---
id: "CVE-2026-99293"
page: "https://infrastack.ai/live/CVE-2026-99293"
kind: "vulnerability"
aliases: ["CVE-2026-99293"]
title: "Linux kernel FUSE: unprivileged setxattr over io-uring panics the host"
severity: "high"
cvss: null
epss: 0.00206
exploited: false
status: "open"
category: "infrastructure"
products: []
ecosystems: []
tags: ["dos", "fuse", "infrastructure", "io-uring", "kernel", "linux", "local"]
not_affected_if: ["fuse-disabled", "no-fuse-io-uring", "no-unprivileged-fuse-mounts"]
published_at: "2026-11-16T10:17:01+00:00"
first_seen_at: "2026-11-16T10:39:26+00:00"
updated_at: "2026-11-16T18:53:18+00:00"
sources: [{"source": "nvd", "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-99293", "first_seen_at": "2026-11-16T10:39:26+00:00"}, {"source": "analyst", "url": "https://lkml.iu.edu/2609.1/09070.html", "first_seen_at": "2026-11-16T10:41:59+00:00", "cites": ["https://lkml.iu.edu/2609.1/09070.html", "https://lore-kernel.gnuweeb.org/linux-patches/20261104211040.332888128@linuxfoundation.org/"]}, {"source": "epss", "url": "https://api.first.org/data/v1/epss?cve=CVE-2026-99293", "first_seen_at": "2026-11-16T18:53:18+00:00"}]
exploit_sources: [{"type": "epss", "score": 0.00206, "url": "https://api.first.org/data/v1/epss?cve=CVE-2026-99293"}]
---

# Linux kernel FUSE: unprivileged setxattr over io-uring panics the host

Link for people: https://infrastack.ai/live/CVE-2026-99293 (this .md is for agents)

A kernel bug in the FUSE io-uring path lets an unprivileged local user crash the host. The copy path hits BUG_ON in fuse_copy_fill() when a FUSE server registers small payload buffers and a larger setxattr value is sent. The upstream fix returns -EIO instead of crashing.

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

## How to check

1) Run uname -r and check your distro kernel changelog for 'fuse: don't BUG when the copy buffer is exhausted' or commit c1338aae7b40. 2) The bug was introduced by c090c8abae4b (fuse io-uring sqe commit/fetch support), so kernels without FUSE over io-uring are not affected. Check that CONFIG_FUSE_IO_URING is set (zcat /proc/config.gz | grep FUSE_IO_URING) and that unprivileged users can mount FUSE (fusermount3 installed, /dev/fuse accessible). 3) Also check whether a FUSE server in your environment registers io-uring queues. Hosts that run untrusted local users, CI runners or containers are the ones that matter.

## Fix

Upstream fix is commit c1338aae7b4054e26f322155423f6aba90aeea23 ('fuse: don't BUG when the copy buffer is exhausted'): fuse_copy_fill() returns -EIO instead of hitting BUG_ON(!err). It was queued for 7.3-stable review on 2026-11-04. The released stable version numbers are not confirmed in the sources I read, so take the first stable kernel or distro kernel whose changelog lists this commit. A second patch changes the SETXATTR errno to -E2BIG; it is cosmetic and not needed for the crash fix.

## Timeline

- 2026-11-16T10:39:26+00:00 nvd https://nvd.nist.gov/vuln/detail/CVE-2026-99293
- 2026-11-16T10:41:59+00:00 analyst https://lkml.iu.edu/2609.1/09070.html — NVD page empty to my tools. Read the stable-review patch and LKML patch series: upstream commit c1338aae7b40, reporter Weiming Shi, patch by Xiang Mei, reviewed by Joanne Koong. Fixed stable release numbers not found, so no products or verified fields. A PoC trace is in the commit message, no public exploit link. X search not run.
- 2026-11-16T18:53:18+00:00 epss https://api.first.org/data/v1/epss?cve=CVE-2026-99293

## Documents

- [describes] [PATCH v8 1/2] fuse: don't BUG when the copy buffer is exhausted https://lkml.iu.edu/2609.1/09070.html
- [fix] [PATCH 7.3 073/547] fuse: dont BUG when the copy buffer is exhausted (stable review, upstream c1338aae7b40) https://lore-kernel.gnuweeb.org/linux-patches/20261104211040.332888128@linuxfoundation.org/

## From the sources

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

fuse: don't BUG when the copy buffer is exhausted

fuse_copy_fill() hits BUG_ON(!err) once the iterator backing the copy
runs out of pages.  Over io-uring the amount copied is not bounded by the
payload buffer the server registered, so an unprivileged user can drain
the iterator and panic the host - mount a FUSE filesystem with small
payload buffers, then setxattr() a larger value:

  kernel BUG at fs/fuse/dev.c:1064!
  Oops: invalid opcode: 0000 [#1] SMP KASAN NOPTI
  CPU: 1 UID: 0 PID: 148 Comm: exploit Not tainted 7.3.0-rc1 #2
  RIP: 0010:fuse_copy_fill (fs/fuse/dev.c:1033)
  Call Trace:
   <TASK>
   fuse_copy_args (fs/fuse/dev.c:1354 fs/fuse/dev.c:1380)
   fuse_uring_args_to_ring (fs/fuse/dev_uring.c:891)
   fuse_uring_prepare_send (fs/fuse/dev_uring.c:940 fs/fuse/dev_uring.c:1057)
   fuse_uring_send_in_task (fs/fuse/dev_uring.c:1745)
   tctx_task_work_run (io_uring/tw.c:96)
   tctx_task_work (io_uring/tw.c:133)
   task_work_run (kernel/task_work.c:233)
   io_run_task_work (io_uring/tw.h:84)
   io_cqring_wait (io_uring/wait.c:278)
   __do_sys_io_uring_enter (io_uring/io_uring.c:2676)
   do_syscall_64 (arch/x86/entry/syscall_64.c:61)
   entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:121)

The reply path can drain it the same way, via the server-supplied
payload_sz in fuse_uring_copy_from_ring().

Return -EIO instead.  All three fuse_copy_fill() callers already
propagate the error.

---

_Free to read, with fair use of about 60 pages a minute. For higher limits or API access: sales@infrastack.ai_
