CVE-2026-99085: sched_ext: Fix CPU hotplug hang when a dying CPU's tasks sit in the BPF scheduler
In the Linux kernel, the following vulnerability has been resolved:
schedext: Fix CPU hotplug hang when a dying CPU's tasks sit in the BPF scheduler
A CPU going down has to empty its own rq. Its hotplug thread waits in schedcpuwaitempty() until nothing else is left, and the only thing that wakes it is balancepush(), which runs from schedule() on the dying CPU and pushes off the migratable tasks the CPU picks.
schedext breaks this. A task sitting on a user DSQ or held by the BPF scheduler is still counted on its rq, but an inactive CPU no longer calls ops.dispatch() and can't pull the task back. The dying CPU goes idle with the task still counted, and one of two things happens:
- Another CPU consumes the task. The rq empties without a schedule() on the dying CPU, so the hotplug thread is never woken and cpudown() hangs holding cpuhotpluglock. - The task is affine only to the dying CPU. Nothing can move it, and the offline stalls until the watchdog ejects the BPF scheduler.
When the rq goes offline, re-enqueue every task on it that isn't already on the local DSQ. An enqueue on an offline rq lands on the local DSQ, so the dying CPU itself picks the tasks and balancepush() pushes them off, the same as for the other sched classes. From then on no schedext path on another CPU can pull a task off the rq.
ops.dispatch() currently stops as soon as the CPU goes inactive, and CPU hotplug then waits for an RCU grace period before taking the rq offline. A BPF-held task affine only to the dying CPU and preempted inside an RCU read-side critical section would block that grace period, and the rq would never go offline. Test SCXRQONLINE directly so that ops.dispatch() keeps running until the rq goes offline. Only the dying CPU's own hotplug thread clears the flag during teardown, and taskcanrunonremoterq() still keeps other rqs' tasks off the CPU.
Affected Software
Remediation
Recommended actions to resolve this vulnerability, in priority order.
- Upgrade
Upgrade
debian/linuxto a version that resolves this vulnerability.Fixed in 6.12.115-2 - Upgrade
Upgrade
debian/linuxto a version that resolves this vulnerability.Fixed in 7.2.12-1 - Upgrade
Upgrade
debian/linux-6.12to a version that resolves this vulnerability.Fixed in 6.12.115-2~deb12u1 - Upgrade
Upgrade
ubuntu/linuxto a version that resolves this vulnerability.Fixed in 7.3.0-10.10 - Upgrade
Upgrade
ubuntu/linux-awsto a version that resolves this vulnerability.Fixed in 7.3.0-1004.4 - Upgrade
Upgrade
ubuntu/linux-azureto a version that resolves this vulnerability.Fixed in 7.3.0-1004.4 - Upgrade
Upgrade
ubuntu/linux-azure-fdeto a version that resolves this vulnerability.Fixed in 7.3.0-1004.5 - Upgrade
Upgrade
ubuntu/linux-gcpto a version that resolves this vulnerability.Fixed in 7.3.0-1004.4 - Upgrade
Upgrade
ubuntu/linux-oracleto a version that resolves this vulnerability.Fixed in 7.3.0-1004.4 - Upgrade
Upgrade
ubuntu/linux-raspito a version that resolves this vulnerability.Fixed in 7.3.0-1004.4 - Upgrade
Upgrade
ubuntu/linux-realtimeto a version that resolves this vulnerability.Fixed in 7.3.0-10.10.1 - Upgrade
Upgrade
ubuntu/linux-riscvto a version that resolves this vulnerability.Fixed in 7.3.0-10.10.1 - Upgrade
Upgrade
debian/linuxto a version that resolves this vulnerability.Fixed in 6.1.176-1Fixed in 6.1.192-2Fixed in 6.12.117-1Fixed in 7.3.4-1 - Upgrade
Upgrade
debian/linux-6.12to a version that resolves this vulnerability.Fixed in 6.12.117-1~deb12u1 - Compensating control
In sched_ext, test SCX_RQ_ONLINE directly so ops.dispatch() continues running until the rq goes offline; when the rq goes offline, re-enqueue every task on it that is not already on the local DSQ.
Event History
Frequently Asked Questions
Which systems are exposed to this hang?
Linux systems using sched_ext with tasks held in a user DSQ or by the BPF scheduler during CPU offlining are affected. The failure occurs when those tasks remain counted on the dying CPU's run queue after that CPU stops dispatching work.
How does task affinity change the observed impact?
If another CPU can consume the task, the run queue can empty without waking the hotplug thread, leaving cpu_down() hung while holding cpu_hotplug_lock. If the task is affine only to the dying CPU, offlining stalls until the watchdog ejects the BPF scheduler.
What should be done to remediate this issue?
Use a Linux kernel containing the sched_ext CPU-hotplug fix. The listed stable-kernel commits identify backported fixes, but the available data does not map them to specific kernel release versions.