[CVE-2026-92510][MODERATE 7.0] RDMA/core: Fix potential use after free in ib_destroy_srq_user() [ Upstream

[CVE-2026-92510][MODERATE 7.0] RDMA/core: Fix potential use after free in ib_destroy_srq_user() [ Upstream

[AL-KERNEL]

CVE: CVE-2026-92510
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: RDMA/core: Fix potential use after free in ib_destroy_srq_user() [ Upstream
Commit: a5bfd9fe05d5cebbd3e447bf7f7b2c663007d8b9
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=a5bfd9fe05d5cebbd3e447bf7f7b2c663007d8b9
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-92510
Analysis date: Thu, 17 Sep 2026 22:42:52 -0400
ActionableScore: 6
ActionableScore lower bound: 5
Actionable bucket: Actionable Moderate / Strong Important candidate requiring manual review
Manual review required: YES

Summary:
No one-sentence report phrase was found.

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

The original Linux kernel CVE announcement for CVE-2026-92510 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-92510

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-2026-92510	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-92510 CHECK WITH IMPACT FROM ORIG NN MODERATE	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H';*CWE-416;CWE-362;*CWE-664;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '7';DESCR 'A race in RDMA SRQ destruction can leave an SRQ reachable through restrack while vendor specific resources are already being destroyed. A local actor can race ib_destroy_srq_user() with the RDMA netlink restrack access path, where rdma_restrack_get() is the main synchronization mechanism for that object. For the CVSS the PR:L value is used because reliable triggering requires local code execution with access to RDMA uverbs or RDMA netlink resource interfaces, but does not require full administrator privileges in deployments that grant RDMA device access to users. The issue is not directly network reachable through RDMA traffic alone and is triggered through local control plane object lifetime operations. Impact is at least local denial of service via kernel crash. In the paranoid interpretation, the race exposes a use-after-free candidate in kernel object lifetime handling and may allow confidentiality or integrity impact, so manual review is required.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 6)	YES	INIT UAF NETWORK LINUS KPANIC  KPANIC INCREASED_TO_HIGH_BASED_ON_GUESSCVSS DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFKP	NO	NO	checked

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

ActionableScore=6
ActionableScoreLower=5
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=UNKNOWN
NamespaceScopedPrivilege=UNKNOWN
ImportantCandidate=YES
ManualReview=REQUIRED
AutoDowngradeAllowed=NO

1. ActionableScore

* Raw conservative score: 5
* Conservative score: 5. No floor or cap changed the score.
* Raw paranoid score: 6
* Paranoid score: 6. No floor or cap changed the score.
* Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**::

2. Signal breakdown

Conservative positive signals:

* +2 Real UAF / active lifetime violation: the commit directly states that SRQ remains accessible through restrack after vendor-specific SRQ resources may already be freed, creating a potential use-after-free window.
* +1 Reliable kernel crash / strong DoS: stale access to an SRQ or its vendor-specific resources during destruction can plausibly fault the kernel.
* +1 Secondary index / registration-table staleness: restrack is a resource tracking index through which netlink users can still discover the SRQ while its internal resources are being destroyed.
* +1 Attacker-influenced lifetime manipulation: a local actor with RDMA object access can create and destroy SRQ objects and can attempt to race this with the restrack observation path.

Conservative negative signals:

* No -2 host-admin penalty: host-admin-equivalent privilege is not positively established from the supplied evidence.
* No -1 race penalty: the bug is a race, but repeated SRQ create/destroy and restrack access attempts appear feasible for an actor with the relevant RDMA access. The prompt requires more than the word race or a short window to subtract.
* No rare-hardware penalty: RDMA is specialized, but the patch does not establish a physical-only or rare-hardware-only security condition sufficient for a scoring reduction.

Paranoid additional signal:

* +1 Structured stale-object exploitation leverage: the stale object is an active structured RDMA object exposed through restrack, and the stale path may consume object state or associated vendor resources after their intended lifetime. This is not enough for CLASS_A, but it leaves meaningful exploitation leverage unresolved.

Important non-awarded signals:

* Controlled reclaim not demonstrated.
* Arbitrary kernel read or write not demonstrated.
* Callback or function-pointer control not demonstrated.
* Confidentiality and integrity impact are not directly awarded as independent ActionableScore points.
* Remote reachability not established. This is local RDMA control plane and netlink lifetime behavior, not direct RDMA packet-triggered exploitation.
* Concrete repeated release/refcount signal not awarded. The patch shows premature visibility through restrack while destruction proceeds, but it does not show first put then exact same surviving reference then second put on that same dangling object.
* Weak LPE was not separately awarded because the unresolved concern is already represented by structured stale-object leverage in the paranoid score.
* Extra availability was not awarded because the same plausible kernel crash is already counted once.

Repeated release/refcount checklist:

* First release can end lifetime: not established as a first put on the exact same reference.
* Exact same reference survives to a second lifetime mutation: not established.
* Second operation is a real lifetime/refcount mutation: not established.
* Concrete repeated release/refcount +1 awarded: no.
* Paranoid repeated-refcount leverage +1 awarded: no.
* FLOOR 5 applied: no explicit floor was needed. The conservative raw score already reaches 5.

3. Reachability analysis

The apparent attacker is a local actor with access to RDMA userspace resources and the RDMA netlink restrack path. Creating and destroying an SRQ is normally associated with RDMA uverbs object management, while the conflicting observation path is netlink resource tracking. The supplied evidence does not prove that host root or host-admin-equivalent privilege is required for the complete trigger, but it also does not prove ordinary unprivileged reachability on all systems.

HostAdminRequired is therefore UNKNOWN. NamespaceScopedPrivilege is also UNKNOWN because the supplied context does not establish that all relevant RDMA object creation, destruction, and restrack access checks are fully satisfiable inside attacker-created namespaces.

There is likely prior environmental setup in the practical sense: RDMA support and an RDMA-capable device or equivalent provider must exist. This is not scored as a rare-hardware penalty because the patch does not establish an unusually rare or physical-only condition. The race window is described as short, but the object lifecycle operations appear repeatable, so a hard-race penalty is not applied.

4. Primitive and severity interpretation

The exact primitive is a lifetime ordering bug in RDMA SRQ destruction. Before the fix, rdma_restrack_del() occurred late in ib_destroy_srq_user(), after srq->device->ops.destroy_srq() could have released vendor-specific resources. During that interval, the SRQ could still be found through restrack, where rdma_restrack_get() is the synchronization mechanism for netlink access. This directly supports RealUAF=YES.

This is an active structured object case, not fixed stale storage. The stale object is an RDMA SRQ with associated device/vendor resources and resource-tracking visibility. The demonstrated consequence is at least local kernel crash or DoS. The speculative consequence is stronger exploitation through stale structured-object access, but controlled reclaim, arbitrary write, or demonstrated privilege escalation are not established.

PrimitiveClass is CLASS_B because this is a real kernel lifetime/UAF primitive with unresolved exploitation leverage, but not a demonstrated strong corruption primitive. ImportantCandidate is YES because the SRQ remains externally reachable through a secondary tracking index while internal resources may be freed, and the resulting active-object UAF should not be treated as ordinary crash-only cleanup. AutoDowngradeAllowed is NO because automatic downgrade would discard a concrete unresolved kernel lifetime primitive.

Likely operational severity is Actionable Moderate at minimum, with Strong Important review warranted if analysis confirms below-host-admin reachability or meaningful stale structured-field consumption.

5. One-sentence report phrase

A race in RDMA SRQ destruction can leave an SRQ visible through restrack after vendor-specific resources may have been freed, creating a local active-object UAF risk with at least kernel crash impact and unresolved exploitation leverage.

6. Manual review recommendation

MANUAL CHECK REQUIRED

Reason: the patch explicitly fixes a potential UAF in RDMA object lifetime ordering, involving an active structured object and a secondary restrack visibility path. Controlled reclaim or LPE is not demonstrated, but the primitive is strong enough that automatic closure as ordinary DoS would be unsafe.

7. Call-site confidence

Call-site confidence: medium

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

Patch: RDMA/core: Fix potential use after free in ib_destroy_srq_user() [ Upstream
Commit: a5bfd9fe05d5cebbd3e447bf7f7b2c663007d8b9
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=a5bfd9fe05d5cebbd3e447bf7f7b2c663007d8b9

Changed files:
  drivers/infiniband/core/verbs.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=a5bfd9fe05d5cebbd3e447bf7f7b2c663007d8b9

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

The original Linux kernel CVE announcement for CVE-2026-92510 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-92510

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:H/PR:L/UI:N/S:U/C:L/I:L/A:H
  The Best / paranoid CVSS vector: AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H
  The Best / paranoid CVSS score: 7

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: 5
  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: 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).