheduler lookup and cause unchecked NULL dereferences in job and timeout workqueues.

6. Manual review advice

NO MANUAL CHECK NEEDED

The patch and call paths establish a privileged crash-only logic error without a material corruption or authorization primitive.

7. Call-site confidence

Call-site confidence: high

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

Patch: accel/rocket: search every core slot when looking up a scheduler
Commit: c137756469d5bcc79e97beb406ae4d14b9585463
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=c137756469d5bcc79e97beb406ae4d14b9585463

Commit description:

sched_to_core() walks rdev->cores[] up to rdev->num_cores, and
rocket_remove() decrements num_cores for every core it removes. Unbind a
core that is not the last one and the cores behind it fall outside the
search, so sched_to_core() returns NULL for a core that is still bound and
still running jobs. Neither caller checks the result:

  rocket_job_run():      rocket_fence_create(core), core->dev
  rocket_job_timedout(): dev_err(core->dev, "NPU job timed out")

Unbinding the middle core of the three on an RK3588 while three clients are
submitting to all of them faults twice, once from the surviving core's
job queue and once from its reset work:

  KASAN: null-ptr-deref in range [0x0000000000000220-0x0000000000000227]
  Workqueue: fdad0000.npu drm_sched_run_job_work [gpu_sched]
  pc : rocket_job_run+0x234/0x838 [rocket]
  Call trace:
   rocket_job_run+0x234/0x838 [rocket]
   drm_sched_run_job_work+0x2cc/0xad8 [gpu_sched]
   process_one_work+0x640/0x14f0

  KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
  Workqueue: rocket-reset-2 drm_sched_job_timedout [gpu_sched]
  pc : rocket_job_timedout+0xf0/0x1e0 [rocket]
  Call trace:
   rocket_job_timedout+0xf0/0x1e0 [rocket]
   drm_sched_job_timedout+0x188/0x6a0 [gpu_sched]

Both are the third core: the workqueue names are its device and its
core->index, and it was left at slot 2 while num_cores had dropped to 2.

Search all the slots that were allocated, the way find_core_for_dev() now
does. A core that is still bound is then found, and the two callers get
the pointer they already assume they have.

This does not make unbinding one core out of several safe. An open client
keeps an entity pointing at the scheduler of the core that went away:
drm_sched reports it as not ready for every job that lands on it, and the
client waits in dma_fence_default_wait for a fence that will never signal.
Stopping the NULL dereference is what belongs in a fix; the rest wants
more thought.

Reported-by: Sidong Yang <sidong.yang@furiosa.ai>
Closes: https://lore.kernel.org/dri-devel/apwUewaRnoTNXHCt@rock-5b-plus/
Fixes: 0810d5a ("accel/rocket: Add job submission IOCTL")
Cc: stable@vger.kernel.org
Signed-off-by: Igor Paunovic <royalnet026@gmail.com>
Assisted-by: LLM sparse checkpatch
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Link: https://patch.msgid.link/20260905150432.7477-1-royalnet026@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed files:
  drivers/accel/rocket/rocket_job.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=c137756469d5bcc79e97beb406ae4d14b9585463

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

The original Linux kernel CVE announcement for CVE-2026-99183 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-99183

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: Not available
  The Best / paranoid CVSS vector: AV:L/AC:L/PR:H/UI:N/S:U/C:N/I:N/A:H
  The Best / paranoid CVSS score: 4.4

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: 0

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).