worker_thread: fix racy try_to_freeze() usage

worker_thread() can miss freeze_process()->signal_wake_up() if it happens
between try_to_freeze() and prepare_to_wait().  We should check freezing()
before entering schedule().

This race was introduced by me in

	[PATCH 1/1] workqueue: don't migrate pending works from the dead CPU

Looks like mm/vmscan.c:kswapd() has the same race.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This commit is contained in:
Oleg Nesterov 2007-05-09 02:34:20 -07:00 committed by Linus Torvalds
parent b9aac8e0d3
commit 85f4186af9

View file

@ -308,14 +308,14 @@ static int worker_thread(void *__cwq)
do_sigaction(SIGCHLD, &sa, (struct k_sigaction *)0); do_sigaction(SIGCHLD, &sa, (struct k_sigaction *)0);
for (;;) { for (;;) {
if (cwq->wq->freezeable)
try_to_freeze();
prepare_to_wait(&cwq->more_work, &wait, TASK_INTERRUPTIBLE); prepare_to_wait(&cwq->more_work, &wait, TASK_INTERRUPTIBLE);
if (!cwq->should_stop && list_empty(&cwq->worklist)) if (!freezing(current) && !cwq->should_stop
&& list_empty(&cwq->worklist))
schedule(); schedule();
finish_wait(&cwq->more_work, &wait); finish_wait(&cwq->more_work, &wait);
try_to_freeze();
if (cwq_should_stop(cwq)) if (cwq_should_stop(cwq))
break; break;