[CVE-2026-74383][MODERATE 7.0] nvme-pci: fix out-of-bounds access in nvme_setup_descriptor_pools [ Upstream

[CVE-2026-74383][MODERATE 7.0] nvme-pci: fix out-of-bounds access in nvme_setup_descriptor_pools [ Upstream

[AL-KERNEL]

CVE: CVE-2026-74383
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: nvme-pci: fix out-of-bounds access in nvme_setup_descriptor_pools [ Upstream
Commit: 3e8aed5edaeeb237f97255d711aac6b5843fc059
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=3e8aed5edaeeb237f97255d711aac6b5843fc059
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-74383
Analysis date: Sat, 15 Aug 2026 20:21:56 -0400
ActionableScore: 5
ActionableScore lower bound: 3
Actionable bucket: Actionable Moderate at minimum
Manual review required: YES

Summary:
An out-of-bounds access in `nvme_setup_descriptor_pools()` on non-NUMA kernels can index `descriptor_pools` with `UINT_MAX` during NVMe namespace allocation, causing a reproducible kernel crash and failed device node creation, with practical impact mainly local DoS.

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

The original Linux kernel CVE announcement for CVE-2026-74383 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-74383

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-2026-74383	MODERATE	CHECK WITH IMPACT FROM ORIG NN MODERATE	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:H';CWE-129;CWE-681;*CWE-787;Other CVSS 'AV:L/AC:L/PR:H/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '6.6';DESCR 'An out of bounds access in nvme_setup_descriptor_pools occurs because NUMA_NO_NODE (-1) from hctx numa_node is passed through an unsigned parameter and becomes UINT_MAX. On CONFIG_NUMA n kernels this indexes dev descriptor_pools far beyond the nr_node_ids sized array during NVMe namespace allocation, which can fault the kernel and leave the namespace without a device node. For the CVSS the PR:H is used for the base score because reliable triggering usually requires administrative control over NVMe probe, driver bind, reset, or hotplug paths. For the CVSS the PR:L is used for the paranoid score only if a less privileged local actor can indirectly force NVMe reprobe in a delegated environment. The issue is not network reachable and is tied to local hardware or local driver control. Impact is at least denial of service via kernel crash. Because the primitive is an out of bounds kernel access, limited confidentiality or integrity impact is kept in the paranoid score and the case should be reviewed manually.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate at minimum (with actual score 4)  SKIP CVE-2026-74383 UNKNOWN SKIP The Fixes patch not applied yet, so unlikely that actual: d977506f8863807129d7a11f4057dfb1b38085ea YES  NO NO unknown 	MAYBE	OOB DANGER NVME DMAorINTERRUPT LINUS  DECREASED_TO_MODERATEREG_BASED_ON_ACTIONABLESCORELESSTHAN5	-	-	checked

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

ActionableScore=5
ActionableScoreLower=3

## 1. ActionableScore

* Conservative score: **3**
* Paranoid score: **5**
* Final recommended bucket: **Actionable Moderate at minimum**::

## 2. Signal breakdown

Conservative signals:

* **Memory corruption, Weak/indirect corruption candidate: +1**. `NUMA_NO_NODE` is converted through an unsigned parameter to `UINT_MAX`, causing an out-of-bounds index into `dev->descriptor_pools[]`.
* **Reliable kernel crash / strong DoS: +1**. The commit shows a reproducible kernel page fault during `nvme_alloc_ns()`.
* **Availability impact realistic: +1**. The NVMe namespace may fail to initialize and be left without a `/dev` node.
* **Important filesystem/storage path: +1**. NVMe namespace initialization is a relevant storage availability path.
* **Privileged kernel/device-management memory corruption path: +1**. The bug occurs in a trusted NVMe PCI driver initialization path involving descriptor/DMA pool setup.
* **Requires admin/root/device-control in typical deployments: -2**. Reliable triggering normally requires probe, hotplug, reset, bind/unbind, boot-time device initialization, or similar privileged/device-management control.

Conservative total: `1 + 1 + 1 + 1 + 1 - 2 = 3`

Paranoid adjustment:

* **Local lower-privilege trigger possibility: +1**. In delegated or virtualized environments, a less privileged local actor may indirectly trigger NVMe reprobe/reset behavior.
* **Privilege penalty reduced from -2 to -1**. This is only for the paranoid interpretation where the attacker is not full host root but can influence device lifecycle indirectly.

Paranoid total: `1 + 1 + 1 + 1 + 1 + 1 - 1 = 5`

Not awarded:

* No remote/network trigger.
* No firmware-mediated external influence.
* No strong LPE signal.
* No strong generic corruption primitive, because the index is fixed by `NUMA_NO_NODE` conversion and the demonstrated result is a fault, not attacker-controlled overwrite or reclaim.
* No Dirty-Pipe-like/page-cache ownership bypass.

## 3. Reachability analysis

The bug is reached when an NVMe controller is probed on a `CONFIG_NUMA=n` kernel. In that case `hctx->numa_node` is `NUMA_NO_NODE (-1)`, but the old helper accepted it as unsigned, producing an invalid `UINT_MAX` array index.

Typical reliable triggering requires local administrative or device-management control, such as booting with the affected NVMe device, hotplug, driver bind/unbind, reset, or reprobe. Namespaces and containers usually do not reduce this to unprivileged access unless the environment delegates device lifecycle control. The affected path is not network reachable.

Call-site confidence: **high**, because the commit identifies the direct caller `nvme_init_hctx_common()` and the failing path through `nvme_alloc_ns()`.

## 4. Severity interpretation

This is stronger than an ordinary low-risk correctness bug because it is a concrete kernel out-of-bounds access in a storage driver and has a demonstrated crash. However, realistic exploitation evidence points mainly to local DoS and failed NVMe namespace initialization.

The theoretical memory corruption concern is limited. The invalid index can point outside the intended array, but the value is not finely attacker-controlled and the reproduced behavior is a page fault. Therefore this is best treated as **borderline/manual-review conservative** and **Actionable Moderate in paranoid triage**, not a strong Important candidate without more evidence.

## 5. One-sentence report phrase

An out-of-bounds access in `nvme_setup_descriptor_pools()` on non-NUMA kernels can index `descriptor_pools` with `UINT_MAX` during NVMe namespace allocation, causing a reproducible kernel crash and failed device node creation, with practical impact mainly local DoS.

## 6. Manual review recommendation

**MANUAL CHECK REQUIRED**

Reason: this is a concrete kernel out-of-bounds access in an important storage path. The demonstrated impact is DoS, but the memory-access class and driver initialization context justify manual review before auto-closing.

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

Patch: nvme-pci: fix out-of-bounds access in nvme_setup_descriptor_pools [ Upstream
Commit: 3e8aed5edaeeb237f97255d711aac6b5843fc059
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=3e8aed5edaeeb237f97255d711aac6b5843fc059

Changed files:
  drivers/nvme/host/pci.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=3e8aed5edaeeb237f97255d711aac6b5843fc059

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

The original Linux kernel CVE announcement for CVE-2026-74383 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-74383

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:H/UI:N/S:U/C:N/I:N/A:H
  The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:H
  The Best / paranoid CVSS score: 6.6

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: 3
  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: NO
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).