In the Linux kernel, the following vulnerability has been resolved: sched_ext: Wait for SCX_OPSS_DISPATCHING before reenqueueing a task ebf1ccff79c4… (CVE-2026-98864)

The Linux kernel scheduler extension had a flaw where the reenqueue path did not wait for the SCX_OPSS_DISPATCHING state to clear before dequeueing a task. This caused a race condition where ops.enqueue() could set the task state to SCX_OPSS_QUEUED before the dispatch completed, but the dispatcher's final store would overwrite this state with SCX_OPSS_NONE, leading to dropped dispatches. The fix involves waiting for SCX_OPSS_DISPATCHING to clear before dequeueing a task for reenqueue, aligning this behavior with ops_dequeue().

This vulnerability could cause tasks to be dropped from dispatching due to premature state changes, potentially affecting task scheduling reliability and system stability. No direct exploitation or broader impact details are provided.

A fix has been implemented in the Linux kernel to address this issue by modifying the scheduler extension to wait for the SCX_OPSS_DISPATCHING state to clear before reenqueueing tasks. Users should apply the official kernel updates containing this fix when available.