[CVE-2026-72473][MODERATE 7.0] xprtrdma: Decouple req recycling from RPC completion [ Upstream

[CVE-2026-72473][MODERATE 7.0] xprtrdma: Decouple req recycling from RPC completion [ Upstream

[AL-KERNEL]

CVE: CVE-2026-72473
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: xprtrdma: Decouple req recycling from RPC completion [ Upstream
Commit: 740975054a1970c0cf15f70ac39724a064f45847
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=740975054a1970c0cf15f70ac39724a064f45847
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-72473
Analysis date: Sun, 16 Aug 2026 01:03:47 -0400
ActionableScore: 7
ActionableScore lower bound: 6
Actionable bucket: Strong Important candidate / Actionable Moderate at minimum
Manual review required: YES

Summary:
A lifetime bug in xprtrdma can recycle an rpcrdma_req before RDMA Send completion, allowing the HCA to DMA-read from a reused send buffer and potentially causing crash, RPC state corruption, or data exposure on NFS/RDMA deployments.

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

The original Linux kernel CVE announcement for CVE-2026-72473 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-72473

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-2026-72473	MODERATE	CHECK WITH IMPACT FROM ORIG NN LOW	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:A/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H';*CWE-416;CWE-362;*CWE-672;Other CVSS 'AV:A/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '7.5';DESCR 'A lifetime race in xprtrdma can recycle an rpcrdma_req before the RDMA Send side has completed. This can happen when a Reply completes the RPC and drops the shared rl_kref while the HCA may still be DMA reading from the Send buffer, especially for Sends that use only pre registered buffers and do not need DMA unmap through sc_unmap_count. This is a DMA use-after-free style condition in the RPC/RDMA transport and can lead to kernel crash, corrupted RPC state, or exposure of data that is read from a reused Send buffer. For the CVSS the PR:N value is used for the paranoid score because a malicious or compromised RPC/RDMA peer on the adjacent RDMA fabric may be able to influence Reply timing without privileges on the victim host. The issue is not Internet remote in typical deployments and usually requires access to an internal RDMA or storage network. Impact is at least denial of service and may include confidentiality or integrity impact due to the UAF lifetime primitive, so manual review is recommended.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Strong Important candidate / Actionable Moderate at minimum (with actual score 7)	YES	REMOTE LOCK INIT LEAK UAF NETWORK DMAorINTERRUPT LOG KPANIC  KPANIC INCREASED_TO_HIGH_BASED_ON_GUESSCVSS DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFKP	NO	NO	checked

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

ActionableScore=7
ActionableScoreLower=6

## 1. ActionableScore

* Conservative score: 6
* Paranoid score: 7
* Final recommended bucket: **Strong Important candidate / Actionable Moderate at minimum**::

## 2. Signal breakdown

Triggered signals:

* Generic memory corruption, strong corruption primitive: +2
  The patch fixes a real lifetime bug where `rpcrdma_req` can be returned to the free pool while the HCA may still be DMA-reading from its send buffer.

* Real lifetime corruption: +1
  This is a refcount and ownership lifetime split bug between RPC completion and RDMA Send completion.

* Reliable kernel crash / strong DoS: +1
  Premature request recycling can corrupt RPC/RDMA state and plausibly lead to crashes or connection failure.

* Local unprivileged trigger: +1
  If an NFS/RDMA mount already exists, ordinary local file operations may generate RPC/RDMA traffic without requiring mount privileges.

* Confidentiality impact plausible: +1
  Because the HCA may DMA-read from a buffer after request recycling, reused send-buffer contents could plausibly be exposed to the RDMA peer.

* Integrity impact plausible: +1
  Reuse of the send buffer before DMA completion can plausibly corrupt outgoing RPC/RDMA message contents.

* Important filesystem/storage path: +1
  The affected code is in the NFS RPC-over-RDMA transport path, which is a storage/filesystem data path.

Subtracted signals:

* Hard or unreliable race / special timing required: -1
  The condition depends on Reply completion and RDMA Send completion ordering.

* Rare AND difficult-to-reach subsystem/configuration: -1
  NFS over RDMA requires RDMA-capable infrastructure and is not a default general-purpose network path.

Paranoid-only interpretation:

* Adjacent network-triggerable RDMA peer influence: +2 instead of relying only on local triggering
  A malicious or compromised RPC/RDMA peer on the RDMA fabric may influence Reply timing, but this is adjacent/internal fabric exposure rather than Internet remote exposure.

## 3. Reachability analysis

The bug is reachable in systems using RPC-over-RDMA, typically NFS/RDMA. A local unprivileged user may be able to trigger RPC activity through normal file access if the RDMA-backed filesystem is already mounted. Mount setup itself usually requires administrator privileges, but triggering traffic after setup does not necessarily require them.

A remote Internet attacker is not a realistic model. The more plausible external model is a malicious or compromised peer on an internal RDMA or storage fabric. Containers and namespaces do not normally make this broadly reachable unless the container has access to the mounted filesystem or delegated storage paths.

Call-site confidence: high. The patch directly shows the refcount split, request recycling, sendctx completion, teardown drain, and release paths.

## 4. Severity interpretation

This is stronger than an ordinary Moderate because it is not a simple leak, warning, or validation fix. The patch corrects a real object lifetime mismatch where hardware may still access a request buffer after the software has made the request reusable.

Realistic impact is most confidently DoS or corrupted RPC/RDMA state. However, because the primitive involves DMA reading from a recycled send buffer, confidentiality and integrity impact are plausible enough to require manual review. This should be treated as Actionable Moderate at minimum and a Strong Important candidate in paranoid triage.

## 5. One-sentence report phrase

A lifetime bug in xprtrdma can recycle an rpcrdma_req before RDMA Send completion, allowing the HCA to DMA-read from a reused send buffer and potentially causing crash, RPC state corruption, or data exposure on NFS/RDMA deployments.

## 6. Manual review recommendation

MANUAL CHECK REQUIRED

Reason: this is a real refcount/lifetime bug in a storage transport with DMA access after request recycling. The exploitation path is configuration-dependent and timing-sensitive, but the primitive is serious enough that it should not be auto-closed.

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

Patch: xprtrdma: Decouple req recycling from RPC completion [ Upstream
Commit: 740975054a1970c0cf15f70ac39724a064f45847
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=740975054a1970c0cf15f70ac39724a064f45847

Changed files:
  net/sunrpc/xprtrdma/backchannel.c
  net/sunrpc/xprtrdma/rpc_rdma.c
  net/sunrpc/xprtrdma/transport.c
  net/sunrpc/xprtrdma/verbs.c
  net/sunrpc/xprtrdma/xprt_rdma.h

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=740975054a1970c0cf15f70ac39724a064f45847

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

The original Linux kernel CVE announcement for CVE-2026-72473 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-72473

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:H/PR:L/UI:N/S:U/C:L/I:L/A:H
  The Best / paranoid CVSS vector: AV:A/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H
  The Best / paranoid CVSS score: 7.5

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: 6
  Paranoid ActionableScore: 7

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 alexanjelausa@gmail.com (and both
send reply to CVE record itself too and see "reply" button below for howto reply).