CVE-2025-38352 is a time-of-check time-of-use race (CWE-367) in the Linux kernel's POSIX CPU timer code that a local attacker can use to escalate privileges. The kernel CNA scores it 7.8; NVD has no Primary score. CISA lists it as exploited and Google reports limited, targeted exploitation on Android. Fixed in 6.1.142, 6.6.94 and other stable releases.
Vulnerability summary
CVSS base score
7.8
High severity
EPSS score (FIRST)
1.4%
Probability of exploitation in the wild in the next 30 days
CISA KEV catalog
Yes
Listed as known exploited
Published
Vulnerability ID: CVE-2025-38352 · Source: NVD
Affected components
| Component | CPE | Version range |
|---|---|---|
| Linux kernel (posix-cpu-timers, CONFIG_POSIX_TIMERS) | cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:* | 2.6.36 to before 5.4.295, 5.5 to before 5.10.239, 5.11 to before 5.15.186, 5.16 to before 6.1.142, 6.2 to before 6.6.94, 6.7 to before 6.12.34, 6.13 to before 6.15.3, and 6.16-rc1 (NVD); unaffected from 6.16 (kernel CNA record) |
| Debian linux | cpe:2.3:o:debian:debian_linux:11.0:*:*:*:*:*:*:* | Fixed in 5.10.244-1 for bullseye (DLA-4327-1), 6.1.147-1 for bookworm (DSA-5973-1), 6.12.35-1 in unstable; linux-6.1 for bullseye fixed in 6.1.153-1~deb11u1 (DLA-4328-1) (Debian security tracker) |
| Android (kernel, Time subcomponent) | — | Fixed at security patch level 2025-09-05 or later (Android Security Bulletin September 2025, A-425282960) |
What does this mean for affected devices?
CVE-2025-38352 is a time-of-check time-of-use race (CWE-367) in the Linux kernel's POSIX CPU timer code, kernel/time/posix-cpu-timers.c. The kernel CNA's record, titled "posix-cpu-timers: fix race between handle_posix_cpu_timers() and posix_cpu_timer_del()", explains that an exiting non-autoreaping task that calls handle_posix_cpu_timers() from IRQ "can be reaped by its parent or debugger right after unlock_task_sighand()", and a concurrent posix_cpu_timer_del() then fails to see that the timer is firing. NVD has analyzed the CVE but publishes no Primary score of its own as of 2026-11-19. The Linux kernel CNA scores it CVSS 3.1 7.8, CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H, and its scenario notes say the bug is reached "purely through local syscalls" with "no network-facing path". CISA's ADP container assigns CWE-367 (NVD, CVE record).
Which kernel versions are affected?
The bug dates from Linux 2.6.36. The kernel CNA's dyad file lists the first fixed release on each stable branch: 5.4.295, 5.10.239, 5.15.186, 6.1.142, 6.6.94, 6.12.34, 6.15.3, and 6.16 onward, where mainline commit f90fff1e152d landed. NVD's CPE ranges match and add 6.16-rc1.
The vulnerable file is built only with CONFIG_POSIX_TIMERS=y, per the kernel/time Makefile. The CNA description adds that the fix "is not needed if CONFIG_POSIX_CPU_TIMERS_TASK_WORK=y". In kernel/time/Kconfig that option defaults to y when POSIX_TIMERS and HAVE_POSIX_CPU_TIMERS_TASK_WORK are set. In the v6.1 source, x86 and arm64 select HAVE_POSIX_CPU_TIMERS_TASK_WORK and 32-bit arm does not; current mainline arm selects it. Check the .config of each image rather than assuming. Our Linux kernel CVE triage guide uses this CVE in its worked example of config-based triage.
Which devices are affected?
Linux and Android devices on an unfixed kernel where local code can run. Google's Android Security Bulletin for September 2025, published 2 September 2025, lists CVE-2025-38352 in the Kernel section under the 2025-09-05 security patch level as an elevation of privilege rated High, subcomponent Time (A-425282960). Debian, per its security tracker, fixed linux in 5.10.244-1 for bullseye (DLA-4327-1), 6.1.147-1 for bookworm (DSA-5973-1) and 6.12.35-1 in unstable, and linux-6.1 for bullseye in 6.1.153-1~deb11u1 (DLA-4328-1).
How do you fix CVE-2025-38352?
Move the product kernel to at least the fixed release of its stable branch, for example 5.10.239, 5.15.186, 6.1.142 or 6.6.94, or backport the branch's commit from the dyad file. Android devices need security patch level 2025-09-05 or later. Since the attack is local, limit who can run code on the device: no untrusted apps or shell accounts, and signed firmware updates only.
Is CVE-2025-38352 exploited?
Yes. CISA added it to the KEV catalog on 2025-09-04 with a due date of 2025-09-25; known ransomware campaign use is listed as Unknown, as of 2026-11-19 (catalog version 2026.11.18). Google's September 2025 bulletin states there are indications that CVE-2025-38352 "may be under limited, targeted exploitation". NVD tags a public GitHub repository as an exploit reference. FIRST EPSS gives it 0.01357, the 72.2nd percentile, in the 2026-11-18 dataset (CISA KEV, FIRST EPSS). Other KEV-listed kernel CVEs from the same triage example are CVE-2024-53104 in uvcvideo, CVE-2025-39964 in af_alg and CVE-2025-39682 in kernel TLS.
Evidence
- NVD: status Analyzed; no [email protected] Primary metric; Secondary CVSS 3.1 7.8 AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H from the Linux kernel CNA; CWE-367 from CISA-ADP; CPE linux_kernel 2.6.36 to before 5.4.295, 5.5-5.10.239, 5.11-5.15.186, 5.16-6.1.142, 6.2-6.6.94, 6.7-6.12.34, 6.13-6.15.3, 6.16-rc1, plus debian_linux 11.0; exploit reference github.com/farazsth98/chronomaly; published 2025-07-22; read from cached API response 2026-11-19. — NVD
- CVE record (CNA Linux, PUBLISHED 2025-07-22): title 'posix-cpu-timers: fix race between handle_posix_cpu_timers() and posix_cpu_timer_del()'; programFiles kernel/time/posix-cpu-timers.c; affected from 2.6.36; unaffected 5.4.295, 5.10.239, 5.15.186, 6.1.142, 6.6.94, 6.12.34, 6.15.3, 6.16; CNA CVSS 3.1 7.8 with scenario 'reached purely through local syscalls'; description says the fix is not needed if CONFIG_POSIX_CPU_TIMERS_TASK_WORK=y; CISA ADP: CWE-367, SSVC exploitation active, automatable no, technical impact total. — CVE.org
- Kernel CNA dyad file: 2.6.36 (0bdd2ed4138e) vulnerable; fixed 5.4.295 78a4b8e3795b, 5.10.239 c076635b3a42, 5.15.186 2f3daa04a932, 6.1.142 764a7a5dfda2, 6.6.94 2c72fe18cc5f, 6.12.34 c29d5318708e, 6.15.3 460188bc042a, 6.16 f90fff1e152d. — Linux kernel vulns.git
- kernel/time/Makefile builds posix-cpu-timers.o only under ifeq ($(CONFIG_POSIX_TIMERS),y); kernel/time/Kconfig: POSIX_CPU_TIMERS_TASK_WORK default y if POSIX_TIMERS && HAVE_POSIX_CPU_TIMERS_TASK_WORK; v6.1 arch/x86 and arch/arm64 Kconfig select HAVE_POSIX_CPU_TIMERS_TASK_WORK, v6.1 arch/arm does not, mainline arch/arm does (read 2026-11-19). — Linux kernel source
- Android Security Bulletin September 2025 (published 2025-09-02, updated 2025-12-08): 'There are indications that the following may be under limited, targeted exploitation. CVE-2025-38352 CVE-2025-48543'; Kernel section, 2025-09-05 patch level, A-425282960, EoP, High, subcomponent Time. — Android Security Bulletin
- Debian: linux fixed bullseye 5.10.244-1 (DLA-4327-1), bookworm 6.1.147-1 (DSA-5973-1), unstable 6.12.35-1; linux-6.1 bullseye 6.1.153-1~deb11u1 (DLA-4328-1). — Debian security tracker
- CISA KEV catalog version 2026.11.18: Linux Kernel Time-of-Check Time-of-Use (TOCTOU) Race Condition Vulnerability, dateAdded 2025-09-04, dueDate 2025-09-25, knownRansomwareCampaignUse Unknown, CWE-367; checked 2026-11-19. — CISA KEV
- EPSS 0.01357, percentile 0.72229, dataset 2026-11-18, read 2026-11-19. — FIRST EPSS

Related advisories
Sources
- NVD
- CVE.org record CVE-2025-38352 (CNA: Linux)
- Android Security Bulletin: September 2025
- Debian security tracker: CVE-2025-38352
- Linux kernel vulns.git: CVE-2025-38352.dyad (vulnerable and fixed commits per branch)
- Linux kernel mainline fix commit f90fff1e152d
- Linux kernel source: kernel/time/Makefile
- Linux kernel source: kernel/time/Kconfig
- CISA KEV
- FIRST EPSS
Frequently asked questions
Which kernel versions fix CVE-2025-38352?
5.4.295, 5.10.239, 5.15.186, 6.1.142, 6.6.94, 6.12.34, 6.15.3 and 6.16 onward; the bug dates from 2.6.36. Each stable branch has its own fixed release, so a 6.1.150 kernel is fixed (per the kernel CNA's CVE record and dyad file).
Does CVE-2025-38352 have an NVD CVSS score?
NVD lists no Primary score of its own as of 2026-11-19. The Linux kernel CNA scores it CVSS 3.1 7.8 with a local attack vector and low privileges required, and CISA's ADP entry assigns CWE-367 (per NVD and the CVE record).
Is a kernel built with CONFIG_POSIX_CPU_TIMERS_TASK_WORK affected?
The CNA description states the fix is not needed when CONFIG_POSIX_CPU_TIMERS_TASK_WORK=y, because exit_task_work() runs before exit_notify(). That option defaults to y on architectures that select HAVE_POSIX_CPU_TIMERS_TASK_WORK, so check the image's .config (per the CVE record and kernel/time/Kconfig).
Which Android security patch level fixes CVE-2025-38352?
Security patch level 2025-09-05 or later. The bulletin lists it as a High-severity elevation of privilege in the kernel's Time subcomponent and notes indications of limited, targeted exploitation (per the Android Security Bulletin for September 2025).
Is CVE-2025-38352 exploited in the wild?
Yes. CISA added it to the KEV catalog on 2025-09-04, with ransomware use listed as Unknown, as of 2026-11-19 (catalog version 2026.11.18). FIRST EPSS gives it 0.01357, the 72.2nd percentile, in the 2026-11-18 dataset (per CISA KEV and FIRST EPSS).
Questions about this page?