KernelScan.io

HIGH Introduced in 6.2

perf SchedCbList UAF

CVE-2026-97961

CVSS 7.8 / 10.0 KernelScan AI

CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

01

In the Linux kernel, the following vulnerability has been resolved: perf/core: Allow list_del during perf_event_overflow() A PMU might use perf_sched_cb_inc() and perf_sched_cb_dec() interface to get the PMU call back function pmu::sched_task invoked at schedule in and schedule out. This is achieved by walking along the list anchored by sched_cb_list. The following scenario might lead to a list corruption. perf_pmu_sched_task() for_each_list_entry(..., &sched_cb_list) +--> __perf_pmu_sched_task() +--> event->pmu->sched_task()) +--> PMU_push_sample() +--> perf_event_overflow() +--> __perf_event_overflow() +--> pmu->stop() +--> perf_sched_cb_dec() remove entry from sched_cb_list while list node in use. This happens when ioctl(fd, PERF_EVENT_IOC_REFRESH, xxx) has been invoked and perf_event::event_limit hits zero. Prevent the list corruption and convert for_each_list_entry() to for_each_list_entry_safe().

02

Engine v0.6.0

Risk summary

A local user with access to perf events can trigger list corruption in the perf core scheduling callback path by setting an event refresh limit that causes the current list entry to be removed during iteration. This can lead to kernel memory corruption, potentially enabling privilege escalation or a system crash. The bug requires a PMU that uses the sched_task callback interface and is reachable via the perf_event_open syscall.

Affectedkernel/events/core.c (perf/core)

Vulnerability analysis

During a context switch, the kernel walks a list of hardware monitoring callbacks. While handling one entry, that callback can trigger an event that ends up removing the current list entry before the walk has finished. Because the walk uses the current entry to find the next one, removing it corrupts the list and can later reference memory that has been freed or reused. The fix changes the list walk to remember the next entry before calling each callback, so removing the current entry during processing no longer harms the iterator. A local user who can open a performance-monitoring event and set a refresh limit can trigger this on a system whose hardware monitoring unit uses scheduling callbacks.

03

BranchIntroducedFixed inPatch commit
6.126.26.12.111723e85fb2c3b
6.186.26.18.536d0b9ce3be22
6.66.26.6.160fa360a018369
7.26.27.2.7e2770c76a024
mainline6.27.359e63416f515
cip/6.126.26.12.111-cip32723e85fb2c3b

04

State of the newest release of each Civil Infrastructure Platform SLTS tree, from CIP's own fix records.

CIP treeState
cip/6.12Fixed in 6.12.111-cip32