---
id: "CVE-2026-98858"
kind: "vulnerability"
aliases: ["CVE-2026-98858"]
title: "Linux kernel io_uring: SQPOLL task_work use-after-free on ring teardown"
severity: "unknown"
cvss: null
epss: null
exploited: false
status: "open"
category: "runtime"
products: []
ecosystems: []
tags: ["io-uring", "linux-kernel", "runtime", "self-hosted"]
not_affected_if: ["io-uring-disabled", "no-sqpoll-rings"]
published_at: "2026-10-18T16:17:40+00:00"
first_seen_at: "2026-10-23T21:57:17+00:00"
updated_at: "2026-10-25T05:12:57+00:00"
sources: [{"source": "nvd", "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-98858", "first_seen_at": "2026-10-23T21:57:17+00:00"}]
---

# Linux kernel io_uring: SQPOLL task_work use-after-free on ring teardown

In the Linux kernel's io_uring subsystem, the SQPOLL thread can complete a request while the ring is being torn down, so the add path reads freed memory. The sources describe a KASAN-reported use-after-free; the text does not state what an attacker can achieve beyond a kernel memory fault.

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

## How to check

- Run `uname -r` and compare against the kernel version that carries the fix, which the sources do not name.
- Check whether io_uring is in use with SQPOLL rings, for example with `grep -r IORING_SETUP_SQPOLL` in the application source, or `cat /proc/sys/kernel/io_uring_disabled`.

## Fix

- Update the Linux kernel to a release that includes the upstream commit fixing io_uring task_work add with SQPOLL; the sources do not give the version number, so unknown.
- If io_uring is not needed, set `sysctl kernel.io_uring_disabled=2` (unknown whether the sources endorse this).

## Timeline

- 2026-10-23T21:57:17+00:00 nvd https://nvd.nist.gov/vuln/detail/CVE-2026-98858

## From the sources

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

io_uring: fix task_work add use-after-free with SQPOLL

For SQPOLL, the sqpoll thread runs task_work off its own list and
doesn't need to be told about it, so it can pop and complete a request
as soon as mpscq_push() has published it. If that was the last request
holding the ring alive, io_ring_exit_work() then frees the ctx, stops
the thread, and with it the tctx, while io_req_normal_work_add() is
still looking at both after the push:

  BUG: KASAN: slab-use-after-free in io_req_normal_work_add+0x439/0x510
  Read of size 4 at addr ff11000000ca0000 by task repro/55
  (...)
  BUG: KASAN: slab-use-after-free in queue_work_on+0x25/0x70
  Write of size 8 at addr ff11000000c3bd00 by task repro/55

Hold an RCU read lock across the add, like io_req_local_work_add() already
does, and have io_ring_exit_work() wait for a grace period for SQPOLL rings
as well before freeing the ctx.

---

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