[GIT PULL] workqueue fixes for v4.0-rc3

[Tejun Heo]

Hello, Linus.

One fix patch for a subtle livelock condition which can happen on
PREEMPT_NONE kernels involving two racing cancel_work calls.  Whoever
comes in the second has to wait for the previous one to finish.  This
was implemented by making the later one block for the same condition
that the former would be (work item completion) and then loop and
retest; unfortunately, depending on the wake up order, the later one
could lock out the former one to finish by busy looping on the cpu.

This is fixed by implementing explicit wait mechanism.  Work item
might not belong anywhere at this point and there's remote possibility
of thundering herd problem.  I originally tried to use bit_waitqueue
but it didn't work for static work items on modules.  It's currently
using single wait queue with filtering wake up function and exclusive
wakeup.  If this ever becomes a problem, which is not very likely, we
can try to figure out a way to piggy back on bit_waitqueue.

Thanks.

The following changes since commit c517d838eb7d07bbe9507871fab3931deccff539:

  Linux 4.0-rc1 (2015-02-22 18:21:14 -0800)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq.git for-4.0-fixes

for you to fetch changes up to 8603e1b30027f943cc9c1eef2b291d42c3347af1:

  workqueue: fix hang involving racing cancel[_delayed]_work_sync()'s for PREEMPT_NONE (2015-03-05 08:04:13 -0500)

----------------------------------------------------------------
Tejun Heo (1):
      workqueue: fix hang involving racing cancel[_delayed]_work_sync()'s for PREEMPT_NONE

 include/linux/workqueue.h |  3 ++-
 kernel/workqueue.c        | 56 +++++++++++++++++++++++++++++++++++++++++++----
 2 files changed, 54 insertions(+), 5 deletions(-)

diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
index 74db135..f597846 100644
--- a/include/linux/workqueue.h
+++ b/include/linux/workqueue.h
@@ -70,7 +70,8 @@ enum {
 	/* data contains off-queue information when !WORK_STRUCT_PWQ */
 	WORK_OFFQ_FLAG_BASE	= WORK_STRUCT_COLOR_SHIFT,
 
-	WORK_OFFQ_CANCELING	= (1 << WORK_OFFQ_FLAG_BASE),
+	__WORK_OFFQ_CANCELING	= WORK_OFFQ_FLAG_BASE,
+	WORK_OFFQ_CANCELING	= (1 << __WORK_OFFQ_CANCELING),
 
 	/*
 	 * When a work item is off queue, its high bits point to the last
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index f288493..41ff75b 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -2728,19 +2728,57 @@ bool flush_work(struct work_struct *work)
 }
 EXPORT_SYMBOL_GPL(flush_work);
 
+struct cwt_wait {
+	wait_queue_t		wait;
+	struct work_struct	*work;
+};
+
+static int cwt_wakefn(wait_queue_t *wait, unsigned mode, int sync, void *key)
+{
+	struct cwt_wait *cwait = container_of(wait, struct cwt_wait, wait);
+
+	if (cwait->work != key)
+		return 0;
+	return autoremove_wake_function(wait, mode, sync, key);
+}
+
 static bool __cancel_work_timer(struct work_struct *work, bool is_dwork)
 {
+	static DECLARE_WAIT_QUEUE_HEAD(cancel_waitq);
 	unsigned long flags;
 	int ret;
 
 	do {
 		ret = try_to_grab_pending(work, is_dwork, &flags);
 		/*
-		 * If someone else is canceling, wait for the same event it
-		 * would be waiting for before retrying.
+		 * If someone else is already canceling, wait for it to
+		 * finish.  flush_work() doesn't work for PREEMPT_NONE
+		 * because we may get scheduled between @work's completion
+		 * and the other canceling task resuming and clearing
+		 * CANCELING - flush_work() will return false immediately
+		 * as @work is no longer busy, try_to_grab_pending() will
+		 * return -ENOENT as @work is still being canceled and the
+		 * other canceling task won't be able to clear CANCELING as
+		 * we're hogging the CPU.
+		 *
+		 * Let's wait for completion using a waitqueue.  As this
+		 * may lead to the thundering herd problem, use a custom
+		 * wake function which matches @work along with exclusive
+		 * wait and wakeup.
 		 */
-		if (unlikely(ret == -ENOENT))
-			flush_work(work);
+		if (unlikely(ret == -ENOENT)) {
+			struct cwt_wait cwait;
+
+			init_wait(&cwait.wait);
+			cwait.wait.func = cwt_wakefn;
+			cwait.work = work;
+
+			prepare_to_wait_exclusive(&cancel_waitq, &cwait.wait,
+						  TASK_UNINTERRUPTIBLE);
+			if (work_is_canceling(work))
+				schedule();
+			finish_wait(&cancel_waitq, &cwait.wait);
+		}
 	} while (unlikely(ret < 0));
 
 	/* tell other tasks trying to grab @work to back off */
@@ -2749,6 +2787,16 @@ static bool __cancel_work_timer(struct work_struct *work, bool is_dwork)
 
 	flush_work(work);
 	clear_work_data(work);
+
+	/*
+	 * Paired with prepare_to_wait() above so that either
+	 * waitqueue_active() is visible here or !work_is_canceling() is
+	 * visible there.
+	 */
+	smp_mb();
+	if (waitqueue_active(&cancel_waitq))
+		__wake_up(&cancel_waitq, TASK_NORMAL, 1, work);
+
 	return ret;
 }
 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/