[CVE-2026-89530][IMPORTANT] svcrdma: Reject inline replies that overflow the pull-up buffer

[AL-KERNEL <alkernel@kernelcve.org>]

CVE: CVE-2026-89530
Priority: IMPORTANT
AL-KERNEL base severity: IMPORTANT
KPANIC flag: YES
Patch: svcrdma: Reject inline replies that overflow the pull-up buffer
Commit: fcd91b9957462d398792201c239dffaeff1cc8b2
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=fcd91b9957462d398792201c239dffaeff1cc8b2
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89530
Analysis date: Fri, 11 Sep 2026 18:31:59 -0400
ActionableScore: 8
ActionableScore lower bound: 7
Actionable bucket: Strong Important candidate
Manual review required: YES

Summary:
No one-sentence report phrase was found.

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

The original Linux kernel CVE announcement for CVE-2026-89530 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89530

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: IMPORTANT
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-2026-89530	IMPORTANT	https://www.kernelcve.org/list/?q=CVE-2026-89530 CHECK WITH IMPACT FROM ORIG NN LOW	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:A/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H';*CWE-787;CWE-122;CWE-125;Other CVSS 'AV:A/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '8.8';DESCR 'An oversized inline reply in the RPC over RDMA server path can overflow sctxt sc_xprt_buf because svc_rdma_xb_linearize may linearize a reply that is larger than sc_max_req_size. A remote RDMA client can trigger this by requesting a large reply such as an NFS READ payload without providing a Write list or Reply chunk, forcing the SGE exhaustion pull up path. This can copy past the end of the pull up buffer and corrupt adjacent slab memory, and the oversized SGE length can also make the RDMA device read beyond the mapped region. For the CVSS the PR:N is used because no local account on the victim is required when the attacker can reach the RPC over RDMA service and issue protocol requests. The attack surface is network adjacent in typical deployments because RPC over RDMA is normally used on isolated storage or data center RDMA fabrics rather than the public Internet. Impact is at least denial of service via kernel crash, and the explicit slab memory overwrite plus device out of bounds read make confidentiality and integrity impact plausible in the paranoid assessment.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Strong Important candidate (with actual score 8)	YES	REMOTE NETWORK DMAorINTERRUPT HARDWARE LINUS KPANIC  KPANIC INCREASED_TO_HIGH_BASED_ON_GUESSCVSS	NO	NO	checked

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

ActionableScore=8
ActionableScoreLower=7
PrimitiveClass=CLASS_A
RealUAF=NO
HostAdminRequired=NO
NamespaceScopedPrivilege=NO
ImportantCandidate=YES
ManualReview=REQUIRED
AutoDowngradeAllowed=NO

1. ActionableScore

* Raw conservative score: 7
* Conservative score: 7. No floor or cap applied.
* Raw paranoid score: 8
* Paranoid score: 8. No floor or cap applied.
* Final recommended bucket: **Strong Important candidate**::

2. Signal breakdown

Conservative positive signals:

* +2 Strong corruption primitive. The patch description directly states that svc_rdma_xb_linearize() can copy past sctxt->sc_xprt_buf and corrupt adjacent slab memory.
* +2 Remote or network-adjacent protocol reachability. An RPC-over-RDMA client can trigger the path by requesting a large inline reply without providing a Write list or Reply chunk.
* +1 Reliable kernel crash or strong DoS. Adjacent slab corruption and invalid DMA/SGE state can realistically crash the kernel or destabilize the server.
* +1 Integrity plausible. The overflow writes beyond a kernel slab buffer into adjacent slab memory.
* +1 Confidentiality plausible. The oversized sc_sges[0].length is posted to the RDMA device, and the device can read beyond the mapped region, creating a plausible disclosure path.

Paranoid additional signal:

* +1 Memory layout invariant restoration. The fix restores the invariant between reply size, pull-up buffer size, SGE length, and DMA-mapped region size. This is an independent safety invariant because the bug has both a CPU copy overflow and an oversized device read side effect.

Material non-awarded signals:

* RealUAF not awarded. This is an out-of-bounds write and DMA out-of-bounds read issue, not a lifetime or stale-object bug.
* Controlled reclaim not demonstrated.
* Arbitrary write not demonstrated.
* Callback or function-pointer control not demonstrated.
* Local privilege escalation not demonstrated.
* Extra Availability not awarded beyond the same crash and corruption consequence.
* No race or timing penalty. The trigger is a protocol/state construction, not a narrow timing race.
* No rare-hardware penalty. RDMA is specialized and usually data-center scoped, but the vulnerable deployment is exactly an RPC-over-RDMA server exposure.

3. Reachability analysis

The attacker is an RPC-over-RDMA client that can reach the vulnerable server endpoint. No local account on the victim is required, so HostAdminRequired=NO. The practical attack surface is network-adjacent rather than Internet-wide in most deployments because RPC-over-RDMA and NFS/RDMA are normally used on isolated storage or data-center RDMA fabrics.

A prior administrator must have configured the RPC-over-RDMA service and RDMA-capable environment, but that is normal service exposure rather than a privilege required by the attacker after the service is reachable. Namespace-scoped privilege is not the relevant model here. The trigger does not depend on a hard race. The client controls the protocol shape by omitting the Write list or Reply chunk and requesting a reply large enough to exceed the pull-up buffer when the SGE-exhaustion path is selected.

4. Primitive and severity interpretation

The exact primitive is an attacker-triggerable oversized linearization of an RPC reply into sctxt->sc_xprt_buf, whose size is only sc_max_req_size. The vulnerable path can copy beyond that buffer and corrupt adjacent slab memory. The oversized length is then stored in the first SGE and posted, allowing the RDMA device to read beyond the mapped region.

This is active kernel memory corruption, not merely a warning, invalid state, or crash-only NULL dereference. It is also not a UAF and does not involve a concrete repeated release/refcount operation. The demonstrated consequence is kernel memory corruption with likely DoS. The speculative but technically defensible consequences include limited or high confidentiality and integrity impact depending on adjacent slab layout, payload influence, and whether the oversized device read exposes kernel memory to the client.

PrimitiveClass=CLASS_A because the patch directly establishes a kernel heap/slab out-of-bounds overwrite with attacker-controlled trigger conditions and a related out-of-bounds device read. ImportantCandidate=YES because this is network-adjacent kernel memory corruption in a server path. AutoDowngradeAllowed=NO because automatically treating this as ordinary DoS would discard a concrete slab corruption primitive and DMA read-beyond-mapping behavior.

5. One-sentence report phrase

An RPC-over-RDMA client can force oversized inline reply pull-up, causing svc_rdma_xb_linearize() to overflow sctxt->sc_xprt_buf, corrupt adjacent slab memory, and potentially make the RDMA device read beyond the mapped region.

6. Manual review recommendation

MANUAL CHECK REQUIRED

Reason: this is directly established network-adjacent kernel slab memory corruption with a plausible confidentiality side effect from the oversized posted SGE, even though arbitrary write, controlled reclaim, and LPE are not demonstrated.

7. Call-site confidence

Call-site confidence: high

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

Patch: svcrdma: Reject inline replies that overflow the pull-up buffer
Commit: fcd91b9957462d398792201c239dffaeff1cc8b2
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=fcd91b9957462d398792201c239dffaeff1cc8b2

Commit description:

An RPC-over-RDMA client can request a reply, such as an NFS READ
payload, without providing a Write list or a Reply chunk to carry
it. When such a reply needs more scatter/gather entries than the
device's Send Queue supports, svc_rdma_pull_up_needed() selects
pull-up and svc_rdma_pull_up_reply_msg() linearizes the whole
reply into sctxt->sc_xprt_buf. That buffer is only sc_max_req_size
bytes, while the reply on this path is bounded only by the client's
request, so svc_rdma_xb_linearize() copies past the end of the
buffer and corrupts adjacent slab memory. The oversized length is
then stored in sc_sges[0].length and posted, so the device also
reads beyond the mapped region.

The SGE-exhaustion branch is the only pull-up path that can exceed
the buffer: the threshold branch pulls up only replies smaller
than RPCRDMA_PULLUP_THRESH, and replies that fit the device's SGE
budget are sent directly without linearization. Make
svc_rdma_pull_up_needed() report -E2BIG when the reply it would
pull up cannot fit sc_max_req_size, and fail the request with
ERR_CHUNK as RFC 8166 Section 4.5.3 directs rather than dropping
the connection.

The helper no longer answers a simple yes/no question: it now
reports pull-up, no pull-up, or -E2BIG for a reply too large to
linearize. Rename svc_rdma_pull_up_needed() to
svc_rdma_check_pull_up() so its name no longer implies a boolean
predicate.

Fixes: e248aa7 ("svcrdma: Remove max_sge check at connect time")
Cc: stable@vger.kernel.org
Reported-by: Chris Mason <clm@meta.com>
Assisted-by: kres:claude-opus-4-7
Link: https://patch.msgid.link/20260623014728.826032-1-cel@kernel.org
Signed-off-by: Chuck Lever <cel@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed files:
  net/sunrpc/xprtrdma/svc_rdma_sendto.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=fcd91b9957462d398792201c239dffaeff1cc8b2

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

The original Linux kernel CVE announcement for CVE-2026-89530 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89530

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:L/I:L/A:H
  The Best / paranoid CVSS vector: AV:A/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
  The Best / paranoid CVSS score: 8.8

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: 7
  Paranoid ActionableScore: 8

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: IMPORTANT
KPANIC detected for this report: YES
Published priority for this report (same as in Subject): IMPORTANT

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).