1 files changed, 38 insertions, 49 deletions

diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 05dc3c17cb62..3f0b6d1936e8 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -125,8 +125,8 @@ struct async_submit_bio {
  * Different roots are used for different purposes and may nest inside each
  * other and they require separate keysets.  As lockdep keys should be
  * static, assign keysets according to the purpose of the root as indicated
- * by btrfs_root->objectid.  This ensures that all special purpose roots
- * have separate keysets.
+ * by btrfs_root->root_key.objectid.  This ensures that all special purpose
+ * roots have separate keysets.
  *
  * Lock-nesting across peer nodes is always done with the immediate parent
  * node locked thus preventing deadlock.  As lockdep doesn't know this, use
@@ -1148,7 +1148,6 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
 	root->state = 0;
 	root->orphan_cleanup_state = 0;
 
-	root->objectid = objectid;
 	root->last_trans = 0;
 	root->highest_objectid = 0;
 	root->nr_delalloc_inodes = 0;
@@ -1665,9 +1664,8 @@ static int cleaner_kthread(void *arg)
 	struct btrfs_root *root = arg;
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	int again;
-	struct btrfs_trans_handle *trans;
 
-	do {
+	while (1) {
 		again = 0;
 
 		/* Make the cleaner go to sleep early. */
@@ -1716,42 +1714,16 @@ static int cleaner_kthread(void *arg)
 		 */
 		btrfs_delete_unused_bgs(fs_info);
 sleep:
+		if (kthread_should_park())
+			kthread_parkme();
+		if (kthread_should_stop())
+			return 0;
 		if (!again) {
 			set_current_state(TASK_INTERRUPTIBLE);
-			if (!kthread_should_stop())
-				schedule();
+			schedule();
 			__set_current_state(TASK_RUNNING);
 		}
-	} while (!kthread_should_stop());
-
-	/*
-	 * Transaction kthread is stopped before us and wakes us up.
-	 * However we might have started a new transaction and COWed some
-	 * tree blocks when deleting unused block groups for example. So
-	 * make sure we commit the transaction we started to have a clean
-	 * shutdown when evicting the btree inode - if it has dirty pages
-	 * when we do the final iput() on it, eviction will trigger a
-	 * writeback for it which will fail with null pointer dereferences
-	 * since work queues and other resources were already released and
-	 * destroyed by the time the iput/eviction/writeback is made.
-	 */
-	trans = btrfs_attach_transaction(root);
-	if (IS_ERR(trans)) {
-		if (PTR_ERR(trans) != -ENOENT)
-			btrfs_err(fs_info,
-				  "cleaner transaction attach returned %ld",
-				  PTR_ERR(trans));
-	} else {
-		int ret;
-
-		ret = btrfs_commit_transaction(trans);
-		if (ret)
-			btrfs_err(fs_info,
-				  "cleaner open transaction commit returned %d",
-				  ret);
 	}
-
-	return 0;
 }
 
 static int transaction_kthread(void *arg)
@@ -2156,9 +2128,8 @@ static void btrfs_init_dev_replace_locks(struct btrfs_fs_info *fs_info)
 {
 	mutex_init(&fs_info->dev_replace.lock_finishing_cancel_unmount);
 	rwlock_init(&fs_info->dev_replace.lock);
-	atomic_set(&fs_info->dev_replace.read_locks, 0);
 	atomic_set(&fs_info->dev_replace.blocking_readers, 0);
-	init_waitqueue_head(&fs_info->replace_wait);
+	init_waitqueue_head(&fs_info->dev_replace.replace_wait);
 	init_waitqueue_head(&fs_info->dev_replace.read_lock_wq);
 }
 
@@ -2648,7 +2619,8 @@ int open_ctree(struct super_block *sb,
 		goto fail_dirty_metadata_bytes;
 	}
 
-	ret = percpu_counter_init(&fs_info->bio_counter, 0, GFP_KERNEL);
+	ret = percpu_counter_init(&fs_info->dev_replace.bio_counter, 0,
+			GFP_KERNEL);
 	if (ret) {
 		err = ret;
 		goto fail_delalloc_bytes;
@@ -3309,7 +3281,7 @@ fail_iput:
 
 	iput(fs_info->btree_inode);
 fail_bio_counter:
-	percpu_counter_destroy(&fs_info->bio_counter);
+	percpu_counter_destroy(&fs_info->dev_replace.bio_counter);
 fail_delalloc_bytes:
 	percpu_counter_destroy(&fs_info->delalloc_bytes);
 fail_dirty_metadata_bytes:
@@ -3932,6 +3904,13 @@ void close_ctree(struct btrfs_fs_info *fs_info)
 	int ret;
 
 	set_bit(BTRFS_FS_CLOSING_START, &fs_info->flags);
+	/*
+	 * We don't want the cleaner to start new transactions, add more delayed
+	 * iputs, etc. while we're closing. We can't use kthread_stop() yet
+	 * because that frees the task_struct, and the transaction kthread might
+	 * still try to wake up the cleaner.
+	 */
+	kthread_park(fs_info->cleaner_kthread);
 
 	/* wait for the qgroup rescan worker to stop */
 	btrfs_qgroup_wait_for_completion(fs_info, false);
@@ -3959,9 +3938,8 @@ void close_ctree(struct btrfs_fs_info *fs_info)
 
 	if (!sb_rdonly(fs_info->sb)) {
 		/*
-		 * If the cleaner thread is stopped and there are
-		 * block groups queued for removal, the deletion will be
-		 * skipped when we quit the cleaner thread.
+		 * The cleaner kthread is stopped, so do one final pass over
+		 * unused block groups.
 		 */
 		btrfs_delete_unused_bgs(fs_info);
 
@@ -3977,6 +3955,7 @@ void close_ctree(struct btrfs_fs_info *fs_info)
 	kthread_stop(fs_info->transaction_kthread);
 	kthread_stop(fs_info->cleaner_kthread);
 
+	ASSERT(list_empty(&fs_info->delayed_iputs));
 	set_bit(BTRFS_FS_CLOSING_DONE, &fs_info->flags);
 
 	btrfs_free_qgroup_config(fs_info);
@@ -4018,7 +3997,7 @@ void close_ctree(struct btrfs_fs_info *fs_info)
 
 	percpu_counter_destroy(&fs_info->dirty_metadata_bytes);
 	percpu_counter_destroy(&fs_info->delalloc_bytes);
-	percpu_counter_destroy(&fs_info->bio_counter);
+	percpu_counter_destroy(&fs_info->dev_replace.bio_counter);
 	cleanup_srcu_struct(&fs_info->subvol_srcu);
 
 	btrfs_free_stripe_hash_table(fs_info);
@@ -4204,7 +4183,7 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 		return ret;
 	}
 
-	while ((node = rb_first(&delayed_refs->href_root)) != NULL) {
+	while ((node = rb_first_cached(&delayed_refs->href_root)) != NULL) {
 		struct btrfs_delayed_ref_head *head;
 		struct rb_node *n;
 		bool pin_bytes = false;
@@ -4222,11 +4201,11 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 			continue;
 		}
 		spin_lock(&head->lock);
-		while ((n = rb_first(&head->ref_tree)) != NULL) {
+		while ((n = rb_first_cached(&head->ref_tree)) != NULL) {
 			ref = rb_entry(n, struct btrfs_delayed_ref_node,
 				       ref_node);
 			ref->in_tree = 0;
-			rb_erase(&ref->ref_node, &head->ref_tree);
+			rb_erase_cached(&ref->ref_node, &head->ref_tree);
 			RB_CLEAR_NODE(&ref->ref_node);
 			if (!list_empty(&ref->add_list))
 				list_del(&ref->add_list);
@@ -4240,7 +4219,7 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 		if (head->processing == 0)
 			delayed_refs->num_heads_ready--;
 		atomic_dec(&delayed_refs->num_entries);
-		rb_erase(&head->href_node, &delayed_refs->href_root);
+		rb_erase_cached(&head->href_node, &delayed_refs->href_root);
 		RB_CLEAR_NODE(&head->href_node);
 		spin_unlock(&head->lock);
 		spin_unlock(&delayed_refs->lock);
@@ -4359,13 +4338,23 @@ static int btrfs_destroy_pinned_extent(struct btrfs_fs_info *fs_info,
 	unpin = pinned_extents;
 again:
 	while (1) {
+		/*
+		 * The btrfs_finish_extent_commit() may get the same range as
+		 * ours between find_first_extent_bit and clear_extent_dirty.
+		 * Hence, hold the unused_bg_unpin_mutex to avoid double unpin
+		 * the same extent range.
+		 */
+		mutex_lock(&fs_info->unused_bg_unpin_mutex);
 		ret = find_first_extent_bit(unpin, 0, &start, &end,
 					    EXTENT_DIRTY, NULL);
-		if (ret)
+		if (ret) {
+			mutex_unlock(&fs_info->unused_bg_unpin_mutex);
 			break;
+		}
 
 		clear_extent_dirty(unpin, start, end);
 		btrfs_error_unpin_extent_range(fs_info, start, end);
+		mutex_unlock(&fs_info->unused_bg_unpin_mutex);
 		cond_resched();
 	}