Executive brief
A vulnerability in the Linux kernel's scheduling mechanism can cause the system to hang or become unresponsive. This occurs when the system is under heavy load and attempts to enable a specific type of custom task scheduler (sched_ext). An attacker or a malfunctioning process could exploit this to cause a denial-of-service, effectively stopping all operations on the affected machine.
Technical details
A race condition and priority inversion issue exists in the Linux kernel's sched_ext (Extensible Scheduler) framework. During the execution of scx_enable(), the calling thread's scheduling class is transitioned from 'fair' to 'ext'. Because the 'fair' class has a higher priority than 'ext', a saturated fair-class workload can indefinitely starve the thread responsible for completing the enablement process. This starvation leads to a kernel hang. The fix involves offloading the enablement logic to a dedicated Real-Time (SCHED_FIFO) kthread to ensure it cannot be preempted or starved by fair or ext class tasks. Patches have been released for various stable branches including 6.12.y, 6.18.y, and 6.19.y.
Affected products
- Linux Linux kernel 6.12 to 6.12.78, 6.13 to 6.18.20, 6.19 to 6.19.9, 7.0-rc1 to 7.0-rc2
Timeline
- 2026-03-03: patched: Initial patch authored by Tejun Heo
- 2026-05-08: advisory: CVE-2026-43392 published by kernel.org