xfrm: hold dev ref until after transport_finish NF_HOOK
Affects the Linux kernel xfrm input path using asynchronous crypto completion, in kernels from 4.15.0 onward. Fixed in 6.6.160, 6.12.94, 6.18.23, and 6.19.13 on their branches.
- Published
- Last modified
- Where the record stands: published, rejected by its publisher, or superseded by another id.What do these mean? →
- published
- Whether CISA lists it in its Known Exploited Vulnerabilities catalog: proof that attackers are using it.What do these mean? →
- Not listed in CISA KEV
- How many real-world setups can hit the bug, from the vendor's advisory and CISA data. It does not know your setup.What do these mean? →
- Specific setups only
- Fix
- Fixed in 3.3, 3.17, 4.15, 6.6.160, 6.12.94, and 2 more
- Whether working attack code is public (Exploit-DB, Metasploit, Nuclei or GitHub). Public code makes attacks easier.What do these mean? →
- None on Exploit-DB, Metasploit or GitHub
- Exploitation: none
- Automatable: no
- Technical impact: total
Summary
In the Linux kernel, the following vulnerability has been resolved: xfrm: hold dev ref until after transport_finish NF_HOOK After async crypto completes, xfrm_input_resume() calls dev_put() immediately on re-entry before the skb reaches transport_finish. The skb->dev pointer is then used inside NF_HOOK and its okfn, which can race with device teardown. Remove the dev_put from the async resumption entry and instead drop the reference after the NF_HOOK call in transport_finish, using a saved device pointer since NF_HOOK may consume the skb. This covers NF_DROP, NF_QUEUE and NF_STOLEN paths that skip the okfn. For non-transport exits (decaps, gro, drop) and secondary async return points, release the reference inline when async is set.
Timeline
- CVE published
- CISA SSVC assessment
Exploitation: none · Automatable: no · Technical impact: total
- Record last updated
Affected versions
- Linux Linux
From 3.2.100 · fixed in 3.3; From 3.16.55 · fixed in 3.17; From 4.14.24 · fixed in 4.15
- Linux Linux
4.15
- KernelLinux
From 4.15.0 · fixed in 6.6.160; From 6.7.0 · fixed in 6.12.94; From 6.13.0 · fixed in 6.18.23; From 6.19.0 · fixed in 6.19.13
References 6
git.kernel.org
Questions about CVE-2026-31663
What does this mean?
The network device referenced by a packet can be torn down while the packet is still being processed. The device reference is released too early on the asynchronous crypto completion path, which can lead to use of a freed device pointer.Source: git.kernel.org
Am I affected, and how do I check?
Kernel ranges listed as affected are 4.15.0 up to 6.6.160, 6.7.0 up to 6.12.94, 6.13.0 up to 6.18.23, and 6.19.0 up to 6.19.13, plus older stable ranges from 3.2.100, 3.16.55, and 4.14.24. Only packets that pass through the asynchronous crypto completion path in xfrm input are involved, so the kernel version (uname -r) and whether that path is used are the things to check.Source: git.kernel.org
What should I do about it?
Move to a kernel release at or beyond the fixed version for the branch in use: 6.6.160, 6.12.94, 6.18.23, or 6.19.13, or a later release. Older branches are fixed in 3.3, 3.17, and 4.15 respectively.Source: git.kernel.org
Is it being exploited?
It is not in CISA's Known Exploited Vulnerabilities catalog, and CISA's SSVC assessment does not mark exploitation as active. That does not mean it has not been attacked: vendors and researchers can report attacks before CISA lists them. No public exploit code is tracked.Source: cve.now feed data