Merge branch 'for-linus-4.4' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs

Pull btrfs fixes from Chris Mason:
 "This has Mark Fasheh's patches to fix quota accounting during subvol
  deletion, which we've been working on for a while now.  The patch is
  pretty small but it's a key fix.

  Otherwise it's a random assortment"

* 'for-linus-4.4' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
  btrfs: fix balance range usage filters in 4.4-rc
  btrfs: qgroup: account shared subtree during snapshot delete
  Btrfs: use btrfs_get_fs_root in resolve_indirect_ref
  btrfs: qgroup: fix quota disable during rescan
  Btrfs: fix race between cleaner kthread and space cache writeout
  Btrfs: fix scrub preventing unused block groups from being deleted
  Btrfs: fix race between scrub and block group deletion
  btrfs: fix rcu warning during device replace
  btrfs: Continue replace when set_block_ro failed
  btrfs: fix clashing number of the enhanced balance usage filter
  Btrfs: fix the number of transaction units needed to remove a block group
  Btrfs: use global reserve when deleting unused block group after ENOSPC
  Btrfs: tests: checking for NULL instead of IS_ERR()
  btrfs: fix signed overflows in btrfs_sync_file

fs/btrfs/backref.c

@@ -355,7 +355,7 @@ static int __resolve_indirect_ref(struct btrfs_fs_info *fs_info,
 
 	index = srcu_read_lock(&fs_info->subvol_srcu);
 
-	root = btrfs_read_fs_root_no_name(fs_info, &root_key);
+	root = btrfs_get_fs_root(fs_info, &root_key, false);
 	if (IS_ERR(root)) {
 		srcu_read_unlock(&fs_info->subvol_srcu, index);
 		ret = PTR_ERR(root);

fs/btrfs/ctree.h

@@ -3416,6 +3416,7 @@ int btrfs_cross_ref_exist(struct btrfs_trans_handle *trans,
 struct btrfs_block_group_cache *btrfs_lookup_block_group(
 						 struct btrfs_fs_info *info,
 						 u64 bytenr);
+void btrfs_get_block_group(struct btrfs_block_group_cache *cache);
 void btrfs_put_block_group(struct btrfs_block_group_cache *cache);
 int get_block_group_index(struct btrfs_block_group_cache *cache);
 struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans,
@@ -3479,6 +3480,9 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
 			   struct btrfs_root *root, u64 bytes_used,
 			   u64 type, u64 chunk_objectid, u64 chunk_offset,
 			   u64 size);
+struct btrfs_trans_handle *btrfs_start_trans_remove_block_group(
+				struct btrfs_fs_info *fs_info,
+				const u64 chunk_offset);
 int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 			     struct btrfs_root *root, u64 group_start,
 			     struct extent_map *em);

fs/btrfs/extent-tree.c

@@ -124,7 +124,7 @@ static int block_group_bits(struct btrfs_block_group_cache *cache, u64 bits)
 	return (cache->flags & bits) == bits;
 }
 
-static void btrfs_get_block_group(struct btrfs_block_group_cache *cache)
+void btrfs_get_block_group(struct btrfs_block_group_cache *cache)
 {
 	atomic_inc(&cache->count);
 }
@@ -5915,19 +5915,6 @@ static int update_block_group(struct btrfs_trans_handle *trans,
 			set_extent_dirty(info->pinned_extents,
 					 bytenr, bytenr + num_bytes - 1,
 					 GFP_NOFS | __GFP_NOFAIL);
-			/*
-			 * No longer have used bytes in this block group, queue
-			 * it for deletion.
-			 */
-			if (old_val == 0) {
-				spin_lock(&info->unused_bgs_lock);
-				if (list_empty(&cache->bg_list)) {
-					btrfs_get_block_group(cache);
-					list_add_tail(&cache->bg_list,
-						      &info->unused_bgs);
-				}
-				spin_unlock(&info->unused_bgs_lock);
-			}
 		}
 
 		spin_lock(&trans->transaction->dirty_bgs_lock);
@@ -5939,6 +5926,22 @@ static int update_block_group(struct btrfs_trans_handle *trans,
 		}
 		spin_unlock(&trans->transaction->dirty_bgs_lock);
 
+		/*
+		 * No longer have used bytes in this block group, queue it for
+		 * deletion. We do this after adding the block group to the
+		 * dirty list to avoid races between cleaner kthread and space
+		 * cache writeout.
+		 */
+		if (!alloc && old_val == 0) {
+			spin_lock(&info->unused_bgs_lock);
+			if (list_empty(&cache->bg_list)) {
+				btrfs_get_block_group(cache);
+				list_add_tail(&cache->bg_list,
+					      &info->unused_bgs);
+			}
+			spin_unlock(&info->unused_bgs_lock);
+		}
+
 		btrfs_put_block_group(cache);
 		total -= num_bytes;
 		bytenr += num_bytes;
@@ -8105,21 +8108,47 @@ static noinline void reada_walk_down(struct btrfs_trans_handle *trans,
 }
 
 /*
- * TODO: Modify related function to add related node/leaf to dirty_extent_root,
- * for later qgroup accounting.
- *
- * Current, this function does nothing.
+ * These may not be seen by the usual inc/dec ref code so we have to
+ * add them here.
  */
+static int record_one_subtree_extent(struct btrfs_trans_handle *trans,
+				     struct btrfs_root *root, u64 bytenr,
+				     u64 num_bytes)
+{
+	struct btrfs_qgroup_extent_record *qrecord;
+	struct btrfs_delayed_ref_root *delayed_refs;
+
+	qrecord = kmalloc(sizeof(*qrecord), GFP_NOFS);
+	if (!qrecord)
+		return -ENOMEM;
+
+	qrecord->bytenr = bytenr;
+	qrecord->num_bytes = num_bytes;
+	qrecord->old_roots = NULL;
+
+	delayed_refs = &trans->transaction->delayed_refs;
+	spin_lock(&delayed_refs->lock);
+	if (btrfs_qgroup_insert_dirty_extent(delayed_refs, qrecord))
+		kfree(qrecord);
+	spin_unlock(&delayed_refs->lock);
+
+	return 0;
+}
+
 static int account_leaf_items(struct btrfs_trans_handle *trans,
 			      struct btrfs_root *root,
 			      struct extent_buffer *eb)
 {
 	int nr = btrfs_header_nritems(eb);
-	int i, extent_type;
+	int i, extent_type, ret;
 	struct btrfs_key key;
 	struct btrfs_file_extent_item *fi;
 	u64 bytenr, num_bytes;
 
+	/* We can be called directly from walk_up_proc() */
+	if (!root->fs_info->quota_enabled)
+		return 0;
+
 	for (i = 0; i < nr; i++) {
 		btrfs_item_key_to_cpu(eb, &key, i);
 
@@ -8138,6 +8167,10 @@ static int account_leaf_items(struct btrfs_trans_handle *trans,
 			continue;
 
 		num_bytes = btrfs_file_extent_disk_num_bytes(eb, fi);
+
+		ret = record_one_subtree_extent(trans, root, bytenr, num_bytes);
+		if (ret)
+			return ret;
 	}
 	return 0;
 }
@@ -8206,8 +8239,6 @@ static int adjust_slots_upwards(struct btrfs_root *root,
 
 /*
  * root_eb is the subtree root and is locked before this function is called.
- * TODO: Modify this function to mark all (including complete shared node)
- * to dirty_extent_root to allow it get accounted in qgroup.
  */
 static int account_shared_subtree(struct btrfs_trans_handle *trans,
 				  struct btrfs_root *root,
@@ -8285,6 +8316,11 @@ static int account_shared_subtree(struct btrfs_trans_handle *trans,
 			btrfs_tree_read_lock(eb);
 			btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
 			path->locks[level] = BTRFS_READ_LOCK_BLOCKING;
+
+			ret = record_one_subtree_extent(trans, root, child_bytenr,
+							root->nodesize);
+			if (ret)
+				goto out;
 		}
 
 		if (level == 0) {
@@ -10256,6 +10292,47 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
+struct btrfs_trans_handle *
+btrfs_start_trans_remove_block_group(struct btrfs_fs_info *fs_info,
+				     const u64 chunk_offset)
+{
+	struct extent_map_tree *em_tree = &fs_info->mapping_tree.map_tree;
+	struct extent_map *em;
+	struct map_lookup *map;
+	unsigned int num_items;
+
+	read_lock(&em_tree->lock);
+	em = lookup_extent_mapping(em_tree, chunk_offset, 1);
+	read_unlock(&em_tree->lock);
+	ASSERT(em && em->start == chunk_offset);
+
+	/*
+	 * We need to reserve 3 + N units from the metadata space info in order
+	 * to remove a block group (done at btrfs_remove_chunk() and at
+	 * btrfs_remove_block_group()), which are used for:
+	 *
+	 * 1 unit for adding the free space inode's orphan (located in the tree
+	 * of tree roots).
+	 * 1 unit for deleting the block group item (located in the extent
+	 * tree).
+	 * 1 unit for deleting the free space item (located in tree of tree
+	 * roots).
+	 * N units for deleting N device extent items corresponding to each
+	 * stripe (located in the device tree).
+	 *
+	 * In order to remove a block group we also need to reserve units in the
+	 * system space info in order to update the chunk tree (update one or
+	 * more device items and remove one chunk item), but this is done at
+	 * btrfs_remove_chunk() through a call to check_system_chunk().
+	 */
+	map = (struct map_lookup *)em->bdev;
+	num_items = 3 + map->num_stripes;
+	free_extent_map(em);
+
+	return btrfs_start_transaction_fallback_global_rsv(fs_info->extent_root,
+							   num_items, 1);
+}
+
 /*
  * Process the unused_bgs list and remove any that don't have any allocated
  * space inside of them.
@@ -10322,8 +10399,8 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 		 * Want to do this before we do anything else so we can recover
 		 * properly if we fail to join the transaction.
 		 */
-		/* 1 for btrfs_orphan_reserve_metadata() */
-		trans = btrfs_start_transaction(root, 1);
+		trans = btrfs_start_trans_remove_block_group(fs_info,
+						     block_group->key.objectid);
 		if (IS_ERR(trans)) {
 			btrfs_dec_block_group_ro(root, block_group);
 			ret = PTR_ERR(trans);

fs/btrfs/file.c

@@ -1882,8 +1882,13 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	struct btrfs_log_ctx ctx;
 	int ret = 0;
 	bool full_sync = 0;
-	const u64 len = end - start + 1;
+	u64 len;
 
+	/*
+	 * The range length can be represented by u64, we have to do the typecasts
+	 * to avoid signed overflow if it's [0, LLONG_MAX] eg. from fsync()
+	 */
+	len = (u64)end - (u64)start + 1;
 	trace_btrfs_sync_file(file, datasync);
 
 	/*
@@ -2071,8 +2076,7 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 			}
 		}
 		if (!full_sync) {
-			ret = btrfs_wait_ordered_range(inode, start,
-						       end - start + 1);
+			ret = btrfs_wait_ordered_range(inode, start, len);
 			if (ret) {
 				btrfs_end_transaction(trans, root);
 				goto out;

fs/btrfs/inode.c

@@ -4046,9 +4046,7 @@ int btrfs_unlink_inode(struct btrfs_trans_handle *trans,
  */
 static struct btrfs_trans_handle *__unlink_start_trans(struct inode *dir)
 {
-	struct btrfs_trans_handle *trans;
 	struct btrfs_root *root = BTRFS_I(dir)->root;
-	int ret;
 
 	/*
 	 * 1 for the possible orphan item
@@ -4057,27 +4055,7 @@ static struct btrfs_trans_handle *__unlink_start_trans(struct inode *dir)
 	 * 1 for the inode ref
 	 * 1 for the inode
 	 */
-	trans = btrfs_start_transaction(root, 5);
-	if (!IS_ERR(trans) || PTR_ERR(trans) != -ENOSPC)
-		return trans;
-
-	if (PTR_ERR(trans) == -ENOSPC) {
-		u64 num_bytes = btrfs_calc_trans_metadata_size(root, 5);
-
-		trans = btrfs_start_transaction(root, 0);
-		if (IS_ERR(trans))
-			return trans;
-		ret = btrfs_cond_migrate_bytes(root->fs_info,
-					       &root->fs_info->trans_block_rsv,
-					       num_bytes, 5);
-		if (ret) {
-			btrfs_end_transaction(trans, root);
-			return ERR_PTR(ret);
-		}
-		trans->block_rsv = &root->fs_info->trans_block_rsv;
-		trans->bytes_reserved = num_bytes;
-	}
-	return trans;
+	return btrfs_start_transaction_fallback_global_rsv(root, 5, 5);
 }
 
 static int btrfs_unlink(struct inode *dir, struct dentry *dentry)

fs/btrfs/qgroup.c

@@ -993,9 +993,10 @@ int btrfs_quota_disable(struct btrfs_trans_handle *trans,
 	mutex_lock(&fs_info->qgroup_ioctl_lock);
 	if (!fs_info->quota_root)
 		goto out;
-	spin_lock(&fs_info->qgroup_lock);
 	fs_info->quota_enabled = 0;
 	fs_info->pending_quota_state = 0;
+	btrfs_qgroup_wait_for_completion(fs_info);
+	spin_lock(&fs_info->qgroup_lock);
 	quota_root = fs_info->quota_root;
 	fs_info->quota_root = NULL;
 	fs_info->qgroup_flags &= ~BTRFS_QGROUP_STATUS_FLAG_ON;
@@ -1461,6 +1462,8 @@ struct btrfs_qgroup_extent_record
 	struct btrfs_qgroup_extent_record *entry;
 	u64 bytenr = record->bytenr;
 
+	assert_spin_locked(&delayed_refs->lock);
+
 	while (*p) {
 		parent_node = *p;
 		entry = rb_entry(parent_node, struct btrfs_qgroup_extent_record,

fs/btrfs/scrub.c

@@ -3432,7 +3432,9 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 static noinline_for_stack int scrub_chunk(struct scrub_ctx *sctx,
 					  struct btrfs_device *scrub_dev,
 					  u64 chunk_offset, u64 length,
-					  u64 dev_offset, int is_dev_replace)
+					  u64 dev_offset,
+					  struct btrfs_block_group_cache *cache,
+					  int is_dev_replace)
 {
 	struct btrfs_mapping_tree *map_tree =
 		&sctx->dev_root->fs_info->mapping_tree;
@@ -3445,8 +3447,18 @@ static noinline_for_stack int scrub_chunk(struct scrub_ctx *sctx,
 	em = lookup_extent_mapping(&map_tree->map_tree, chunk_offset, 1);
 	read_unlock(&map_tree->map_tree.lock);
 
-	if (!em)
-		return -EINVAL;
+	if (!em) {
+		/*
+		 * Might have been an unused block group deleted by the cleaner
+		 * kthread or relocation.
+		 */
+		spin_lock(&cache->lock);
+		if (!cache->removed)
+			ret = -EINVAL;
+		spin_unlock(&cache->lock);
+
+		return ret;
+	}
 
 	map = (struct map_lookup *)em->bdev;
 	if (em->start != chunk_offset)
@@ -3483,6 +3495,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 	u64 length;
 	u64 chunk_offset;
 	int ret = 0;
+	int ro_set;
 	int slot;
 	struct extent_buffer *l;
 	struct btrfs_key key;
@@ -3568,7 +3581,21 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		scrub_pause_on(fs_info);
 		ret = btrfs_inc_block_group_ro(root, cache);
 		scrub_pause_off(fs_info);
-		if (ret) {
+
+		if (ret == 0) {
+			ro_set = 1;
+		} else if (ret == -ENOSPC) {
+			/*
+			 * btrfs_inc_block_group_ro return -ENOSPC when it
+			 * failed in creating new chunk for metadata.
+			 * It is not a problem for scrub/replace, because
+			 * metadata are always cowed, and our scrub paused
+			 * commit_transactions.
+			 */
+			ro_set = 0;
+		} else {
+			btrfs_warn(fs_info, "failed setting block group ro, ret=%d\n",
+				   ret);
 			btrfs_put_block_group(cache);
 			break;
 		}
@@ -3577,7 +3604,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		dev_replace->cursor_left = found_key.offset;
 		dev_replace->item_needs_writeback = 1;
 		ret = scrub_chunk(sctx, scrub_dev, chunk_offset, length,
-				  found_key.offset, is_dev_replace);
+				  found_key.offset, cache, is_dev_replace);
 
 		/*
 		 * flush, submit all pending read and write bios, afterwards
@@ -3611,7 +3638,30 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 
 		scrub_pause_off(fs_info);
 
-		btrfs_dec_block_group_ro(root, cache);
+		if (ro_set)
+			btrfs_dec_block_group_ro(root, cache);
+
+		/*
+		 * We might have prevented the cleaner kthread from deleting
+		 * this block group if it was already unused because we raced
+		 * and set it to RO mode first. So add it back to the unused
+		 * list, otherwise it might not ever be deleted unless a manual
+		 * balance is triggered or it becomes used and unused again.
+		 */
+		spin_lock(&cache->lock);
+		if (!cache->removed && !cache->ro && cache->reserved == 0 &&
+		    btrfs_block_group_used(&cache->item) == 0) {
+			spin_unlock(&cache->lock);
+			spin_lock(&fs_info->unused_bgs_lock);
+			if (list_empty(&cache->bg_list)) {
+				btrfs_get_block_group(cache);
+				list_add_tail(&cache->bg_list,
+					      &fs_info->unused_bgs);
+			}
+			spin_unlock(&fs_info->unused_bgs_lock);
+		} else {
+			spin_unlock(&cache->lock);
+		}
 
 		btrfs_put_block_group(cache);
 		if (ret)

fs/btrfs/tests/free-space-tests.c

@@ -898,8 +898,10 @@ int btrfs_test_free_space_cache(void)
 	}
 
 	root = btrfs_alloc_dummy_root();
-	if (!root)
+	if (IS_ERR(root)) {
+		ret = PTR_ERR(root);
 		goto out;
+	}
 
 	root->fs_info = btrfs_alloc_dummy_fs_info();
 	if (!root->fs_info)

fs/btrfs/transaction.c

@@ -592,6 +592,38 @@ struct btrfs_trans_handle *btrfs_start_transaction(struct btrfs_root *root,
 	return start_transaction(root, num_items, TRANS_START,
 				 BTRFS_RESERVE_FLUSH_ALL);
 }
+struct btrfs_trans_handle *btrfs_start_transaction_fallback_global_rsv(
+					struct btrfs_root *root,
+					unsigned int num_items,
+					int min_factor)
+{
+	struct btrfs_trans_handle *trans;
+	u64 num_bytes;
+	int ret;
+
+	trans = btrfs_start_transaction(root, num_items);
+	if (!IS_ERR(trans) || PTR_ERR(trans) != -ENOSPC)
+		return trans;
+
+	trans = btrfs_start_transaction(root, 0);
+	if (IS_ERR(trans))
+		return trans;
+
+	num_bytes = btrfs_calc_trans_metadata_size(root, num_items);
+	ret = btrfs_cond_migrate_bytes(root->fs_info,
+				       &root->fs_info->trans_block_rsv,
+				       num_bytes,
+				       min_factor);
+	if (ret) {
+		btrfs_end_transaction(trans, root);
+		return ERR_PTR(ret);
+	}
+
+	trans->block_rsv = &root->fs_info->trans_block_rsv;
+	trans->bytes_reserved = num_bytes;
+
+	return trans;
+}
 
 struct btrfs_trans_handle *btrfs_start_transaction_lflush(
 					struct btrfs_root *root,

fs/btrfs/transaction.h

@@ -185,6 +185,10 @@ int btrfs_end_transaction(struct btrfs_trans_handle *trans,
 			  struct btrfs_root *root);
 struct btrfs_trans_handle *btrfs_start_transaction(struct btrfs_root *root,
 						   unsigned int num_items);
+struct btrfs_trans_handle *btrfs_start_transaction_fallback_global_rsv(
+					struct btrfs_root *root,
+					unsigned int num_items,
+					int min_factor);
 struct btrfs_trans_handle *btrfs_start_transaction_lflush(
 					struct btrfs_root *root,
 					unsigned int num_items);

fs/btrfs/volumes.c

@@ -1973,8 +1973,7 @@ void btrfs_rm_dev_replace_remove_srcdev(struct btrfs_fs_info *fs_info,
 	if (srcdev->writeable) {
 		fs_devices->rw_devices--;
 		/* zero out the old super if it is writable */
-		btrfs_scratch_superblocks(srcdev->bdev,
-					rcu_str_deref(srcdev->name));
+		btrfs_scratch_superblocks(srcdev->bdev, srcdev->name->str);
 	}
 
 	if (srcdev->bdev)
@@ -2024,8 +2023,7 @@ void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 	btrfs_sysfs_rm_device_link(fs_info->fs_devices, tgtdev);
 
 	if (tgtdev->bdev) {
-		btrfs_scratch_superblocks(tgtdev->bdev,
-					rcu_str_deref(tgtdev->name));
+		btrfs_scratch_superblocks(tgtdev->bdev, tgtdev->name->str);
 		fs_info->fs_devices->open_devices--;
 	}
 	fs_info->fs_devices->num_devices--;
@@ -2853,7 +2851,8 @@ static int btrfs_relocate_chunk(struct btrfs_root *root, u64 chunk_offset)
 	if (ret)
 		return ret;
 
-	trans = btrfs_start_transaction(root, 0);
+	trans = btrfs_start_trans_remove_block_group(root->fs_info,
+						     chunk_offset);
 	if (IS_ERR(trans)) {
 		ret = PTR_ERR(trans);
 		btrfs_std_error(root->fs_info, ret, NULL);
@@ -3123,7 +3122,7 @@ static int chunk_profiles_filter(u64 chunk_type,
 	return 1;
 }
 
-static int chunk_usage_filter(struct btrfs_fs_info *fs_info, u64 chunk_offset,
+static int chunk_usage_range_filter(struct btrfs_fs_info *fs_info, u64 chunk_offset,
 			      struct btrfs_balance_args *bargs)
 {
 	struct btrfs_block_group_cache *cache;
@@ -3156,7 +3155,7 @@ static int chunk_usage_filter(struct btrfs_fs_info *fs_info, u64 chunk_offset,
 	return ret;
 }
 
-static int chunk_usage_range_filter(struct btrfs_fs_info *fs_info,
+static int chunk_usage_filter(struct btrfs_fs_info *fs_info,
 		u64 chunk_offset, struct btrfs_balance_args *bargs)
 {
 	struct btrfs_block_group_cache *cache;

fs/btrfs/volumes.h

@@ -382,7 +382,7 @@ struct map_lookup {
 #define BTRFS_BALANCE_ARGS_LIMIT	(1ULL << 5)
 #define BTRFS_BALANCE_ARGS_LIMIT_RANGE	(1ULL << 6)
 #define BTRFS_BALANCE_ARGS_STRIPES_RANGE (1ULL << 7)
-#define BTRFS_BALANCE_ARGS_USAGE_RANGE	(1ULL << 8)
+#define BTRFS_BALANCE_ARGS_USAGE_RANGE	(1ULL << 10)
 
 #define BTRFS_BALANCE_ARGS_MASK			\
 	(BTRFS_BALANCE_ARGS_PROFILES |		\