CVE-2026-43482

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> sched_ext: Disable preemption between scx_claim_exit() and kicking helper work<br /> <br /> scx_claim_exit() atomically sets exit_kind, which prevents scx_error() from<br /> triggering further error handling. After claiming exit, the caller must kick<br /> the helper kthread work which initiates bypass mode and teardown.<br /> <br /> If the calling task gets preempted between claiming exit and kicking the<br /> helper work, and the BPF scheduler fails to schedule it back (since error<br /> handling is now disabled), the helper work is never queued, bypass mode<br /> never activates, tasks stop being dispatched, and the system wedges.<br /> <br /> Disable preemption across scx_claim_exit() and the subsequent work kicking<br /> in all callers - scx_disable() and scx_vexit(). Add<br /> lockdep_assert_preemption_disabled() to scx_claim_exit() to enforce the<br /> requirement.