md: flush ->event_work before stopping array.

commit ee5d004fd0591536a061451eba2b187092e9127c upstream.

The 'event_work' worker used by dm-raid may still be running
when the array is stopped.  This can result in an oops.

So flush the workqueue on which it is run after detaching
and before destroying the device.

Reported-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: NeilBrown <neilb@suse.com>
Fixes: 9d09e663d550 ("dm: raid456 basic support")
[ luis: backported to 3.16: adjusted context ]
Signed-off-by: Luis Henriques <luis.henriques@canonical.com>

diff --git a/drivers/md/md.c b/drivers/md/md.c
index b6eddaf97541a421e0c0337cd65eb5a1c39bc301..9fc6c5de06386a70003c379fbea06e0fbf83aac9 100644 (file)
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -5264,6 +5264,8 @@ EXPORT_SYMBOL_GPL(md_stop_writes);
 
 static void __md_stop(struct mddev *mddev)
 {
+       /* Ensure ->event_work is done */
+       flush_workqueue(md_misc_wq);
        mddev->ready = 0;
        mddev->pers->stop(mddev);
        if (mddev->pers->sync_request && mddev->to_remove == NULL)