Логотип exploitDog
Консоль
Логотип exploitDog

exploitDog

redhat логотип

CVE-2026-23225

Опубликовано: 18 фев. 2026
Источник: redhat
CVSS3: 7

Описание

In the Linux kernel, the following vulnerability has been resolved: sched/mmcid: Don't assume CID is CPU owned on mode switch Shinichiro reported a KASAN UAF, which is actually an out of bounds access in the MMCID management code. CPU0CPU1 T1 runs in userspace T0: fork(T4) -> Switch to per CPU CID mode fixup() set MM_CID_TRANSIT on T1/CPU1 T4 exit() T3 exit() T2 exit() T1 exit() switch to per task mode ---> Out of bounds access. As T1 has not scheduled after T0 set the TRANSIT bit, it exits with the TRANSIT bit set. sched_mm_cid_remove_user() clears the TRANSIT bit in the task and drops the CID, but it does not touch the per CPU storage. That's functionally correct because a CID is only owned by the CPU when the ONCPU bit is set, which is mutually exclusive with the TRANSIT flag. Now sched_mm_cid_exit() assumes that the CID is CPU owned because the prior mode was per CPU. It invokes mm_drop_cid_on_cpu() which clears the not set ONCPU bit and then invokes clear_bit() with an insanely large bit number because TRANSIT is set (bit 29). Prevent that by actually validating that the CID is CPU owned in mm_drop_cid_on_cpu().

A flaw was found in the Linux kernel. A local user can exploit a race condition in the sched/mmcid mode switch path. This occurs when a task exits with a specific flag (MM_CID_TRANSIT) still set, leading to an incorrect assumption that the Context ID (CID) is CPU-owned. This can result in an out-of-bounds bit operation, causing the kernel to crash, leading to a Denial of Service (DoS). In a worst-case scenario, this flaw could also lead to kernel memory corruption, potentially impacting system integrity and confidentiality.

Отчет

A local race in the sched mmcid mode switch path can lead to an out of bounds bit operation when a task exits with the MM_CID_TRANSIT flag still set and the code incorrectly assumes the CID is CPU owned. This can cause clear_bit to be called with an extremely large bit index which can crash the kernel and may corrupt kernel memory. For the CVSS the PR is L because a normal local user can create and exit tasks and potentially hit the timing window during CID mode transitions. The issue is not network reachable and requires local execution on the host. Impact is at least denial of service due to a kernel crash. A conservative worst case also considers possible integrity and confidentiality impact if the out of bounds write can be steered into sensitive memory.

Затронутые пакеты

ПлатформаПакетСостояниеРекомендацияРелиз
Red Hat Enterprise Linux 10kernelNot affected
Red Hat Enterprise Linux 6kernelNot affected
Red Hat Enterprise Linux 7kernelNot affected
Red Hat Enterprise Linux 7kernel-rtNot affected
Red Hat Enterprise Linux 8kernelNot affected
Red Hat Enterprise Linux 8kernel-rtNot affected
Red Hat Enterprise Linux 9kernelNot affected
Red Hat Enterprise Linux 9kernel-rtNot affected

Показывать по

Дополнительная информация

Статус:

Moderate
Дефект:
CWE-787
https://bugzilla.redhat.com/show_bug.cgi?id=2440676kernel: Kernel: Denial of Service and potential memory corruption via mmcid mode switch race condition

7 High

CVSS3

Связанные уязвимости

ubuntu
около 1 месяца назад

In the Linux kernel, the following vulnerability has been resolved: sched/mmcid: Don't assume CID is CPU owned on mode switch Shinichiro reported a KASAN UAF, which is actually an out of bounds access in the MMCID management code. CPU0 CPU1 T1 runs in userspace T0: fork(T4) -> Switch to per CPU CID mode fixup() set MM_CID_TRANSIT on T1/CPU1 T4 exit() T3 exit() T2 exit() T1 exit() switch to per task mode ---> Out of bounds access. As T1 has not scheduled after T0 set the TRANSIT bit, it exits with the TRANSIT bit set. sched_mm_cid_remove_user() clears the TRANSIT bit in the task and drops the CID, but it does not touch the per CPU storage. That's functionally correct because a CID is only owned by the CPU when the ONCPU bit is set, which is mutually exclusive with the TRANSIT flag. Now sched_mm_cid_exit() assumes that the CID is CPU owned because the prior mode was per CPU. It invokes mm_drop_cid_on_cpu() which clears the not set ONCPU bit and then invokes clear_bit() with an in...

nvd
около 1 месяца назад

In the Linux kernel, the following vulnerability has been resolved: sched/mmcid: Don't assume CID is CPU owned on mode switch Shinichiro reported a KASAN UAF, which is actually an out of bounds access in the MMCID management code. CPU0 CPU1 T1 runs in userspace T0: fork(T4) -> Switch to per CPU CID mode fixup() set MM_CID_TRANSIT on T1/CPU1 T4 exit() T3 exit() T2 exit() T1 exit() switch to per task mode ---> Out of bounds access. As T1 has not scheduled after T0 set the TRANSIT bit, it exits with the TRANSIT bit set. sched_mm_cid_remove_user() clears the TRANSIT bit in the task and drops the CID, but it does not touch the per CPU storage. That's functionally correct because a CID is only owned by the CPU when the ONCPU bit is set, which is mutually exclusive with the TRANSIT flag. Now sched_mm_cid_exit() assumes that the CID is CPU owned because the prior mode was per CPU. It invokes mm_drop_cid_on_cpu() which clears the not set O

CVSS3: 5.3
msrc
около 1 месяца назад

sched/mmcid: Don't assume CID is CPU owned on mode switch

debian
около 1 месяца назад

In the Linux kernel, the following vulnerability has been resolved: s ...

github
около 1 месяца назад

In the Linux kernel, the following vulnerability has been resolved: sched/mmcid: Don't assume CID is CPU owned on mode switch Shinichiro reported a KASAN UAF, which is actually an out of bounds access in the MMCID management code. CPU0 CPU1 T1 runs in userspace T0: fork(T4) -> Switch to per CPU CID mode fixup() set MM_CID_TRANSIT on T1/CPU1 T4 exit() T3 exit() T2 exit() T1 exit() switch to per task mode ---> Out of bounds access. As T1 has not scheduled after T0 set the TRANSIT bit, it exits with the TRANSIT bit set. sched_mm_cid_remove_user() clears the TRANSIT bit in the task and drops the CID, but it does not touch the per CPU storage. That's functionally correct because a CID is only owned by the CPU when the ONCPU bit is set, which is mutually exclusive with the TRANSIT flag. Now sched_mm_cid_exit() assumes that the CID is CPU owned because the prior mode was per CPU. It invokes mm_drop_cid_on_cpu() which clears the not se...

7 High

CVSS3