CVE-2025-38352KEV · OVERDUECWE-367

Linux Kernel Time-of-Check Time-of-Use (TOCTOU) Race Condition Vulnerability

High · published July 22, 2025

Patch now

Confirmed exploited, and the score agrees

CVSS calls it high at 7.8. It is confirmed in active exploitation. It sits in the 71.1th percentile for exploit probability.

393
days past CISA
deadline
CVSS v3.1
7.8
EPSS
1%
Percentile
71.1
In the wild
Confirmed
What it is

⚡ A race condition in the Linux kernel's handling of CPU timers could let a sneaky task escape before it’s properly dealt with. 🚨 Think of it like a restaurant where a waiter forgets to double-check the order before serving it, leading to a mix-up just as the chef is about to clean up the kitchen! This vulnerability lets a task exit before it’s fully cleaned up, allowing potential chaos. An attacker could exploit this race condition to manipulate CPU timers, potentially leading to system instability or unauthorized access as tasks slip through the cracks. If a non-autoreaping task gets the jump on its cleanup, it could leave behind a mess that could be leveraged for further attacks.

Put simply

Think of it like a restaurant where a waiter forgets to double-check the order before serving it, leading to a mix-up just as the chef is about to clean up the kitchen! This vulnerability lets a task exit before it’s fully cleaned up, allowing potential chaos. This race condition occurs when an exiting non-autoreaping task calls handle_posix_cpu_timers() during an IRQ, potentially allowing it to be reaped by its parent or debugger while concurrent posix_cpu_timer_del() runs, causing failure to detect necessary timer states.

What to do

An attacker could exploit this race condition to manipulate CPU timers, potentially leading to system instability or unauthorized access as tasks slip through the cracks. If a non-autoreaping task gets the jump on its cleanup, it could leave behind a mess that could be leveraged for further attacks. To patch this vulnerability, make sure to apply the latest updates to your Linux kernel. Specifically, check for the fix that adds a tsk->exit_state check into run_posix_cpu_timers(). If CONFIG_POSIX_CPU_TIMERS_TASK_WORK=y, the patch won't be necessary, but it's still best to stay updated! You’ve got this! Stay vigilant, apply those patches, and keep your system secure! 🛡️

The record
Technical detail
CVSS v3.1
7.8 · HIGH
Vector
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
CVSS v4.0
Not supplied
EPSS
0.01341 · 71.1th percentile
Weakness
CWE-367 · Time-of-check Time-of-use (TOCTOU) Race Condition
Published
2025-07-22T12:15Z
KEV added
2025-09-04 · due 2025-09-25
Affected products (9)
ProductVersionsFixed in
linux/linux_kernel≥ 2.6.36, < 5.4.2955.4.295
linux/linux_kernel≥ 5.5, < 5.10.2395.10.239
linux/linux_kernel≥ 5.11, < 5.15.1865.15.186
linux/linux_kernel≥ 5.16, < 6.1.1426.1.142
linux/linux_kernel≥ 6.2, < 6.6.946.6.94
linux/linux_kernel≥ 6.7, < 6.12.346.12.34
linux/linux_kernel≥ 6.13, < 6.15.36.15.3
linux/linux_kernelall versions—
debian/debian_linuxall versions—
References (12)
EPSS history
Timeline
  • 04 SEP 00:00Z
    Added to CISA KEV — remediate by Sep 25
    kev
  • 22 JUL 08:04Z
    posix-cpu-timers: fix race between handle_posix_cpu_timers() and posix_cpu_timer_del()
    cvelistv5