[CVE-2026-98070][MODERATE 7.0] net/rds: acquire RDS_IN_XMIT in rds_tcp_reset_callbacks() [ Upstream

[CVE-2026-98070][MODERATE 7.0] net/rds: acquire RDS_IN_XMIT in rds_tcp_reset_callbacks() [ Upstream

[AL-KERNEL]

CVE: CVE-2026-98070
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: net/rds: acquire RDS_IN_XMIT in rds_tcp_reset_callbacks() [ Upstream
Commit: d625112564c3e980e02504270222b49b82690cee
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=d625112564c3e980e02504270222b49b82690cee
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98070
Analysis date: Sat, 26 Sep 2026 16:26:06 -0400
ActionableScore: 5
ActionableScore lower bound: 4
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-2026-98070 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98070

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-2026-98070	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-98070 CHECK WITH IMPACT FROM ORIG NN MODERATE	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-362;*CWE-416;CWE-667;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '7.5';DESCR 'A race in rds_tcp_reset_callbacks can allow rds_send_xmit to reacquire RDS_IN_XMIT immediately after the reset path observes it as clear. On weakly ordered systems the sender and resetter can miss each others state updates, allowing transmission to run concurrently with the socket swap and rds_send_path_reset while cp_xmit state is being rewritten. The reset path also cached tc t_sock before waiting, so concurrent teardown could release the socket and leave a stale pointer that may create a use after free condition. The demonstrated impact is at least denial of service through kernel corruption or a crash. In the paranoid interpretation the stale socket reference and concurrent state modification may also affect confidentiality and integrity and require manual review for privilege escalation potential. For the CVSS the PR:N value is used because an adjacent RDS TCP peer may influence simultaneous connection establishment and a duelling SYN without credentials on the target host. This network scenario requires RDS TCP to be enabled and reachable, an overlapping connection or teardown event, and successful race timing. RDS TCP is normally used inside trusted cluster or data center networks rather than exposed to the public Internet.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 5)  SKIP CVE-2026-98070 UNKNOWN SKIP No affected files built, so skip this CVE NO  - - unknown 	MAYBE	REMOTE LOCK DEADLOCK INIT NETWORK HARDWARE LINUS KPANIC  KPANIC INCREASED_TO_HIGH_BASED_ON_GUESSCVSS DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFDH	-	-	checked

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

ActionableScore=5
ActionableScoreLower=4
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=UNKNOWN
NamespaceScopedPrivilege=UNKNOWN
ImportantCandidate=YES
ManualReview=REQUIRED
AutoDowngradeAllowed=NO

1. ActionableScore

* Raw conservative score: 3
* Conservative score: 4
* Applied conservative floor: 3 -> 4 under the Real UAF floor for a stale active socket reference used outside its valid lifetime.
* Raw paranoid score: 3
* Paranoid score: 5
* Applied paranoid floor: 3 -> 5 because the stale active socket is consumed during callback restoration and socket replacement while transmit state is concurrently rewritten.
* Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**::

2. Signal breakdown

* +2 Real UAF or lifetime violation. `tc->t_sock` is cached before waiting, while concurrent teardown can release the socket and clear `t_sock`. The cached `osock` can therefore outlive the socket.
* +1 Active-state synchronization violation. Sampling `RDS_IN_XMIT` does not acquire it, allowing `rds_send_xmit()` to overlap the socket swap and `rds_send_path_reset()` while `cp_xmit_*` state is rewritten.
* +1 Realistic kernel crash or strong DoS. A stale socket consumed by callback restoration or concurrent modification of transmit bookkeeping can plausibly cause an invalid dereference or fatal state corruption.
* -1 Material timing and architecture dependency. The primary race requires an overlap involving reset, transmit, or teardown and is specifically enabled by store-buffering behavior on weakly ordered architectures.

Important non-awarded signals:

* Direct remote reachability is not established. A peer may influence connection timing, but the complete trigger also depends on local RDS connection state and concurrent kernel activity.
* Local unprivileged reachability is not established from the supplied call path.
* Controlled reclaim, attacker-shaped replacement contents, arbitrary write, disclosure, and callback hijacking are not demonstrated.
* Structured stale-object leverage is not separately awarded because attacker influence over reclaimed socket storage or replacement contents is not established.
* Confidentiality and integrity points are not awarded from the speculative CVSS worst case.
* No extra Availability point is awarded because it would duplicate the same crash or corruption consequence.
* There is no concrete repeated release or refcount operation on the exact same dangling reference. The dedicated repeated-release signal and its paranoid bonus do not apply.

3. Reachability analysis

The reset path is reached while resolving a duelling RDS TCP connection and overlaps with transmit or teardown activity. A network peer can influence connection establishment and SYN timing, but the evidence does not establish that arbitrary traffic alone can create the complete vulnerable state without an existing RDS configuration and local connection activity.

The supplied code does not resolve whether an ordinary local user can create and drive the required RDS TCP connection, whether a capability is required, or which namespace would govern such a check. Therefore `HostAdminRequired` and `NamespaceScopedPrivilege` remain `UNKNOWN`, with no privilege penalty and no unprivileged-trigger bonus.

The race is timing-sensitive and particularly relevant to weakly ordered architectures. It may be repeatable through connection churn, but attacker control over both racing paths is not established.

4. Primitive and severity interpretation

The primary primitive is an active structured-object lifetime violation: a cached `struct socket` pointer can become stale after teardown releases the underlying socket. The stale pointer is subsequently used in a broader socket callback replacement path rather than merely read for diagnostics.

A second, related primitive is failure to acquire the transmit exclusion bit. This allows active transmission to overlap `rds_send_path_reset()` while transmit bookkeeping is being rewritten. The fix changes observation of the bit into ownership held across the entire socket swap and reset operation.

This is CLASS_B rather than CLASS_A. A real UAF candidate and active-state synchronization failure are established, but controlled reclaim, directed corruption, arbitrary read or write, and privilege escalation are not.

The conservative interpretation supports a crash or DoS capable active-object UAF. The paranoid interpretation preserves the issue at Actionable Moderate because stale socket processing and concurrent state mutation leave meaningful exploitation leverage unresolved. This makes it an Important candidate for review, but not a demonstrated Important vulnerability.

`AutoDowngradeAllowed=NO` because automatically reducing the issue to an ordinary race or crash would discard the concrete stale active socket lifetime violation and concurrent mutation of live transmit state.

5. One-sentence report phrase

A synchronization race in RDS TCP can permit transmit processing to overlap socket replacement and path reset while a socket pointer released by concurrent teardown remains cached, creating a host kernel UAF and crash risk with unresolved exploitation leverage.

6. Manual review recommendation

MANUAL CHECK REQUIRED

The stale active socket is used during callback and connection-state manipulation, and the available patch context does not resolve whether reuse can produce effects beyond a kernel crash.

7. Call-site confidence

Call-site confidence: medium

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

Patch: net/rds: acquire RDS_IN_XMIT in rds_tcp_reset_callbacks() [ Upstream
Commit: d625112564c3e980e02504270222b49b82690cee
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=d625112564c3e980e02504270222b49b82690cee

Changed files:
  net/rds/tcp.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=d625112564c3e980e02504270222b49b82690cee

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

The original Linux kernel CVE announcement for CVE-2026-98070 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98070

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:N/I:N/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: 4
  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: 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).