CVE-2025-38114: e1000: Move cancel_work_sync to avoid deadlock
In the Linux kernel, the following vulnerability has been resolved:
e1000: Move cancelworksync to avoid deadlock
Previously, e1000down called cancelworksync for the e1000 reset task (via e1000downandstop), which takes RTNL.
As reported by users and syzbot, a deadlock is possible in the following scenario:
CPU 0: - RTNL is held - e1000close - e1000down - cancelworksync (cancel / wait for e1000resettask())
CPU 1: - processonework - e1000resettask - take RTNL
To remedy this, avoid calling cancelworksync from e1000down (e1000resettask does nothing if the device is down anyway). Instead, call cancelworksync for e1000resettask when the device is being removed.
Affected Software
Event History
Frequently Asked Questions
What is the severity of CVE-2025-38114?
CVE-2025-38114 is categorized as a medium severity vulnerability due to potential deadlock issues affecting system stability.
How do I fix CVE-2025-38114?
To fix CVE-2025-38114, update the Linux kernel to the latest version that includes the resolved patch.
What systems are affected by CVE-2025-38114?
CVE-2025-38114 affects systems running the Linux kernel where the e1000 driver is utilized.
What is the impact of CVE-2025-38114?
The impact of CVE-2025-38114 is a potential system deadlock during the reset task of the e1000 driver.
Who reported CVE-2025-38114?
CVE-2025-38114 was reported by users and syzbot, highlighting concerns over the deadlock issue.