1 files changed, 39 insertions, 56 deletions

diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index 87bd37b70738..808370ada888 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -894,8 +894,6 @@ static int btrfs_clean_quota_tree(struct btrfs_trans_handle *trans,
 	if (!path)
 		return -ENOMEM;
 
-	path->leave_spinning = 1;
-
 	key.objectid = 0;
 	key.offset = 0;
 	key.type = 0;
@@ -1944,34 +1942,22 @@ static int qgroup_trace_extent_swap(struct btrfs_trans_handle* trans,
 		struct btrfs_key dst_key;
 
 		if (src_path->nodes[cur_level] == NULL) {
-			struct btrfs_key first_key;
 			struct extent_buffer *eb;
 			int parent_slot;
-			u64 child_gen;
-			u64 child_bytenr;
 
 			eb = src_path->nodes[cur_level + 1];
 			parent_slot = src_path->slots[cur_level + 1];
-			child_bytenr = btrfs_node_blockptr(eb, parent_slot);
-			child_gen = btrfs_node_ptr_generation(eb, parent_slot);
-			btrfs_node_key_to_cpu(eb, &first_key, parent_slot);
 
-			eb = read_tree_block(fs_info, child_bytenr, child_gen,
-					     cur_level, &first_key);
+			eb = btrfs_read_node_slot(eb, parent_slot);
 			if (IS_ERR(eb)) {
 				ret = PTR_ERR(eb);
 				goto out;
-			} else if (!extent_buffer_uptodate(eb)) {
-				free_extent_buffer(eb);
-				ret = -EIO;
-				goto out;
 			}
 
 			src_path->nodes[cur_level] = eb;
 
 			btrfs_tree_read_lock(eb);
-			btrfs_set_lock_blocking_read(eb);
-			src_path->locks[cur_level] = BTRFS_READ_LOCK_BLOCKING;
+			src_path->locks[cur_level] = BTRFS_READ_LOCK;
 		}
 
 		src_path->slots[cur_level] = dst_path->slots[cur_level];
@@ -2066,10 +2052,8 @@ static int qgroup_trace_new_subtree_blocks(struct btrfs_trans_handle* trans,
 
 	/* Read the tree block if needed */
 	if (dst_path->nodes[cur_level] == NULL) {
-		struct btrfs_key first_key;
 		int parent_slot;
 		u64 child_gen;
-		u64 child_bytenr;
 
 		/*
 		 * dst_path->nodes[root_level] must be initialized before
@@ -2088,31 +2072,23 @@ static int qgroup_trace_new_subtree_blocks(struct btrfs_trans_handle* trans,
 		  */
 		eb = dst_path->nodes[cur_level + 1];
 		parent_slot = dst_path->slots[cur_level + 1];
-		child_bytenr = btrfs_node_blockptr(eb, parent_slot);
 		child_gen = btrfs_node_ptr_generation(eb, parent_slot);
-		btrfs_node_key_to_cpu(eb, &first_key, parent_slot);
 
 		/* This node is old, no need to trace */
 		if (child_gen < last_snapshot)
 			goto out;
 
-		eb = read_tree_block(fs_info, child_bytenr, child_gen,
-				     cur_level, &first_key);
+		eb = btrfs_read_node_slot(eb, parent_slot);
 		if (IS_ERR(eb)) {
 			ret = PTR_ERR(eb);
 			goto out;
-		} else if (!extent_buffer_uptodate(eb)) {
-			free_extent_buffer(eb);
-			ret = -EIO;
-			goto out;
 		}
 
 		dst_path->nodes[cur_level] = eb;
 		dst_path->slots[cur_level] = 0;
 
 		btrfs_tree_read_lock(eb);
-		btrfs_set_lock_blocking_read(eb);
-		dst_path->locks[cur_level] = BTRFS_READ_LOCK_BLOCKING;
+		dst_path->locks[cur_level] = BTRFS_READ_LOCK;
 		need_cleanup = true;
 	}
 
@@ -2256,38 +2232,28 @@ walk_down:
 	level = root_level;
 	while (level >= 0) {
 		if (path->nodes[level] == NULL) {
-			struct btrfs_key first_key;
 			int parent_slot;
-			u64 child_gen;
 			u64 child_bytenr;
 
 			/*
-			 * We need to get child blockptr/gen from parent before
-			 * we can read it.
+			 * We need to get child blockptr from parent before we
+			 * can read it.
 			  */
 			eb = path->nodes[level + 1];
 			parent_slot = path->slots[level + 1];
 			child_bytenr = btrfs_node_blockptr(eb, parent_slot);
-			child_gen = btrfs_node_ptr_generation(eb, parent_slot);
-			btrfs_node_key_to_cpu(eb, &first_key, parent_slot);
 
-			eb = read_tree_block(fs_info, child_bytenr, child_gen,
-					     level, &first_key);
+			eb = btrfs_read_node_slot(eb, parent_slot);
 			if (IS_ERR(eb)) {
 				ret = PTR_ERR(eb);
 				goto out;
-			} else if (!extent_buffer_uptodate(eb)) {
-				free_extent_buffer(eb);
-				ret = -EIO;
-				goto out;
 			}
 
 			path->nodes[level] = eb;
 			path->slots[level] = 0;
 
 			btrfs_tree_read_lock(eb);
-			btrfs_set_lock_blocking_read(eb);
-			path->locks[level] = BTRFS_READ_LOCK_BLOCKING;
+			path->locks[level] = BTRFS_READ_LOCK;
 
 			ret = btrfs_qgroup_trace_extent(trans, child_bytenr,
 							fs_info->nodesize,
@@ -3224,6 +3190,12 @@ out:
 	return ret;
 }
 
+static bool rescan_should_stop(struct btrfs_fs_info *fs_info)
+{
+	return btrfs_fs_closing(fs_info) ||
+		test_bit(BTRFS_FS_STATE_REMOUNTING, &fs_info->fs_state);
+}
+
 static void btrfs_qgroup_rescan_worker(struct btrfs_work *work)
 {
 	struct btrfs_fs_info *fs_info = container_of(work, struct btrfs_fs_info,
@@ -3232,6 +3204,7 @@ static void btrfs_qgroup_rescan_worker(struct btrfs_work *work)
 	struct btrfs_trans_handle *trans = NULL;
 	int err = -ENOMEM;
 	int ret = 0;
+	bool stopped = false;
 
 	path = btrfs_alloc_path();
 	if (!path)
@@ -3244,7 +3217,7 @@ static void btrfs_qgroup_rescan_worker(struct btrfs_work *work)
 	path->skip_locking = 1;
 
 	err = 0;
-	while (!err && !btrfs_fs_closing(fs_info)) {
+	while (!err && !(stopped = rescan_should_stop(fs_info))) {
 		trans = btrfs_start_transaction(fs_info->fs_root, 0);
 		if (IS_ERR(trans)) {
 			err = PTR_ERR(trans);
@@ -3287,7 +3260,7 @@ out:
 	}
 
 	mutex_lock(&fs_info->qgroup_rescan_lock);
-	if (!btrfs_fs_closing(fs_info))
+	if (!stopped)
 		fs_info->qgroup_flags &= ~BTRFS_QGROUP_STATUS_FLAG_RESCAN;
 	if (trans) {
 		ret = update_qgroup_status_item(trans);
@@ -3306,7 +3279,7 @@ out:
 
 	btrfs_end_transaction(trans);
 
-	if (btrfs_fs_closing(fs_info)) {
+	if (stopped) {
 		btrfs_info(fs_info, "qgroup scan paused");
 	} else if (err >= 0) {
 		btrfs_info(fs_info, "qgroup scan completed%s",
@@ -3565,16 +3538,6 @@ static int try_flush_qgroup(struct btrfs_root *root)
 	bool can_commit = true;
 
 	/*
-	 * We don't want to run flush again and again, so if there is a running
-	 * one, we won't try to start a new flush, but exit directly.
-	 */
-	if (test_and_set_bit(BTRFS_ROOT_QGROUP_FLUSHING, &root->state)) {
-		wait_event(root->qgroup_flush_wait,
-			!test_bit(BTRFS_ROOT_QGROUP_FLUSHING, &root->state));
-		return 0;
-	}
-
-	/*
 	 * If current process holds a transaction, we shouldn't flush, as we
 	 * assume all space reservation happens before a transaction handle is
 	 * held.
@@ -3588,6 +3551,26 @@ static int try_flush_qgroup(struct btrfs_root *root)
 	    current->journal_info != BTRFS_SEND_TRANS_STUB)
 		can_commit = false;
 
+	/*
+	 * We don't want to run flush again and again, so if there is a running
+	 * one, we won't try to start a new flush, but exit directly.
+	 */
+	if (test_and_set_bit(BTRFS_ROOT_QGROUP_FLUSHING, &root->state)) {
+		/*
+		 * We are already holding a transaction, thus we can block other
+		 * threads from flushing.  So exit right now. This increases
+		 * the chance of EDQUOT for heavy load and near limit cases.
+		 * But we can argue that if we're already near limit, EDQUOT is
+		 * unavoidable anyway.
+		 */
+		if (!can_commit)
+			return 0;
+
+		wait_event(root->qgroup_flush_wait,
+			!test_bit(BTRFS_ROOT_QGROUP_FLUSHING, &root->state));
+		return 0;
+	}
+
 	ret = btrfs_start_delalloc_snapshot(root);
 	if (ret < 0)
 		goto out;
@@ -4242,7 +4225,7 @@ int btrfs_qgroup_trace_subtree_after_cow(struct btrfs_trans_handle *trans,
 	spin_unlock(&blocks->lock);
 
 	/* Read out reloc subtree root */
-	reloc_eb = read_tree_block(fs_info, block->reloc_bytenr,
+	reloc_eb = read_tree_block(fs_info, block->reloc_bytenr, 0,
 				   block->reloc_generation, block->level,
 				   &block->first_key);
 	if (IS_ERR(reloc_eb)) {