md: protect against crash upon fsync on ro array

If an fsync occurs on a read-only array, we need to send a
completion for the IO and may not increment the active IO count.
Otherwise, we hit a bug trace and can't stop the MD array anymore.

By advice of Christoph Hellwig we return success upon a flush
request but we return -EROFS for other writes.
We detect flush requests by checking if the bio has zero sectors.

This patch is suitable to any -stable kernel to which it applies.

Cc: Christoph Hellwig <[e-mail]>
Cc: Ben Hutchings <[e-mail]>
Cc: NeilBrown <[e-mail]>
Cc: [e-mail]
Signed-off-by: Sebastian Riemer <[e-mail]>
Reported-by: Ben Hutchings <[e-mail]>
Acked-by: Paul Menzel <[e-mail]>
Signed-off-by: NeilBrown <[e-mail]>

drivers/md/md.c

···
 		bio_io_error(bio);
 		return;
 	}
+	if (mddev->ro == 1 && unlikely(rw == WRITE)) {
+		bio_endio(bio, bio_sectors(bio) == 0 ? 0 : -EROFS);
+		return;
+	}
 	smp_rmb(); /* Ensure implications of  'active' are visible */
 	rcu_read_lock();
 	if (mddev->suspended) {