destroy_workqueue() can livelock
PATCH 1/1 destroy_workqueue() can livelock
Pointed out by Michal Schmidt <[email protected]>.
The bug was introduced in 2.6.22 by me.
cleanup_workqueue_thread() does flush_cpu_workqueue(cwq) in a loop until ->worklist becomes empty. This is live-lockable, a re-niced caller can get CPU after wake_up() and insert a new barrier before the lower-priority cwq->thread has a chance to clear ->current_work.
Change cleanup_workqueue_thread() to do flush_cpu_workqueue(cwq) only once. We can rely on the fact that run_workqueue() won't return until it flushes all works. So it is safe to call kthread_stop() after that, the "should stop" request won't be noticed until run_workqueue() returns.
kernel/workqueue.c renamed from kernel/workqueue.c~LIVELOCK +5 -6
@@ -739,18 +739,17 @@ static void cleanup_workqueue_thread(strif (cwq->thread ==NULL )return ; +flush_cpu_workqueue (cwq);/* -* If the caller is CPU_DEAD the single flush_cpu_workqueue() -* is not enough, a concurrent flush_workqueue() can insert a -* barrier after us. +* If the caller is CPU_DEAD and cwq->worklist was not empty, +* a concurrent flush_workqueue() can insert a barrier after us. +* However, in that case run_workqueue() won't return and check +* kthread_should_stop() until it flushes all work_struct's. * When ->worklist becomes empty it is safe to exit because no * more work_structs can be queued on this cwq: flush_workqueue * checks list_empty(), and a "normal" queue_work() can't use * a dead CPU. */ -while (flush_cpu_workqueue (cwq)) - ; -kthread_stop (cwq->thread); cwq->thread =NULL ; }
Quoted 9 lines · Oleg Nesterov
Pointed out by Michal Schmidt <[email protected]>.
The bug was introduced in 2.6.22 by me.
cleanup_workqueue_thread()doesflush_cpu_workqueue(cwq)in a loop until ->worklist becomes empty. This is live-lockable, a re-niced caller can get CPU afterwake_up()and insert a new barrier before the lower-prioritycwq->threadhas a chance to clear ->current_work.
Quoted 6 lines
Change
cleanup_workqueue_thread()to doflush_cpu_workqueue(cwq)only once. We can rely on the fact thatrun_workqueue()won't return until it flushes all works. So it is safe to callkthread_stop()after that, the "should stop" request won't be noticed untilrun_workqueue()returns.Signed-off-by Oleg Nesterov <[email protected]>
I confirm the patch fixes the bug I was seeing.
Michal