* [CVE-2026-97955][MODERATE 7.0] net: mana: restore the XDP program pointer when pre-allocation fails [ Upstream
@ 2026-09-26 20:21 21% AL-KERNEL
0 siblings, 0 replies; 1+ results
From: AL-KERNEL @ 2026-09-26 20:21 UTC (permalink / raw)
To: kernel-cve
CVE: CVE-2026-97955
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: net: mana: restore the XDP program pointer when pre-allocation fails [ Upstream
Commit: 5b3cdc9abd82b7e49e0f1a21d6eb75466c5ecef8
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=5b3cdc9abd82b7e49e0f1a21d6eb75466c5ecef8
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97955
Analysis date: Sat, 26 Sep 2026 16:21:05 -0400
ActionableScore: 5
ActionableScore lower bound: unknown
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-2026-97955 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97955
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-2026-97955 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-97955 CHECK WITH IMPACT FROM ORIG NN LOW 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-664;*CWE-672;Other CVSS 'AV:L/AC:H/PR:H/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '7';DESCR 'A use-after-free in the Microsoft MANA XDP configuration path can occur when mana_xdp_set publishes a new bpf_prog pointer before RX buffer pre-allocation completes. If pre-allocation fails the XDP attach is reported as failed and the BPF core can free the program while apc->bpf_prog still references it. A later interface bring up or TX timeout reset can consume the stale pointer and call bpf_prog_add() on freed memory. This can at least cause a local denial of service and may permit confidentiality integrity or privilege impact if freed memory is reclaimed in a controllable way. For the CVSS the PR:L paranoid score assumes a local user or service account with delegated ability to configure an interface and attach an XDP program. In standard host deployments attaching XDP to a physical interface normally requires CAP_NET_ADMIN. The issue is not directly network reachable. Triggering also requires an RX buffer allocation failure followed by a later queue reconfiguration or reset. Impact is at least a local denial of service via use-after-free and in the worst case may allow broader impact due to memory corruption.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 5) YES REMOTE BPF ERRORPATH UAF NETWORK KPANIC INCREASED_FROM_LOW_BASED_ON_REQUIREMANUALCHECK YES NO checked
======================================================================
ACTIONABLESCORE ANALYSIS
======================================================================
ActionableScore=5
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=UNKNOWN
NamespaceScopedPrivilege=UNKNOWN
ImportantCandidate=YES
ManualReview=REQUIRED
AutoDowngradeAllowed=NO
1. ActionableScore
* Raw conservative score: 4
* Conservative score: 5
* Raw paranoid score: 5
* Paranoid score: 5
* Conservative floor applied: 4 -> 5. The FLOOR 5 condition applies because a real UAF is established and a later concrete stale refcount operation through the same dangling `bpf_prog` pointer is described.
* Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**::
2. Signal breakdown
Positive signals:
* +2 Real UAF / active lifetime violation. `mana_xdp_set()` publishes the new `bpf_prog` pointer before RX buffer pre-allocation. On allocation failure, the BPF core drops its reference and the program can be freed while `apc->bpf_prog` still references it.
* +1 Concrete repeated lifetime/refcount mutation. The first BPF-core reference drop may end the program lifetime. The exact dangling reference remains in `apc->bpf_prog`. A later `mana_xdp_get()` consumer invokes `bpf_prog_add()` on that stale pointer, which is a refcount-affecting operation under normal BPF helper semantics.
* +1 Realistic kernel crash / strong DoS. Accessing and incrementing BPF program lifetime state through freed memory can result in an Oops, KASAN report, or kernel crash.
Paranoid-only interpretation:
* +1 Paranoid stale-refcount leverage was considered because `bpf_prog_add()` may mutate refcount-related storage after the original program lifetime ended. It is not separately reflected in the final score because the conservative FLOOR 5 already raises the conservative result to 5.
No negative privilege, race, hardware, or subsystem penalties were applied:
* Host-admin-equivalent privilege is not positively established from the supplied material. XDP attachment normally requires privileged BPF and netdevice control, but the exact capability scope and target network namespace are not shown.
* No hard race penalty applies. This is an ordered error-path and later-use sequence, not an externally uncontrollable timing race.
* No rare-hardware penalty applies. MANA is device-specific, but the input does not establish physical-only access or genuinely rare attacker-unreachable hardware.
Important non-awarded signals:
* Controlled reclaim or attacker-controlled replacement of the freed `bpf_prog` object is not demonstrated.
* Arbitrary or directed kernel write is not demonstrated.
* Callback, function pointer, or BPF instruction execution through reclaimed memory is not demonstrated.
* Confidentiality and integrity impact are not separately awarded merely because the stale refcount operation could theoretically affect reused memory.
* Remote reachability is not established. The triggering operation is local XDP configuration and later local interface reconfiguration or reset.
* Structured stale-object leverage is not separately awarded. `struct bpf_prog` is complex, but the supplied evidence only establishes stale use by `bpf_prog_add()`, not attacker-shaped post-free fields consumed as an active replacement object.
* Weak LPE concern is not separately awarded because it would duplicate the concrete stale refcount-mutation signal.
* Extra Availability is not awarded because it would double-count the same kernel crash / DoS consequence.
Repeated lifetime analysis:
* First lifetime operation can end lifetime: yes. The failed attach causes the BPF core to drop the caller-held reference, and the commit explicitly states that the program can be freed.
* Exact reference survives: yes. `apc->bpf_prog` remains set to the program that may have been freed.
* Second operation mutates lifetime/refcount state: yes. The supplied description states that a later consumer calls `bpf_prog_add()` on freed memory. Under standard BPF semantics this acquires or increments a program reference.
* Concrete repeated release/refcount signal awarded: yes, +1.
* Additional paranoid stale-refcount leverage: technically defensible, but it does not increase the post-floor numeric result.
* FLOOR 5 applied: yes.
3. Reachability analysis
The immediate trigger is a local XDP attachment attempt through an operation such as `ip link set dev ethX xdp obj ...`, combined with RX buffer pre-allocation failure. A later interface bring-up or TX timeout reset reaches the stale pointer through `mana_chn_setxdp()` or a similar consumer of `mana_xdp_get()`.
The supplied material does not establish that an ordinary unprivileged user can perform the complete chain. BPF program loading and XDP attachment are generally privilege-gated, and controlling a MANA device normally requires netdevice administration rights. However, the evidence does not resolve whether the relevant device is host-owned, delegated to a restricted service account, moved into an attacker-controlled network namespace, or otherwise exposed below host-admin level.
Therefore:
* HostAdminRequired is UNKNOWN, not YES. Host-level CAP_NET_ADMIN or equivalent may be required in common deployments, but this is not fully proven by the supplied context.
* NamespaceScopedPrivilege is UNKNOWN. Netdevice control may be namespace-scoped in some configurations, but the complete BPF load, XDP attach, allocation-failure, reset, and stale-consumption sequence is not established as namespace-reachable.
* Prior privileged setup is possible because a MANA device and XDP-capable environment must exist, but the prompt does not establish a separate host-admin provisioning step that the attacker cannot influence.
* No direct remote packet-only trigger is established.
* The ordered failure sequence is repeatable in principle if an attacker can repeatedly attach XDP programs and influence memory pressure or RX pre-allocation failure, but attacker control over allocation failure and reset frequency is not established.
4. Primitive and severity interpretation
The exact primitive is a dangling `struct bpf_prog *` retained in `apc->bpf_prog` after the BPF core may have freed the attached program. The later path does not merely read a stale pointer and immediately fault. It calls `bpf_prog_add()` on the stale program, which performs a lifetime/refcount-related operation on potentially freed storage.
This is an active structured-object UAF, not fixed async stale storage. The stale use occurs after a failed XDP setup and during later queue allocation, interface bring-up, or reset processing. The currently demonstrated consequence is kernel memory safety failure and likely local DoS. Controlled reclaim, controlled object substitution, arbitrary kernel read or write, BPF instruction execution from reclaimed storage, confidentiality disclosure, integrity modification, and privilege escalation are not established.
The issue is CLASS_B rather than CLASS_A. A real post-lifetime refcount mutation is concrete and materially stronger than an ordinary stale cleanup dereference, but there is no evidence of controlled reclaim, refcount takeover, callback control, directed corruption, or a demonstrated LPE chain.
ImportantCandidate is YES because a concrete dangling active-object reference receives a later refcount-affecting operation, leaving meaningful post-lifetime corruption leverage unresolved. AutoDowngradeAllowed is NO because automatically reducing this to an ordinary allocation-failure or crash-only issue would discard the established stale lifetime mutation primitive. The likely operational severity is Actionable Moderate, with Important-level manual review needed to determine whether MANA/XDP object lifetime and allocator reuse can provide practical exploitation leverage.
5. One-sentence report phrase
A failed MANA XDP attach can leave `apc->bpf_prog` referencing a freed BPF program, and a later interface bring-up or reset may call `bpf_prog_add()` on the stale object, causing a local kernel use-after-free with unresolved refcount-corruption leverage.
6. Manual review recommendation
MANUAL CHECK REQUIRED
A concrete UAF and subsequent refcount-affecting operation through the same potentially freed `bpf_prog` reference are established, while the practical scope of XDP privilege delegation and post-free object reuse remains unresolved.
7. Call-site confidence
Call-site confidence: medium
======================================================================
UPSTREAM PATCH SUMMARY
======================================================================
Patch: net: mana: restore the XDP program pointer when pre-allocation fails [ Upstream
Commit: 5b3cdc9abd82b7e49e0f1a21d6eb75466c5ecef8
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=5b3cdc9abd82b7e49e0f1a21d6eb75466c5ecef8
Changed files:
drivers/net/ethernet/microsoft/mana/mana_bpf.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=5b3cdc9abd82b7e49e0f1a21d6eb75466c5ecef8
======================================================================
DETAILED REPORT METHODOLOGY
======================================================================
The original Linux kernel CVE announcement for CVE-2026-97955 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97955
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:H/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: unknown
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 [email protected] (and both
send reply to CVE record itself too and see "reply" button below for howto reply).
^ permalink raw reply [relevance 21%]
Results 1-1 of 1 | reverse | sort options + mbox downloads above
-- links below jump to the message on this page --
2026-09-26 20:21 21% [CVE-2026-97955][MODERATE 7.0] net: mana: restore the XDP program pointer when pre-allocation fails [ Upstream AL-KERNEL
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox