[CVE-2026-72466][MODERATE 7.0] xprtrdma: Fix bcall rep leak and unbounded peek [ Upstream

[CVE-2026-72466][MODERATE 7.0] xprtrdma: Fix bcall rep leak and unbounded peek [ Upstream

[AL-KERNEL]

CVE: CVE-2026-72466
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: xprtrdma: Fix bcall rep leak and unbounded peek [ Upstream
Commit: 0cee8f9c3b14bd6dee9c4310090a7f45b89b834f
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=0cee8f9c3b14bd6dee9c4310090a7f45b89b834f
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-72466
Analysis date: Sat, 15 Aug 2026 08:34:58 -0400
ActionableScore: 6
ActionableScore lower bound: 4
Actionable bucket: Strong Important candidate / Actionable Moderate at minimum**:: Paranoid handling is Actionable Moderate. Strong Important is not proven because there is no write primitive, UAF reclaim, object replacement, callback control, or demonstrated privilege escalation.
Manual review required: YES

Summary:
A short or malformed RPC/RDMA reply can cause `rpcrdma_is_bcall()` to read past the real wire payload, misclassify the frame, leak a DMA-mapped receive buffer, and drain the Receive queue, leading primarily to network-adjacent denial of service for NFS/RDMA transports.

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

The original Linux kernel CVE announcement for CVE-2026-72466 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-72466

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-2026-72466	MODERATE	CHECK WITH IMPACT FROM ORIG NN IMPORTANT	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:A/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:H';CWE-125;*CWE-401;**CWE-400;CWE-703;Other CVSS 'AV:A/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '7.6';DESCR 'A short or malformed RPC/RDMA reply can make rpcrdma_is_bcall read up to 5 XDR words past the real wire payload into stale receive buffer contents and misclassify the frame as a backchannel call. A second short header path can return true without transferring the rep to rpcrdma_bc_receive_call, so rpcrdma_reply_handler skips normal cleanup and receive reposting. This can orphan a DMA mapped receive buffer and drain the Receive queue, causing RNR NAKs and denial of service for the RPC/RDMA transport. For the CVSS the PR:N is used because no local privileges are required on the victim when the attacker can act as or tamper with an RPC/RDMA peer. The issue is network adjacent rather than general Internet reachable in typical deployments because NFS over RDMA is usually used inside trusted storage or data center fabrics. Impact is primarily availability loss, with paranoid C:L and I:L due to the bounded out-of-bounds read and protocol state confusion, but no strong evidence of arbitrary write or privilege escalation.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Strong Important candidate / Actionable Moderate at minimum (with actual score 5)	YES	REMOTE LEAK UAF NETWORK  DECREASED_TO_MODERATE70_BASED_ON_ACTIONABLESCORELOWERTHAN6 DECREASED_TO_MODERATEREG_BASED_ON_FALSEPOSCHECKOFKP	NO	NO	checked

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

ActionableScore=6
ActionableScoreLower=4

## 1. ActionableScore

* Conservative score: 4
* Paranoid score: 6
* Final recommended bucket: **Strong Important candidate / Actionable Moderate at minimum**:: Paranoid handling is Actionable Moderate. Strong Important is not proven because there is no write primitive, UAF reclaim, object replacement, callback control, or demonstrated privilege escalation.

## 2. Signal breakdown

Conservative signals:

* Remote reachable / network-triggerable: +2
  A malformed RPC/RDMA reply from a peer can trigger the affected decode path. This is network-adjacent rather than public Internet exposure, but still externally triggerable over the RDMA transport.
* Reliable kernel crash / strong DoS: +1
  The receive buffer can be orphaned and receives are not reposted, allowing the Receive queue to drain and causing RNR NAKs / transport failure.
* Availability impact realistic: +1
  The impact is not just a transient warning. It can degrade or break RPC/RDMA transport availability until recovery or teardown.
* Important filesystem/storage path: +1
  This affects RPC/RDMA, commonly used for NFS over RDMA storage traffic.
* Rare AND difficult-to-reach subsystem/configuration: -1
  NFS over RDMA requires RDMA-capable deployment and is usually limited to storage or data-center fabrics.

Conservative total: 4

Paranoid additional signals:

* Confidentiality impact plausible: +1
  The bounded OOB read can inspect stale receive-buffer contents, although no direct user-visible disclosure primitive is shown.
* Integrity impact plausible: +1
  Stale data can influence protocol classification as a backchannel call, creating protocol state confusion, although no arbitrary write or privilege escalation is demonstrated.

Paranoid total: 6

Not awarded:

* Generic strong memory corruption: +0
  The patch shows bounded OOB read and resource/ownership leak behavior, not OOB write, UAF reclaim, double free, type confusion, or arbitrary write.
* Privilege escalation plausible: +0
  No concrete LPE primitive is shown.
* Dirty-Pipe-like / COW ownership violation: +0
  No page-cache or shared-file ownership bypass is involved.

## 3. Reachability analysis

A malicious or compromised RPC/RDMA peer, or an attacker able to tamper with RPC/RDMA traffic, can send a short or malformed reply that reaches `rpcrdma_is_bcall()`. No local privileges are required on the victim for that network-adjacent scenario, so PR is effectively none for the transport peer case. This is not a general Internet-facing path in typical deployments because NFS over RDMA is normally used inside trusted storage or data-center networks. Namespace or container delegation is not the main factor here. The main exposure condition is whether RPC/RDMA / NFS over RDMA is enabled and reachable by untrusted or semi-trusted peers.

Call-site confidence: high. The commit describes the relevant caller contract in `rpcrdma_reply_handler()`, including the bare return that skips `rpcrdma_rep_put()` and `rpcrdma_post_recvs()`.

## 4. Severity interpretation

This behaves above an ordinary Moderate because it is externally triggerable in RPC/RDMA deployments and can cause a persistent transport-level DoS by leaking receive buffers and draining the Receive queue. It does not reach a clear Important-class memory corruption case because the OOB access is read-only and bounded, and the patch does not show attacker-controlled overwrite, reclaim, type confusion, callback hijack, or privilege escalation. The conservative interpretation is Borderline / manual review recommended. The paranoid interpretation is Actionable Moderate and should not be auto-closed.

## 5. One-sentence report phrase

A short or malformed RPC/RDMA reply can cause `rpcrdma_is_bcall()` to read past the real wire payload, misclassify the frame, leak a DMA-mapped receive buffer, and drain the Receive queue, leading primarily to network-adjacent denial of service for NFS/RDMA transports.

## 6. Manual review recommendation

MANUAL CHECK REQUIRED

Reason: the issue is network-adjacent, affects a storage transport path, includes bounded OOB read plus receive-buffer lifetime/ownership mishandling, and has a realistic DoS impact even though privilege escalation is not currently supported by the patch evidence.

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

Patch: xprtrdma: Fix bcall rep leak and unbounded peek [ Upstream
Commit: 0cee8f9c3b14bd6dee9c4310090a7f45b89b834f
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=0cee8f9c3b14bd6dee9c4310090a7f45b89b834f

Changed files:
  net/sunrpc/xprtrdma/rpc_rdma.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=0cee8f9c3b14bd6dee9c4310090a7f45b89b834f

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

The original Linux kernel CVE announcement for CVE-2026-72466 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-72466

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:A/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
  The Best / paranoid CVSS vector: AV:A/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:H
  The Best / paranoid CVSS score: 7.6

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: 4
  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: 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).