1 files changed, 150 insertions, 130 deletions

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 8e344b4b3444..cf3733c90c47 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -70,11 +70,7 @@
 #include "md-bitmap.h"
 #include "md-cluster.h"
 
-/* pers_list is a list of registered personalities protected
- * by pers_lock.
- * pers_lock does extra service to protect accesses to
- * mddev->thread when the mutex cannot be held.
- */
+/* pers_list is a list of registered personalities protected by pers_lock. */
 static LIST_HEAD(pers_list);
 static DEFINE_SPINLOCK(pers_lock);
 
@@ -87,23 +83,13 @@ static struct module *md_cluster_mod;
 static DECLARE_WAIT_QUEUE_HEAD(resync_wait);
 static struct workqueue_struct *md_wq;
 static struct workqueue_struct *md_misc_wq;
-static struct workqueue_struct *md_rdev_misc_wq;
+struct workqueue_struct *md_bitmap_wq;
 
 static int remove_and_add_spares(struct mddev *mddev,
 				 struct md_rdev *this);
 static void mddev_detach(struct mddev *mddev);
-
-enum md_ro_state {
-	MD_RDWR,
-	MD_RDONLY,
-	MD_AUTO_READ,
-	MD_MAX_STATE
-};
-
-static bool md_is_rdwr(struct mddev *mddev)
-{
-	return (mddev->ro == MD_RDWR);
-}
+static void export_rdev(struct md_rdev *rdev, struct mddev *mddev);
+static void md_wakeup_thread_directly(struct md_thread __rcu *thread);
 
 /*
  * Default number of read corrections we'll attempt on an rdev
@@ -360,10 +346,6 @@ EXPORT_SYMBOL_GPL(md_new_event);
 static LIST_HEAD(all_mddevs);
 static DEFINE_SPINLOCK(all_mddevs_lock);
 
-static bool is_md_suspended(struct mddev *mddev)
-{
-	return percpu_ref_is_dying(&mddev->active_io);
-}
 /* Rather than calling directly into the personality make_request function,
  * IO requests come here first so that we can check if the device is
  * being suspended pending a reconfiguration.
@@ -457,13 +439,19 @@ static void md_submit_bio(struct bio *bio)
  */
 void mddev_suspend(struct mddev *mddev)
 {
-	WARN_ON_ONCE(mddev->thread && current == mddev->thread->tsk);
-	lockdep_assert_held(&mddev->reconfig_mutex);
+	struct md_thread *thread = rcu_dereference_protected(mddev->thread,
+			lockdep_is_held(&mddev->reconfig_mutex));
+
+	WARN_ON_ONCE(thread && current == thread->tsk);
 	if (mddev->suspended++)
 		return;
 	wake_up(&mddev->sb_wait);
 	set_bit(MD_ALLOW_SB_UPDATE, &mddev->flags);
 	percpu_ref_kill(&mddev->active_io);
+
+	if (mddev->pers->prepare_suspend)
+		mddev->pers->prepare_suspend(mddev);
+
 	wait_event(mddev->sb_wait, percpu_ref_is_zero(&mddev->active_io));
 	mddev->pers->quiesce(mddev, 1);
 	clear_bit_unlock(MD_ALLOW_SB_UPDATE, &mddev->flags);
@@ -655,9 +643,11 @@ void mddev_init(struct mddev *mddev)
 {
 	mutex_init(&mddev->open_mutex);
 	mutex_init(&mddev->reconfig_mutex);
+	mutex_init(&mddev->delete_mutex);
 	mutex_init(&mddev->bitmap_info.mutex);
 	INIT_LIST_HEAD(&mddev->disks);
 	INIT_LIST_HEAD(&mddev->all_mddevs);
+	INIT_LIST_HEAD(&mddev->deleting);
 	timer_setup(&mddev->safemode_timer, md_safemode_timeout, 0);
 	atomic_set(&mddev->active, 1);
 	atomic_set(&mddev->openers, 0);
@@ -759,6 +749,24 @@ static void mddev_free(struct mddev *mddev)
 
 static const struct attribute_group md_redundancy_group;
 
+static void md_free_rdev(struct mddev *mddev)
+{
+	struct md_rdev *rdev;
+	struct md_rdev *tmp;
+
+	mutex_lock(&mddev->delete_mutex);
+	if (list_empty(&mddev->deleting))
+		goto out;
+
+	list_for_each_entry_safe(rdev, tmp, &mddev->deleting, same_set) {
+		list_del_init(&rdev->same_set);
+		kobject_del(&rdev->kobj);
+		export_rdev(rdev, mddev);
+	}
+out:
+	mutex_unlock(&mddev->delete_mutex);
+}
+
 void mddev_unlock(struct mddev *mddev)
 {
 	if (mddev->to_remove) {
@@ -800,13 +808,10 @@ void mddev_unlock(struct mddev *mddev)
 	} else
 		mutex_unlock(&mddev->reconfig_mutex);
 
-	/* As we've dropped the mutex we need a spinlock to
-	 * make sure the thread doesn't disappear
-	 */
-	spin_lock(&pers_lock);
+	md_free_rdev(mddev);
+
 	md_wakeup_thread(mddev->thread);
 	wake_up(&mddev->sb_wait);
-	spin_unlock(&pers_lock);
 }
 EXPORT_SYMBOL_GPL(mddev_unlock);
 
@@ -938,7 +943,7 @@ void md_super_write(struct mddev *mddev, struct md_rdev *rdev,
 	atomic_inc(&rdev->nr_pending);
 
 	bio->bi_iter.bi_sector = sector;
-	bio_add_page(bio, page, size, 0);
+	__bio_add_page(bio, page, size, 0);
 	bio->bi_private = rdev;
 	bio->bi_end_io = super_written;
 
@@ -979,7 +984,7 @@ int sync_page_io(struct md_rdev *rdev, sector_t sector, int size,
 		bio.bi_iter.bi_sector = sector + rdev->new_data_offset;
 	else
 		bio.bi_iter.bi_sector = sector + rdev->data_offset;
-	bio_add_page(&bio, page, size, 0);
+	__bio_add_page(&bio, page, size, 0);
 
 	submit_bio_wait(&bio);
 
@@ -2440,16 +2445,12 @@ static int bind_rdev_to_array(struct md_rdev *rdev, struct mddev *mddev)
 	return err;
 }
 
-static void rdev_delayed_delete(struct work_struct *ws)
-{
-	struct md_rdev *rdev = container_of(ws, struct md_rdev, del_work);
-	kobject_del(&rdev->kobj);
-	kobject_put(&rdev->kobj);
-}
-
 void md_autodetect_dev(dev_t dev);
 
-static void export_rdev(struct md_rdev *rdev)
+/* just for claiming the bdev */
+static struct md_rdev claim_rdev;
+
+static void export_rdev(struct md_rdev *rdev, struct mddev *mddev)
 {
 	pr_debug("md: export_rdev(%pg)\n", rdev->bdev);
 	md_rdev_clear(rdev);
@@ -2457,13 +2458,15 @@ static void export_rdev(struct md_rdev *rdev)
 	if (test_bit(AutoDetected, &rdev->flags))
 		md_autodetect_dev(rdev->bdev->bd_dev);
 #endif
-	blkdev_put(rdev->bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL);
+	blkdev_put(rdev->bdev, mddev->major_version == -2 ? &claim_rdev : rdev);
 	rdev->bdev = NULL;
 	kobject_put(&rdev->kobj);
 }
 
 static void md_kick_rdev_from_array(struct md_rdev *rdev)
 {
+	struct mddev *mddev = rdev->mddev;
+
 	bd_unlink_disk_holder(rdev->bdev, rdev->mddev->gendisk);
 	list_del_rcu(&rdev->same_set);
 	pr_debug("md: unbind<%pg>\n", rdev->bdev);
@@ -2477,15 +2480,17 @@ static void md_kick_rdev_from_array(struct md_rdev *rdev)
 	rdev->sysfs_unack_badblocks = NULL;
 	rdev->sysfs_badblocks = NULL;
 	rdev->badblocks.count = 0;
-	/* We need to delay this, otherwise we can deadlock when
-	 * writing to 'remove' to "dev/state".  We also need
-	 * to delay it due to rcu usage.
-	 */
+
 	synchronize_rcu();
-	INIT_WORK(&rdev->del_work, rdev_delayed_delete);
-	kobject_get(&rdev->kobj);
-	queue_work(md_rdev_misc_wq, &rdev->del_work);
-	export_rdev(rdev);
+
+	/*
+	 * kobject_del() will wait for all in progress writers to be done, where
+	 * reconfig_mutex is held, hence it can't be called under
+	 * reconfig_mutex and it's delayed to mddev_unlock().
+	 */
+	mutex_lock(&mddev->delete_mutex);
+	list_add(&rdev->same_set, &mddev->deleting);
+	mutex_unlock(&mddev->delete_mutex);
 }
 
 static void export_array(struct mddev *mddev)
@@ -3553,6 +3558,7 @@ rdev_attr_store(struct kobject *kobj, struct attribute *attr,
 {
 	struct rdev_sysfs_entry *entry = container_of(attr, struct rdev_sysfs_entry, attr);
 	struct md_rdev *rdev = container_of(kobj, struct md_rdev, kobj);
+	struct kernfs_node *kn = NULL;
 	ssize_t rv;
 	struct mddev *mddev = rdev->mddev;
 
@@ -3560,6 +3566,10 @@ rdev_attr_store(struct kobject *kobj, struct attribute *attr,
 		return -EIO;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EACCES;
+
+	if (entry->store == state_store && cmd_match(page, "remove"))
+		kn = sysfs_break_active_protection(kobj, attr);
+
 	rv = mddev ? mddev_lock(mddev) : -ENODEV;
 	if (!rv) {
 		if (rdev->mddev == NULL)
@@ -3568,6 +3578,10 @@ rdev_attr_store(struct kobject *kobj, struct attribute *attr,
 			rv = entry->store(rdev, page, length);
 		mddev_unlock(mddev);
 	}
+
+	if (kn)
+		sysfs_unbreak_active_protection(kn);
+
 	return rv;
 }
 
@@ -3612,6 +3626,7 @@ int md_rdev_init(struct md_rdev *rdev)
 	return badblocks_init(&rdev->badblocks, 0);
 }
 EXPORT_SYMBOL_GPL(md_rdev_init);
+
 /*
  * Import a device. If 'super_format' >= 0, then sanity check the superblock
  *
@@ -3624,7 +3639,6 @@ EXPORT_SYMBOL_GPL(md_rdev_init);
  */
 static struct md_rdev *md_import_device(dev_t newdev, int super_format, int super_minor)
 {
-	static struct md_rdev claim_rdev; /* just for claiming the bdev */
 	struct md_rdev *rdev;
 	sector_t size;
 	int err;
@@ -3640,9 +3654,8 @@ static struct md_rdev *md_import_device(dev_t newdev, int super_format, int supe
 	if (err)
 		goto out_clear_rdev;
 
-	rdev->bdev = blkdev_get_by_dev(newdev,
-			FMODE_READ | FMODE_WRITE | FMODE_EXCL,
-			super_format == -2 ? &claim_rdev : rdev);
+	rdev->bdev = blkdev_get_by_dev(newdev, BLK_OPEN_READ | BLK_OPEN_WRITE,
+			super_format == -2 ? &claim_rdev : rdev, NULL);
 	if (IS_ERR(rdev->bdev)) {
 		pr_warn("md: could not open device unknown-block(%u,%u).\n",
 			MAJOR(newdev), MINOR(newdev));
@@ -3679,7 +3692,7 @@ static struct md_rdev *md_import_device(dev_t newdev, int super_format, int supe
 	return rdev;
 
 out_blkdev_put:
-	blkdev_put(rdev->bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL);
+	blkdev_put(rdev->bdev, super_format == -2 ? &claim_rdev : rdev);
 out_clear_rdev:
 	md_rdev_clear(rdev);
 out_free_rdev:
@@ -3794,8 +3807,9 @@ int strict_strtoul_scaled(const char *cp, unsigned long *res, int scale)
 static ssize_t
 safe_delay_show(struct mddev *mddev, char *page)
 {
-	int msec = (mddev->safemode_delay*1000)/HZ;
-	return sprintf(page, "%d.%03d\n", msec/1000, msec%1000);
+	unsigned int msec = ((unsigned long)mddev->safemode_delay*1000)/HZ;
+
+	return sprintf(page, "%u.%03u\n", msec/1000, msec%1000);
 }
 static ssize_t
 safe_delay_store(struct mddev *mddev, const char *cbuf, size_t len)
@@ -3807,7 +3821,7 @@ safe_delay_store(struct mddev *mddev, const char *cbuf, size_t len)
 		return -EINVAL;
 	}
 
-	if (strict_strtoul_scaled(cbuf, &msec, 3) < 0)
+	if (strict_strtoul_scaled(cbuf, &msec, 3) < 0 || msec > UINT_MAX / HZ)
 		return -EINVAL;
 	if (msec == 0)
 		mddev->safemode_delay = 0;
@@ -4477,6 +4491,8 @@ max_corrected_read_errors_store(struct mddev *mddev, const char *buf, size_t len
 	rv = kstrtouint(buf, 10, &n);
 	if (rv < 0)
 		return rv;
+	if (n > INT_MAX)
+		return -EINVAL;
 	atomic_set(&mddev->max_corr_read_errors, n);
 	return len;
 }
@@ -4491,20 +4507,6 @@ null_show(struct mddev *mddev, char *page)
 	return -EINVAL;
 }
 
-/* need to ensure rdev_delayed_delete() has completed */
-static void flush_rdev_wq(struct mddev *mddev)
-{
-	struct md_rdev *rdev;
-
-	rcu_read_lock();
-	rdev_for_each_rcu(rdev, mddev)
-		if (work_pending(&rdev->del_work)) {
-			flush_workqueue(md_rdev_misc_wq);
-			break;
-		}
-	rcu_read_unlock();
-}
-
 static ssize_t
 new_dev_store(struct mddev *mddev, const char *buf, size_t len)
 {
@@ -4532,7 +4534,6 @@ new_dev_store(struct mddev *mddev, const char *buf, size_t len)
 	    minor != MINOR(dev))
 		return -EOVERFLOW;
 
-	flush_rdev_wq(mddev);
 	err = mddev_lock(mddev);
 	if (err)
 		return err;
@@ -4560,7 +4561,7 @@ new_dev_store(struct mddev *mddev, const char *buf, size_t len)
 	err = bind_rdev_to_array(rdev, mddev);
  out:
 	if (err)
-		export_rdev(rdev);
+		export_rdev(rdev, mddev);
 	mddev_unlock(mddev);
 	if (!err)
 		md_new_event();
@@ -4804,11 +4805,21 @@ action_store(struct mddev *mddev, const char *page, size_t len)
 			return -EINVAL;
 		err = mddev_lock(mddev);
 		if (!err) {
-			if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
+			if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery)) {
 				err =  -EBUSY;
-			else {
+			} else if (mddev->reshape_position == MaxSector ||
+				   mddev->pers->check_reshape == NULL ||
+				   mddev->pers->check_reshape(mddev)) {
 				clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
 				err = mddev->pers->start_reshape(mddev);
+			} else {
+				/*
+				 * If reshape is still in progress, and
+				 * md_check_recovery() can continue to reshape,
+				 * don't restart reshape because data can be
+				 * corrupted for raid456.
+				 */
+				clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
 			}
 			mddev_unlock(mddev);
 		}
@@ -5592,7 +5603,6 @@ struct mddev *md_alloc(dev_t dev, char *name)
 	 * removed (mddev_delayed_delete).
 	 */
 	flush_workqueue(md_misc_wq);
-	flush_workqueue(md_rdev_misc_wq);
 
 	mutex_lock(&disks_mutex);
 	mddev = mddev_alloc(dev);
@@ -6269,10 +6279,12 @@ static int md_set_readonly(struct mddev *mddev, struct block_device *bdev)
 	}
 	if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
 		set_bit(MD_RECOVERY_INTR, &mddev->recovery);
-	if (mddev->sync_thread)
-		/* Thread might be blocked waiting for metadata update
-		 * which will now never happen */
-		wake_up_process(mddev->sync_thread->tsk);
+
+	/*
+	 * Thread might be blocked waiting for metadata update which will now
+	 * never happen
+	 */
+	md_wakeup_thread_directly(mddev->sync_thread);
 
 	if (mddev->external && test_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags))
 		return -EBUSY;
@@ -6333,10 +6345,12 @@ static int do_md_stop(struct mddev *mddev, int mode,
 	}
 	if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
 		set_bit(MD_RECOVERY_INTR, &mddev->recovery);
-	if (mddev->sync_thread)
-		/* Thread might be blocked waiting for metadata update
-		 * which will now never happen */
-		wake_up_process(mddev->sync_thread->tsk);
+
+	/*
+	 * Thread might be blocked waiting for metadata update which will now
+	 * never happen
+	 */
+	md_wakeup_thread_directly(mddev->sync_thread);
 
 	mddev_unlock(mddev);
 	wait_event(resync_wait, (mddev->sync_thread == NULL &&
@@ -6498,7 +6512,7 @@ static void autorun_devices(int part)
 			rdev_for_each_list(rdev, tmp, &candidates) {
 				list_del_init(&rdev->same_set);
 				if (bind_rdev_to_array(rdev, mddev))
-					export_rdev(rdev);
+					export_rdev(rdev, mddev);
 			}
 			autorun_array(mddev);
 			mddev_unlock(mddev);
@@ -6508,7 +6522,7 @@ static void autorun_devices(int part)
 		 */
 		rdev_for_each_list(rdev, tmp, &candidates) {
 			list_del_init(&rdev->same_set);
-			export_rdev(rdev);
+			export_rdev(rdev, mddev);
 		}
 		mddev_put(mddev);
 	}
@@ -6696,13 +6710,13 @@ int md_add_new_disk(struct mddev *mddev, struct mdu_disk_info_s *info)
 				pr_warn("md: %pg has different UUID to %pg\n",
 					rdev->bdev,
 					rdev0->bdev);
-				export_rdev(rdev);
+				export_rdev(rdev, mddev);
 				return -EINVAL;
 			}
 		}
 		err = bind_rdev_to_array(rdev, mddev);
 		if (err)
-			export_rdev(rdev);
+			export_rdev(rdev, mddev);
 		return err;
 	}
 
@@ -6733,7 +6747,6 @@ int md_add_new_disk(struct mddev *mddev, struct mdu_disk_info_s *info)
 			if (info->state & (1<<MD_DISK_SYNC)  &&
 			    info->raid_disk < mddev->raid_disks) {
 				rdev->raid_disk = info->raid_disk;
-				set_bit(In_sync, &rdev->flags);
 				clear_bit(Bitmap_sync, &rdev->flags);
 			} else
 				rdev->raid_disk = -1;
@@ -6746,7 +6759,7 @@ int md_add_new_disk(struct mddev *mddev, struct mdu_disk_info_s *info)
 			/* This was a hot-add request, but events doesn't
 			 * match, so reject it.
 			 */
-			export_rdev(rdev);
+			export_rdev(rdev, mddev);
 			return -EINVAL;
 		}
 
@@ -6772,7 +6785,7 @@ int md_add_new_disk(struct mddev *mddev, struct mdu_disk_info_s *info)
 				}
 			}
 			if (has_journal || mddev->bitmap) {
-				export_rdev(rdev);
+				export_rdev(rdev, mddev);
 				return -EBUSY;
 			}
 			set_bit(Journal, &rdev->flags);
@@ -6787,7 +6800,7 @@ int md_add_new_disk(struct mddev *mddev, struct mdu_disk_info_s *info)
 				/* --add initiated by this node */
 				err = md_cluster_ops->add_new_disk(mddev, rdev);
 				if (err) {
-					export_rdev(rdev);
+					export_rdev(rdev, mddev);
 					return err;
 				}
 			}
@@ -6797,7 +6810,7 @@ int md_add_new_disk(struct mddev *mddev, struct mdu_disk_info_s *info)
 		err = bind_rdev_to_array(rdev, mddev);
 
 		if (err)
-			export_rdev(rdev);
+			export_rdev(rdev, mddev);
 
 		if (mddev_is_clustered(mddev)) {
 			if (info->state & (1 << MD_DISK_CANDIDATE)) {
@@ -6860,7 +6873,7 @@ int md_add_new_disk(struct mddev *mddev, struct mdu_disk_info_s *info)
 
 		err = bind_rdev_to_array(rdev, mddev);
 		if (err) {
-			export_rdev(rdev);
+			export_rdev(rdev, mddev);
 			return err;
 		}
 	}
@@ -6985,7 +6998,7 @@ static int hot_add_disk(struct mddev *mddev, dev_t dev)
 	return 0;
 
 abort_export:
-	export_rdev(rdev);
+	export_rdev(rdev, mddev);
 	return err;
 }
 
@@ -7486,7 +7499,7 @@ static int __md_set_array_info(struct mddev *mddev, void __user *argp)
 	return err;
 }
 
-static int md_ioctl(struct block_device *bdev, fmode_t mode,
+static int md_ioctl(struct block_device *bdev, blk_mode_t mode,
 			unsigned int cmd, unsigned long arg)
 {
 	int err = 0;
@@ -7555,9 +7568,6 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode,
 
 	}
 
-	if (cmd == ADD_NEW_DISK || cmd == HOT_ADD_DISK)
-		flush_rdev_wq(mddev);
-
 	if (cmd == HOT_REMOVE_DISK)
 		/* need to ensure recovery thread has run */
 		wait_event_interruptible_timeout(mddev->sb_wait,
@@ -7718,7 +7728,7 @@ out:
 	return err;
 }
 #ifdef CONFIG_COMPAT
-static int md_compat_ioctl(struct block_device *bdev, fmode_t mode,
+static int md_compat_ioctl(struct block_device *bdev, blk_mode_t mode,
 		    unsigned int cmd, unsigned long arg)
 {
 	switch (cmd) {
@@ -7767,13 +7777,13 @@ out_unlock:
 	return err;
 }
 
-static int md_open(struct block_device *bdev, fmode_t mode)
+static int md_open(struct gendisk *disk, blk_mode_t mode)
 {
 	struct mddev *mddev;
 	int err;
 
 	spin_lock(&all_mddevs_lock);
-	mddev = mddev_get(bdev->bd_disk->private_data);
+	mddev = mddev_get(disk->private_data);
 	spin_unlock(&all_mddevs_lock);
 	if (!mddev)
 		return -ENODEV;
@@ -7789,7 +7799,7 @@ static int md_open(struct block_device *bdev, fmode_t mode)
 	atomic_inc(&mddev->openers);
 	mutex_unlock(&mddev->open_mutex);
 
-	bdev_check_media_change(bdev);
+	disk_check_media_change(disk);
 	return 0;
 
 out_unlock:
@@ -7799,7 +7809,7 @@ out:
 	return err;
 }
 
-static void md_release(struct gendisk *disk, fmode_t mode)
+static void md_release(struct gendisk *disk)
 {
 	struct mddev *mddev = disk->private_data;
 
@@ -7886,13 +7896,29 @@ static int md_thread(void *arg)
 	return 0;
 }
 
-void md_wakeup_thread(struct md_thread *thread)
+static void md_wakeup_thread_directly(struct md_thread __rcu *thread)
 {
-	if (thread) {
-		pr_debug("md: waking up MD thread %s.\n", thread->tsk->comm);
-		set_bit(THREAD_WAKEUP, &thread->flags);
-		wake_up(&thread->wqueue);
+	struct md_thread *t;
+
+	rcu_read_lock();
+	t = rcu_dereference(thread);
+	if (t)
+		wake_up_process(t->tsk);
+	rcu_read_unlock();
+}
+
+void md_wakeup_thread(struct md_thread __rcu *thread)
+{
+	struct md_thread *t;
+
+	rcu_read_lock();
+	t = rcu_dereference(thread);
+	if (t) {
+		pr_debug("md: waking up MD thread %s.\n", t->tsk->comm);
+		set_bit(THREAD_WAKEUP, &t->flags);
+		wake_up(&t->wqueue);
 	}
+	rcu_read_unlock();
 }
 EXPORT_SYMBOL(md_wakeup_thread);
 
@@ -7922,22 +7948,15 @@ struct md_thread *md_register_thread(void (*run) (struct md_thread *),
 }
 EXPORT_SYMBOL(md_register_thread);
 
-void md_unregister_thread(struct md_thread **threadp)
+void md_unregister_thread(struct md_thread __rcu **threadp)
 {
-	struct md_thread *thread;
+	struct md_thread *thread = rcu_dereference_protected(*threadp, true);
 
-	/*
-	 * Locking ensures that mddev_unlock does not wake_up a
-	 * non-existent thread
-	 */
-	spin_lock(&pers_lock);
-	thread = *threadp;
-	if (!thread) {
-		spin_unlock(&pers_lock);
+	if (!thread)
 		return;
-	}
-	*threadp = NULL;
-	spin_unlock(&pers_lock);
+
+	rcu_assign_pointer(*threadp, NULL);
+	synchronize_rcu();
 
 	pr_debug("interrupting MD-thread pid %d\n", task_pid_nr(thread->tsk));
 	kthread_stop(thread->tsk);
@@ -9100,6 +9119,7 @@ void md_do_sync(struct md_thread *thread)
 	spin_unlock(&mddev->lock);
 
 	wake_up(&resync_wait);
+	wake_up(&mddev->sb_wait);
 	md_wakeup_thread(mddev->thread);
 	return;
 }
@@ -9202,9 +9222,8 @@ static void md_start_sync(struct work_struct *ws)
 {
 	struct mddev *mddev = container_of(ws, struct mddev, del_work);
 
-	mddev->sync_thread = md_register_thread(md_do_sync,
-						mddev,
-						"resync");
+	rcu_assign_pointer(mddev->sync_thread,
+			   md_register_thread(md_do_sync, mddev, "resync"));
 	if (!mddev->sync_thread) {
 		pr_warn("%s: could not start resync thread...\n",
 			mdname(mddev));
@@ -9619,9 +9638,10 @@ static int __init md_init(void)
 	if (!md_misc_wq)
 		goto err_misc_wq;
 
-	md_rdev_misc_wq = alloc_workqueue("md_rdev_misc", 0, 0);
-	if (!md_rdev_misc_wq)
-		goto err_rdev_misc_wq;
+	md_bitmap_wq = alloc_workqueue("md_bitmap", WQ_MEM_RECLAIM | WQ_UNBOUND,
+				       0);
+	if (!md_bitmap_wq)
+		goto err_bitmap_wq;
 
 	ret = __register_blkdev(MD_MAJOR, "md", md_probe);
 	if (ret < 0)
@@ -9641,8 +9661,8 @@ static int __init md_init(void)
 err_mdp:
 	unregister_blkdev(MD_MAJOR, "md");
 err_md:
-	destroy_workqueue(md_rdev_misc_wq);
-err_rdev_misc_wq:
+	destroy_workqueue(md_bitmap_wq);
+err_bitmap_wq:
 	destroy_workqueue(md_misc_wq);
 err_misc_wq:
 	destroy_workqueue(md_wq);
@@ -9938,8 +9958,8 @@ static __exit void md_exit(void)
 	}
 	spin_unlock(&all_mddevs_lock);
 
-	destroy_workqueue(md_rdev_misc_wq);
 	destroy_workqueue(md_misc_wq);
+	destroy_workqueue(md_bitmap_wq);
 	destroy_workqueue(md_wq);
 }