CVE-20XX-?
agent?

CVE-2026-99737

unscored No fix listed Other OS & kernel

What it is

In the Linux kernel, the following vulnerability has been resolved: mm/shmem: fix truncate loop after boundary swapin failure If swapping in a large entry at the new EOF fails, shmem_undo_range() leaves the entry in place. Its final scan skips the entry, then restarts from the same start index and finds it again. Truncation can loop forever while holding i_rwsem. Advance start past the skipped entry so that a restarted scan can finish, as we already do for folios that cannot be split. Note that this does not fix the missing zeroing when boundary swapin fails, it merely prevents the endless loop.

At a glance

Fix
No patch or advisory is linked yet. Look at mitigations or restrict exposure.

Timeline

  1. PublishedDisclosed publicly on CVE.org and NVD
  2. Last updatedMost recent change to the record

Why it ranks here

Details

Published
Last modified
Record status
Received

Affected

Is it in your code?

Hand this to your coding agent. It searches your project for the affected products and checks versions. Nothing leaves your machine.

Read https://cve-list.com/agent/cve/CVE-2026-99737.md and follow it. Check whether this project is affected by CVE-2026-99737. Read only. Don't change anything and don't upload my code. Tell me yes, no or not sure, with the file as proof.

· More options · Status badge

Meme: I wake up, CVE-2026-99737
Download Square Reddit

References

NVD record · CVE.org record