---
id: "CVE-2026-98953"
kind: "vulnerability"
aliases: ["CVE-2026-98953"]
title: "Linux kernel: devlink reference leak when nested notify work is pending"
severity: "unknown"
cvss: null
epss: null
exploited: false
status: "open"
category: "runtime"
products: []
ecosystems: []
tags: ["devlink", "linux-kernel", "network-hardware", "runtime"]
not_affected_if: ["no-devlink-usage"]
published_at: "2026-10-23T13:17:16+00:00"
first_seen_at: "2026-10-23T21:57:33+00:00"
updated_at: "2026-10-25T04:57:52+00:00"
sources: [{"source": "nvd", "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-98953", "first_seen_at": "2026-10-23T21:57:33+00:00"}]
---

# Linux kernel: devlink reference leak when nested notify work is pending

In the Linux kernel's devlink subsystem, a reference taken for queued notify work is leaked when the work is already pending, so each register and unregister cycle of a nested devlink instance leaks a devlink_rel object and its index. The sources describe a resource leak; they do not state a code execution or privilege escalation path.

_Summary, check and fix written by a model from the sources; labelled as such._

## How to check

- `uname -r` and compare against the fixed kernel builds listed in the NVD entry
- unknown whether the sources list affected kernel version ranges

## Fix

- Update the Linux kernel to a build that includes the upstream devlink fix (fixed version: unknown)
- unknown whether distribution kernel packages have shipped the fix

## Timeline

- 2026-10-23T21:57:33+00:00 nvd https://nvd.nist.gov/vuln/detail/CVE-2026-98953

## From the sources

In the Linux kernel, the following vulnerability has been resolved:

devlink: fix devlink_rel reference leak when notify work is pending

devlink_rel_nested_in_notify_work_schedule() takes a reference on the
devlink_rel for the notify work and then queues the work, ignoring the
return value of schedule_delayed_work(). If the work is already pending,
nothing new is queued, the work runs only once and drops only one
reference, so the extra one is leaked together with the devlink_rel and
its index in devlink_rels.

This is easy to hit. devl_register() of a nested instance queues the
work, and if the instance is unregistered before the work has run,
devlink_rel_put() queues it again while it is still pending. The work
also keeps rescheduling itself for as long as the parent devlink lock
cannot be taken, which widens the window. Registering and unregistering
a nested devlink instance 100 times while holding the parent lock leaks
all 100 devlink_rel objects.

The reschedule path in devlink_rel_nested_in_notify_work() has the same
problem: if the work was queued again while it was running, the
reference it holds is never dropped.

Drop the reference in both places when the work was already pending.
The pending work holds its own reference, so this can not be the last
one.

---

_Free to read at a human pace. An API with keys and an MCP server are in preparation; early access: hello@infrastack.ai_