From e339a6b097c515a31ce230d498c44ff2e89f1cf4 Mon Sep 17 00:00:00 2001
From: Josef Bacik <jbacik@fb.com>
Date: Wed, 2 Jul 2014 10:54:25 -0700
Subject: Btrfs: __btrfs_mod_ref should always use no_quota

Before I extended the no_quota arg to btrfs_dec/inc_ref because I didn't
understand how snapshot delete was using it and assumed that we needed the
quota operations there.  With Mark's work this has turned out to be not the
case, we _always_ need to use no_quota for btrfs_dec/inc_ref, so just drop the
argument and make __btrfs_mod_ref call it's process function with no_quota set
always.  Thanks,

Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
---
 fs/btrfs/extent-tree.c | 24 +++++++++++-------------
 1 file changed, 11 insertions(+), 13 deletions(-)

(limited to 'fs/btrfs/extent-tree.c')

diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 813537f362f9..591893fe58cb 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -3057,7 +3057,7 @@ out:
 static int __btrfs_mod_ref(struct btrfs_trans_handle *trans,
 			   struct btrfs_root *root,
 			   struct extent_buffer *buf,
-			   int full_backref, int inc, int no_quota)
+			   int full_backref, int inc)
 {
 	u64 bytenr;
 	u64 num_bytes;
@@ -3111,7 +3111,7 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans,
 			key.offset -= btrfs_file_extent_offset(buf, fi);
 			ret = process_func(trans, root, bytenr, num_bytes,
 					   parent, ref_root, key.objectid,
-					   key.offset, no_quota);
+					   key.offset, 1);
 			if (ret)
 				goto fail;
 		} else {
@@ -3119,7 +3119,7 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans,
 			num_bytes = btrfs_level_size(root, level - 1);
 			ret = process_func(trans, root, bytenr, num_bytes,
 					   parent, ref_root, level - 1, 0,
-					   no_quota);
+					   1);
 			if (ret)
 				goto fail;
 		}
@@ -3130,15 +3130,15 @@ fail:
 }
 
 int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
-		  struct extent_buffer *buf, int full_backref, int no_quota)
+		  struct extent_buffer *buf, int full_backref)
 {
-	return __btrfs_mod_ref(trans, root, buf, full_backref, 1, no_quota);
+	return __btrfs_mod_ref(trans, root, buf, full_backref, 1);
 }
 
 int btrfs_dec_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
-		  struct extent_buffer *buf, int full_backref, int no_quota)
+		  struct extent_buffer *buf, int full_backref)
 {
-	return __btrfs_mod_ref(trans, root, buf, full_backref, 0, no_quota);
+	return __btrfs_mod_ref(trans, root, buf, full_backref, 0);
 }
 
 static int write_one_cache_group(struct btrfs_trans_handle *trans,
@@ -7532,9 +7532,9 @@ static noinline int walk_down_proc(struct btrfs_trans_handle *trans,
 	/* wc->stage == UPDATE_BACKREF */
 	if (!(wc->flags[level] & flag)) {
 		BUG_ON(!path->locks[level]);
-		ret = btrfs_inc_ref(trans, root, eb, 1, wc->for_reloc);
+		ret = btrfs_inc_ref(trans, root, eb, 1);
 		BUG_ON(ret); /* -ENOMEM */
-		ret = btrfs_dec_ref(trans, root, eb, 0, wc->for_reloc);
+		ret = btrfs_dec_ref(trans, root, eb, 0);
 		BUG_ON(ret); /* -ENOMEM */
 		ret = btrfs_set_disk_extent_flags(trans, root, eb->start,
 						  eb->len, flag,
@@ -7769,11 +7769,9 @@ static noinline int walk_up_proc(struct btrfs_trans_handle *trans,
 	if (wc->refs[level] == 1) {
 		if (level == 0) {
 			if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF)
-				ret = btrfs_dec_ref(trans, root, eb, 1,
-						    wc->for_reloc);
+				ret = btrfs_dec_ref(trans, root, eb, 1);
 			else
-				ret = btrfs_dec_ref(trans, root, eb, 0,
-						    wc->for_reloc);
+				ret = btrfs_dec_ref(trans, root, eb, 0);
 			BUG_ON(ret); /* -ENOMEM */
 		}
 		/* make block locked assertion in clean_tree_block happy */
-- 
cgit v1.2.3


From 1152651a081720ef6a8c76bb7da676e8c900ac30 Mon Sep 17 00:00:00 2001
From: Mark Fasheh <mfasheh@suse.de>
Date: Thu, 17 Jul 2014 12:39:01 -0700
Subject: btrfs: qgroup: account shared subtrees during snapshot delete

During its tree walk, btrfs_drop_snapshot() will skip any shared
subtrees it encounters. This is incorrect when we have qgroups
turned on as those subtrees need to have their contents
accounted. In particular, the case we're concerned with is when
removing our snapshot root leaves the subtree with only one root
reference.

In those cases we need to find the last remaining root and add
each extent in the subtree to the corresponding qgroup exclusive
counts.

This patch implements the shared subtree walk and a new qgroup
operation, BTRFS_QGROUP_OPER_SUB_SUBTREE. When an operation of
this type is encountered during qgroup accounting, we search for
any root references to that extent and in the case that we find
only one reference left, we go ahead and do the math on it's
exclusive counts.

Signed-off-by: Mark Fasheh <mfasheh@suse.de>
Reviewed-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
---
 fs/btrfs/extent-tree.c | 261 +++++++++++++++++++++++++++++++++++++++++++++++++
 fs/btrfs/qgroup.c      | 164 +++++++++++++++++++++++++++++++
 fs/btrfs/qgroup.h      |   1 +
 3 files changed, 426 insertions(+)

(limited to 'fs/btrfs/extent-tree.c')

diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 591893fe58cb..102ed3143976 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -7478,6 +7478,220 @@ reada:
 	wc->reada_slot = slot;
 }
 
+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, ret;
+	struct btrfs_key key;
+	struct btrfs_file_extent_item *fi;
+	u64 bytenr, num_bytes;
+
+	for (i = 0; i < nr; i++) {
+		btrfs_item_key_to_cpu(eb, &key, i);
+
+		if (key.type != BTRFS_EXTENT_DATA_KEY)
+			continue;
+
+		fi = btrfs_item_ptr(eb, i, struct btrfs_file_extent_item);
+		/* filter out non qgroup-accountable extents  */
+		extent_type = btrfs_file_extent_type(eb, fi);
+
+		if (extent_type == BTRFS_FILE_EXTENT_INLINE)
+			continue;
+
+		bytenr = btrfs_file_extent_disk_bytenr(eb, fi);
+		if (!bytenr)
+			continue;
+
+		num_bytes = btrfs_file_extent_disk_num_bytes(eb, fi);
+
+		ret = btrfs_qgroup_record_ref(trans, root->fs_info,
+					      root->objectid,
+					      bytenr, num_bytes,
+					      BTRFS_QGROUP_OPER_SUB_SUBTREE, 0);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
+/*
+ * Walk up the tree from the bottom, freeing leaves and any interior
+ * nodes which have had all slots visited. If a node (leaf or
+ * interior) is freed, the node above it will have it's slot
+ * incremented. The root node will never be freed.
+ *
+ * At the end of this function, we should have a path which has all
+ * slots incremented to the next position for a search. If we need to
+ * read a new node it will be NULL and the node above it will have the
+ * correct slot selected for a later read.
+ *
+ * If we increment the root nodes slot counter past the number of
+ * elements, 1 is returned to signal completion of the search.
+ */
+static int adjust_slots_upwards(struct btrfs_root *root,
+				struct btrfs_path *path, int root_level)
+{
+	int level = 0;
+	int nr, slot;
+	struct extent_buffer *eb;
+
+	if (root_level == 0)
+		return 1;
+
+	while (level <= root_level) {
+		eb = path->nodes[level];
+		nr = btrfs_header_nritems(eb);
+		path->slots[level]++;
+		slot = path->slots[level];
+		if (slot >= nr || level == 0) {
+			/*
+			 * Don't free the root -  we will detect this
+			 * condition after our loop and return a
+			 * positive value for caller to stop walking the tree.
+			 */
+			if (level != root_level) {
+				btrfs_tree_unlock_rw(eb, path->locks[level]);
+				path->locks[level] = 0;
+
+				free_extent_buffer(eb);
+				path->nodes[level] = NULL;
+				path->slots[level] = 0;
+			}
+		} else {
+			/*
+			 * We have a valid slot to walk back down
+			 * from. Stop here so caller can process these
+			 * new nodes.
+			 */
+			break;
+		}
+
+		level++;
+	}
+
+	eb = path->nodes[root_level];
+	if (path->slots[root_level] >= btrfs_header_nritems(eb))
+		return 1;
+
+	return 0;
+}
+
+/*
+ * root_eb is the subtree root and is locked before this function is called.
+ */
+static int account_shared_subtree(struct btrfs_trans_handle *trans,
+				  struct btrfs_root *root,
+				  struct extent_buffer *root_eb,
+				  u64 root_gen,
+				  int root_level)
+{
+	int ret = 0;
+	int level;
+	struct extent_buffer *eb = root_eb;
+	struct btrfs_path *path = NULL;
+
+	BUG_ON(root_level < 0 || root_level > BTRFS_MAX_LEVEL);
+	BUG_ON(root_eb == NULL);
+
+	if (!root->fs_info->quota_enabled)
+		return 0;
+
+	if (!extent_buffer_uptodate(root_eb)) {
+		ret = btrfs_read_buffer(root_eb, root_gen);
+		if (ret)
+			goto out;
+	}
+
+	if (root_level == 0) {
+		ret = account_leaf_items(trans, root, root_eb);
+		goto out;
+	}
+
+	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
+
+	/*
+	 * Walk down the tree.  Missing extent blocks are filled in as
+	 * we go. Metadata is accounted every time we read a new
+	 * extent block.
+	 *
+	 * When we reach a leaf, we account for file extent items in it,
+	 * walk back up the tree (adjusting slot pointers as we go)
+	 * and restart the search process.
+	 */
+	extent_buffer_get(root_eb); /* For path */
+	path->nodes[root_level] = root_eb;
+	path->slots[root_level] = 0;
+	path->locks[root_level] = 0; /* so release_path doesn't try to unlock */
+walk_down:
+	level = root_level;
+	while (level >= 0) {
+		if (path->nodes[level] == NULL) {
+			int child_bsize = root->nodesize;
+			int parent_slot;
+			u64 child_gen;
+			u64 child_bytenr;
+
+			/* We need to get child blockptr/gen 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);
+
+			eb = read_tree_block(root, child_bytenr, child_bsize,
+					     child_gen);
+			if (!eb || !extent_buffer_uptodate(eb)) {
+				ret = -EIO;
+				goto out;
+			}
+
+			path->nodes[level] = eb;
+			path->slots[level] = 0;
+
+			btrfs_tree_read_lock(eb);
+			btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
+			path->locks[level] = BTRFS_READ_LOCK_BLOCKING;
+
+			ret = btrfs_qgroup_record_ref(trans, root->fs_info,
+						root->objectid,
+						child_bytenr,
+						child_bsize,
+						BTRFS_QGROUP_OPER_SUB_SUBTREE,
+						0);
+			if (ret)
+				goto out;
+
+		}
+
+		if (level == 0) {
+			ret = account_leaf_items(trans, root, path->nodes[level]);
+			if (ret)
+				goto out;
+
+			/* Nonzero return here means we completed our search */
+			ret = adjust_slots_upwards(root, path, root_level);
+			if (ret)
+				break;
+
+			/* Restart search with new slots */
+			goto walk_down;
+		}
+
+		level--;
+	}
+
+	ret = 0;
+out:
+	btrfs_free_path(path);
+
+	return ret;
+}
+
 /*
  * helper to process tree block while walking down the tree.
  *
@@ -7581,6 +7795,7 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
 	int level = wc->level;
 	int reada = 0;
 	int ret = 0;
+	bool need_account = false;
 
 	generation = btrfs_node_ptr_generation(path->nodes[level],
 					       path->slots[level]);
@@ -7626,6 +7841,7 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
 
 	if (wc->stage == DROP_REFERENCE) {
 		if (wc->refs[level - 1] > 1) {
+			need_account = true;
 			if (level == 1 &&
 			    (wc->flags[0] & BTRFS_BLOCK_FLAG_FULL_BACKREF))
 				goto skip;
@@ -7689,6 +7905,16 @@ skip:
 			parent = 0;
 		}
 
+		if (need_account) {
+			ret = account_shared_subtree(trans, root, next,
+						     generation, level - 1);
+			if (ret) {
+				printk_ratelimited(KERN_ERR "BTRFS: %s Error "
+					"%d accounting shared subtree. Quota "
+					"is out of sync, rescan required.\n",
+					root->fs_info->sb->s_id, ret);
+			}
+		}
 		ret = btrfs_free_extent(trans, root, bytenr, blocksize, parent,
 				root->root_key.objectid, level - 1, 0, 0);
 		BUG_ON(ret); /* -ENOMEM */
@@ -7773,6 +7999,13 @@ static noinline int walk_up_proc(struct btrfs_trans_handle *trans,
 			else
 				ret = btrfs_dec_ref(trans, root, eb, 0);
 			BUG_ON(ret); /* -ENOMEM */
+			ret = account_leaf_items(trans, root, eb);
+			if (ret) {
+				printk_ratelimited(KERN_ERR "BTRFS: %s Error "
+					"%d accounting leaf items. Quota "
+					"is out of sync, rescan required.\n",
+					root->fs_info->sb->s_id, ret);
+			}
 		}
 		/* make block locked assertion in clean_tree_block happy */
 		if (!path->locks[level] &&
@@ -7898,6 +8131,8 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 	int level;
 	bool root_dropped = false;
 
+	btrfs_debug(root->fs_info, "Drop subvolume %llu", root->objectid);
+
 	path = btrfs_alloc_path();
 	if (!path) {
 		err = -ENOMEM;
@@ -8023,6 +8258,24 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 				goto out_end_trans;
 			}
 
+			/*
+			 * Qgroup update accounting is run from
+			 * delayed ref handling. This usually works
+			 * out because delayed refs are normally the
+			 * only way qgroup updates are added. However,
+			 * we may have added updates during our tree
+			 * walk so run qgroups here to make sure we
+			 * don't lose any updates.
+			 */
+			ret = btrfs_delayed_qgroup_accounting(trans,
+							      root->fs_info);
+			if (ret)
+				printk_ratelimited(KERN_ERR "BTRFS: Failure %d "
+						   "running qgroup updates "
+						   "during snapshot delete. "
+						   "Quota is out of sync, "
+						   "rescan required.\n", ret);
+
 			btrfs_end_transaction_throttle(trans, tree_root);
 			if (!for_reloc && btrfs_need_cleaner_sleep(root)) {
 				pr_debug("BTRFS: drop snapshot early exit\n");
@@ -8076,6 +8329,14 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 	}
 	root_dropped = true;
 out_end_trans:
+	ret = btrfs_delayed_qgroup_accounting(trans, tree_root->fs_info);
+	if (ret)
+		printk_ratelimited(KERN_ERR "BTRFS: Failure %d "
+				   "running qgroup updates "
+				   "during snapshot delete. "
+				   "Quota is out of sync, "
+				   "rescan required.\n", ret);
+
 	btrfs_end_transaction_throttle(trans, tree_root);
 out_free:
 	kfree(wc);
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index 98cb6b2630f9..65b62c467e28 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -1201,6 +1201,50 @@ out:
 	mutex_unlock(&fs_info->qgroup_ioctl_lock);
 	return ret;
 }
+
+static int comp_oper_exist(struct btrfs_qgroup_operation *oper1,
+			   struct btrfs_qgroup_operation *oper2)
+{
+	/*
+	 * Ignore seq and type here, we're looking for any operation
+	 * at all related to this extent on that root.
+	 */
+	if (oper1->bytenr < oper2->bytenr)
+		return -1;
+	if (oper1->bytenr > oper2->bytenr)
+		return 1;
+	if (oper1->ref_root < oper2->ref_root)
+		return -1;
+	if (oper1->ref_root > oper2->ref_root)
+		return 1;
+	return 0;
+}
+
+static int qgroup_oper_exists(struct btrfs_fs_info *fs_info,
+			      struct btrfs_qgroup_operation *oper)
+{
+	struct rb_node *n;
+	struct btrfs_qgroup_operation *cur;
+	int cmp;
+
+	spin_lock(&fs_info->qgroup_op_lock);
+	n = fs_info->qgroup_op_tree.rb_node;
+	while (n) {
+		cur = rb_entry(n, struct btrfs_qgroup_operation, n);
+		cmp = comp_oper_exist(cur, oper);
+		if (cmp < 0) {
+			n = n->rb_right;
+		} else if (cmp) {
+			n = n->rb_left;
+		} else {
+			spin_unlock(&fs_info->qgroup_op_lock);
+			return -EEXIST;
+		}
+	}
+	spin_unlock(&fs_info->qgroup_op_lock);
+	return 0;
+}
+
 static int comp_oper(struct btrfs_qgroup_operation *oper1,
 		     struct btrfs_qgroup_operation *oper2)
 {
@@ -1290,6 +1334,23 @@ int btrfs_qgroup_record_ref(struct btrfs_trans_handle *trans,
 	oper->seq = atomic_inc_return(&fs_info->qgroup_op_seq);
 	INIT_LIST_HEAD(&oper->elem.list);
 	oper->elem.seq = 0;
+
+	if (type == BTRFS_QGROUP_OPER_SUB_SUBTREE) {
+		/*
+		 * If any operation for this bytenr/ref_root combo
+		 * exists, then we know it's not exclusively owned and
+		 * shouldn't be queued up.
+		 *
+		 * This also catches the case where we have a cloned
+		 * extent that gets queued up multiple times during
+		 * drop snapshot.
+		 */
+		if (qgroup_oper_exists(fs_info, oper)) {
+			kfree(oper);
+			return 0;
+		}
+	}
+
 	ret = insert_qgroup_oper(fs_info, oper);
 	if (ret) {
 		/* Shouldn't happen so have an assert for developers */
@@ -1883,6 +1944,106 @@ out:
 	return ret;
 }
 
+/*
+ * Process a reference to a shared subtree. This type of operation is
+ * queued during snapshot removal when we encounter extents which are
+ * shared between more than one root.
+ */
+static int qgroup_subtree_accounting(struct btrfs_trans_handle *trans,
+				     struct btrfs_fs_info *fs_info,
+				     struct btrfs_qgroup_operation *oper)
+{
+	struct ulist *roots = NULL;
+	struct ulist_node *unode;
+	struct ulist_iterator uiter;
+	struct btrfs_qgroup_list *glist;
+	struct ulist *parents;
+	int ret = 0;
+	struct btrfs_qgroup *qg;
+	u64 root_obj = 0;
+	struct seq_list elem = {};
+
+	parents = ulist_alloc(GFP_NOFS);
+	if (!parents)
+		return -ENOMEM;
+
+	btrfs_get_tree_mod_seq(fs_info, &elem);
+	ret = btrfs_find_all_roots(trans, fs_info, oper->bytenr,
+				   elem.seq, &roots);
+	btrfs_put_tree_mod_seq(fs_info, &elem);
+	if (ret < 0)
+		return ret;
+
+	if (roots->nnodes != 1)
+		goto out;
+
+	ULIST_ITER_INIT(&uiter);
+	unode = ulist_next(roots, &uiter); /* Only want 1 so no need to loop */
+	/*
+	 * If we find our ref root then that means all refs
+	 * this extent has to the root have not yet been
+	 * deleted. In that case, we do nothing and let the
+	 * last ref for this bytenr drive our update.
+	 *
+	 * This can happen for example if an extent is
+	 * referenced multiple times in a snapshot (clone,
+	 * etc). If we are in the middle of snapshot removal,
+	 * queued updates for such an extent will find the
+	 * root if we have not yet finished removing the
+	 * snapshot.
+	 */
+	if (unode->val == oper->ref_root)
+		goto out;
+
+	root_obj = unode->val;
+	BUG_ON(!root_obj);
+
+	spin_lock(&fs_info->qgroup_lock);
+	qg = find_qgroup_rb(fs_info, root_obj);
+	if (!qg)
+		goto out_unlock;
+
+	qg->excl += oper->num_bytes;
+	qg->excl_cmpr += oper->num_bytes;
+	qgroup_dirty(fs_info, qg);
+
+	/*
+	 * Adjust counts for parent groups. First we find all
+	 * parents, then in the 2nd loop we do the adjustment
+	 * while adding parents of the parents to our ulist.
+	 */
+	list_for_each_entry(glist, &qg->groups, next_group) {
+		ret = ulist_add(parents, glist->group->qgroupid,
+				ptr_to_u64(glist->group), GFP_ATOMIC);
+		if (ret < 0)
+			goto out_unlock;
+	}
+
+	ULIST_ITER_INIT(&uiter);
+	while ((unode = ulist_next(parents, &uiter))) {
+		qg = u64_to_ptr(unode->aux);
+		qg->excl += oper->num_bytes;
+		qg->excl_cmpr += oper->num_bytes;
+		qgroup_dirty(fs_info, qg);
+
+		/* Add any parents of the parents */
+		list_for_each_entry(glist, &qg->groups, next_group) {
+			ret = ulist_add(parents, glist->group->qgroupid,
+					ptr_to_u64(glist->group), GFP_ATOMIC);
+			if (ret < 0)
+				goto out_unlock;
+		}
+	}
+
+out_unlock:
+	spin_unlock(&fs_info->qgroup_lock);
+
+out:
+	ulist_free(roots);
+	ulist_free(parents);
+	return ret;
+}
+
 /*
  * btrfs_qgroup_account_ref is called for every ref that is added to or deleted
  * from the fs. First, all roots referencing the extent are searched, and
@@ -1920,6 +2081,9 @@ static int btrfs_qgroup_account(struct btrfs_trans_handle *trans,
 	case BTRFS_QGROUP_OPER_SUB_SHARED:
 		ret = qgroup_shared_accounting(trans, fs_info, oper);
 		break;
+	case BTRFS_QGROUP_OPER_SUB_SUBTREE:
+		ret = qgroup_subtree_accounting(trans, fs_info, oper);
+		break;
 	default:
 		ASSERT(0);
 	}
diff --git a/fs/btrfs/qgroup.h b/fs/btrfs/qgroup.h
index 5952ff1fbd7a..18cc68ca3090 100644
--- a/fs/btrfs/qgroup.h
+++ b/fs/btrfs/qgroup.h
@@ -44,6 +44,7 @@ enum btrfs_qgroup_operation_type {
 	BTRFS_QGROUP_OPER_ADD_SHARED,
 	BTRFS_QGROUP_OPER_SUB_EXCL,
 	BTRFS_QGROUP_OPER_SUB_SHARED,
+	BTRFS_QGROUP_OPER_SUB_SUBTREE,
 };
 
 struct btrfs_qgroup_operation {
-- 
cgit v1.2.3


From 95669976bd7d30ae265db938ecb46a6b7f8cb893 Mon Sep 17 00:00:00 2001
From: Miao Xie <miaox@cn.fujitsu.com>
Date: Thu, 24 Jul 2014 11:37:14 +0800
Subject: Btrfs: don't consider the missing device when allocating new chunks

The original code allocated new chunks by the number of the writable devices
and missing devices to make sure that any RAID levels on a degraded FS continue
to be honored, but it introduced a problem that it stopped us to allocating
new chunks, the steps to reproduce is following:

 # mkfs.btrfs -m raid1 -d raid1 -f <dev0> <dev1>
 # mkfs.btrfs -f <dev1>			//Removing <dev1> from the original fs
 # mount -o degraded <dev0> <mnt>
 # dd if=/dev/null of=<mnt>/tmpfile bs=1M

It is because we allocate new chunks only on the writable devices, if we take
the number of missing devices into account, and want to allocate new chunks
with higher RAID level, we will fail becaue we don't have enough writable
device. Fix it by ignoring the number of missing devices when allocating
new chunks.

Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
---
 fs/btrfs/extent-tree.c | 16 ++--------------
 1 file changed, 2 insertions(+), 14 deletions(-)

(limited to 'fs/btrfs/extent-tree.c')

diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 102ed3143976..5524434da059 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -3586,13 +3586,7 @@ static u64 get_restripe_target(struct btrfs_fs_info *fs_info, u64 flags)
  */
 static u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags)
 {
-	/*
-	 * we add in the count of missing devices because we want
-	 * to make sure that any RAID levels on a degraded FS
-	 * continue to be honored.
-	 */
-	u64 num_devices = root->fs_info->fs_devices->rw_devices +
-		root->fs_info->fs_devices->missing_devices;
+	u64 num_devices = root->fs_info->fs_devices->rw_devices;
 	u64 target;
 	u64 tmp;
 
@@ -8440,13 +8434,7 @@ static u64 update_block_group_flags(struct btrfs_root *root, u64 flags)
 	if (stripped)
 		return extended_to_chunk(stripped);
 
-	/*
-	 * we add in the count of missing devices because we want
-	 * to make sure that any RAID levels on a degraded FS
-	 * continue to be honored.
-	 */
-	num_devices = root->fs_info->fs_devices->rw_devices +
-		root->fs_info->fs_devices->missing_devices;
+	num_devices = root->fs_info->fs_devices->rw_devices;
 
 	stripped = BTRFS_BLOCK_GROUP_RAID0 |
 		BTRFS_BLOCK_GROUP_RAID5 | BTRFS_BLOCK_GROUP_RAID6 |
-- 
cgit v1.2.3


From 9e0af23764344f7f1b68e4eefbe7dc865018b63d Mon Sep 17 00:00:00 2001
From: Liu Bo <bo.li.liu@oracle.com>
Date: Fri, 15 Aug 2014 23:36:53 +0800
Subject: Btrfs: fix task hang under heavy compressed write

This has been reported and discussed for a long time, and this hang occurs in
both 3.15 and 3.16.

Btrfs now migrates to use kernel workqueue, but it introduces this hang problem.

Btrfs has a kind of work queued as an ordered way, which means that its
ordered_func() must be processed in the way of FIFO, so it usually looks like --

normal_work_helper(arg)
    work = container_of(arg, struct btrfs_work, normal_work);

    work->func() <---- (we name it work X)
    for ordered_work in wq->ordered_list
            ordered_work->ordered_func()
            ordered_work->ordered_free()

The hang is a rare case, first when we find free space, we get an uncached block
group, then we go to read its free space cache inode for free space information,
so it will

file a readahead request
    btrfs_readpages()
         for page that is not in page cache
                __do_readpage()
                     submit_extent_page()
                           btrfs_submit_bio_hook()
                                 btrfs_bio_wq_end_io()
                                 submit_bio()
                                 end_workqueue_bio() <--(ret by the 1st endio)
                                      queue a work(named work Y) for the 2nd
                                      also the real endio()

So the hang occurs when work Y's work_struct and work X's work_struct happens
to share the same address.

A bit more explanation,

A,B,C -- struct btrfs_work
arg   -- struct work_struct

kthread:
worker_thread()
    pick up a work_struct from @worklist
    process_one_work(arg)
	worker->current_work = arg;  <-- arg is A->normal_work
	worker->current_func(arg)
		normal_work_helper(arg)
		     A = container_of(arg, struct btrfs_work, normal_work);

		     A->func()
		     A->ordered_func()
		     A->ordered_free()  <-- A gets freed

		     B->ordered_func()
			  submit_compressed_extents()
			      find_free_extent()
				  load_free_space_inode()
				      ...   <-- (the above readhead stack)
				      end_workqueue_bio()
					   btrfs_queue_work(work C)
		     B->ordered_free()

As if work A has a high priority in wq->ordered_list and there are more ordered
works queued after it, such as B->ordered_func(), its memory could have been
freed before normal_work_helper() returns, which means that kernel workqueue
code worker_thread() still has worker->current_work pointer to be work
A->normal_work's, ie. arg's address.

Meanwhile, work C is allocated after work A is freed, work C->normal_work
and work A->normal_work are likely to share the same address(I confirmed this
with ftrace output, so I'm not just guessing, it's rare though).

When another kthread picks up work C->normal_work to process, and finds our
kthread is processing it(see find_worker_executing_work()), it'll think
work C as a collision and skip then, which ends up nobody processing work C.

So the situation is that our kthread is waiting forever on work C.

Besides, there're other cases that can lead to deadlock, but the real problem
is that all btrfs workqueue shares one work->func, -- normal_work_helper,
so this makes each workqueue to have its own helper function, but only a
wraper pf normal_work_helper.

With this patch, I no long hit the above hang.

Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Chris Mason <clm@fb.com>
---
 fs/btrfs/async-thread.c  | 44 ++++++++++++++++++++++++++++++++--------
 fs/btrfs/async-thread.h  | 28 ++++++++++++++++++++++++-
 fs/btrfs/delayed-inode.c |  4 ++--
 fs/btrfs/disk-io.c       | 53 ++++++++++++++++++++++++++----------------------
 fs/btrfs/extent-tree.c   |  7 ++++---
 fs/btrfs/inode.c         | 35 +++++++++++++++++++++-----------
 fs/btrfs/ordered-data.c  |  1 +
 fs/btrfs/qgroup.c        |  1 +
 fs/btrfs/raid56.c        |  9 +++++---
 fs/btrfs/reada.c         |  3 ++-
 fs/btrfs/scrub.c         | 14 +++++++------
 fs/btrfs/volumes.c       |  3 ++-
 12 files changed, 141 insertions(+), 61 deletions(-)

(limited to 'fs/btrfs/extent-tree.c')

diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c
index 5a201d81049c..fbd76ded9a34 100644
--- a/fs/btrfs/async-thread.c
+++ b/fs/btrfs/async-thread.c
@@ -22,7 +22,6 @@
 #include <linux/list.h>
 #include <linux/spinlock.h>
 #include <linux/freezer.h>
-#include <linux/workqueue.h>
 #include "async-thread.h"
 #include "ctree.h"
 
@@ -55,8 +54,39 @@ struct btrfs_workqueue {
 	struct __btrfs_workqueue *high;
 };
 
-static inline struct __btrfs_workqueue
-*__btrfs_alloc_workqueue(const char *name, int flags, int max_active,
+static void normal_work_helper(struct btrfs_work *work);
+
+#define BTRFS_WORK_HELPER(name)					\
+void btrfs_##name(struct work_struct *arg)				\
+{									\
+	struct btrfs_work *work = container_of(arg, struct btrfs_work,	\
+					       normal_work);		\
+	normal_work_helper(work);					\
+}
+
+BTRFS_WORK_HELPER(worker_helper);
+BTRFS_WORK_HELPER(delalloc_helper);
+BTRFS_WORK_HELPER(flush_delalloc_helper);
+BTRFS_WORK_HELPER(cache_helper);
+BTRFS_WORK_HELPER(submit_helper);
+BTRFS_WORK_HELPER(fixup_helper);
+BTRFS_WORK_HELPER(endio_helper);
+BTRFS_WORK_HELPER(endio_meta_helper);
+BTRFS_WORK_HELPER(endio_meta_write_helper);
+BTRFS_WORK_HELPER(endio_raid56_helper);
+BTRFS_WORK_HELPER(rmw_helper);
+BTRFS_WORK_HELPER(endio_write_helper);
+BTRFS_WORK_HELPER(freespace_write_helper);
+BTRFS_WORK_HELPER(delayed_meta_helper);
+BTRFS_WORK_HELPER(readahead_helper);
+BTRFS_WORK_HELPER(qgroup_rescan_helper);
+BTRFS_WORK_HELPER(extent_refs_helper);
+BTRFS_WORK_HELPER(scrub_helper);
+BTRFS_WORK_HELPER(scrubwrc_helper);
+BTRFS_WORK_HELPER(scrubnc_helper);
+
+static struct __btrfs_workqueue *
+__btrfs_alloc_workqueue(const char *name, int flags, int max_active,
 			 int thresh)
 {
 	struct __btrfs_workqueue *ret = kzalloc(sizeof(*ret), GFP_NOFS);
@@ -232,13 +262,11 @@ static void run_ordered_work(struct __btrfs_workqueue *wq)
 	spin_unlock_irqrestore(lock, flags);
 }
 
-static void normal_work_helper(struct work_struct *arg)
+static void normal_work_helper(struct btrfs_work *work)
 {
-	struct btrfs_work *work;
 	struct __btrfs_workqueue *wq;
 	int need_order = 0;
 
-	work = container_of(arg, struct btrfs_work, normal_work);
 	/*
 	 * We should not touch things inside work in the following cases:
 	 * 1) after work->func() if it has no ordered_free
@@ -262,7 +290,7 @@ static void normal_work_helper(struct work_struct *arg)
 		trace_btrfs_all_work_done(work);
 }
 
-void btrfs_init_work(struct btrfs_work *work,
+void btrfs_init_work(struct btrfs_work *work, btrfs_work_func_t uniq_func,
 		     btrfs_func_t func,
 		     btrfs_func_t ordered_func,
 		     btrfs_func_t ordered_free)
@@ -270,7 +298,7 @@ void btrfs_init_work(struct btrfs_work *work,
 	work->func = func;
 	work->ordered_func = ordered_func;
 	work->ordered_free = ordered_free;
-	INIT_WORK(&work->normal_work, normal_work_helper);
+	INIT_WORK(&work->normal_work, uniq_func);
 	INIT_LIST_HEAD(&work->ordered_list);
 	work->flags = 0;
 }
diff --git a/fs/btrfs/async-thread.h b/fs/btrfs/async-thread.h
index 9c6b66d15fb0..e9e31c94758f 100644
--- a/fs/btrfs/async-thread.h
+++ b/fs/btrfs/async-thread.h
@@ -19,12 +19,14 @@
 
 #ifndef __BTRFS_ASYNC_THREAD_
 #define __BTRFS_ASYNC_THREAD_
+#include <linux/workqueue.h>
 
 struct btrfs_workqueue;
 /* Internal use only */
 struct __btrfs_workqueue;
 struct btrfs_work;
 typedef void (*btrfs_func_t)(struct btrfs_work *arg);
+typedef void (*btrfs_work_func_t)(struct work_struct *arg);
 
 struct btrfs_work {
 	btrfs_func_t func;
@@ -38,11 +40,35 @@ struct btrfs_work {
 	unsigned long flags;
 };
 
+#define BTRFS_WORK_HELPER_PROTO(name)					\
+void btrfs_##name(struct work_struct *arg)
+
+BTRFS_WORK_HELPER_PROTO(worker_helper);
+BTRFS_WORK_HELPER_PROTO(delalloc_helper);
+BTRFS_WORK_HELPER_PROTO(flush_delalloc_helper);
+BTRFS_WORK_HELPER_PROTO(cache_helper);
+BTRFS_WORK_HELPER_PROTO(submit_helper);
+BTRFS_WORK_HELPER_PROTO(fixup_helper);
+BTRFS_WORK_HELPER_PROTO(endio_helper);
+BTRFS_WORK_HELPER_PROTO(endio_meta_helper);
+BTRFS_WORK_HELPER_PROTO(endio_meta_write_helper);
+BTRFS_WORK_HELPER_PROTO(endio_raid56_helper);
+BTRFS_WORK_HELPER_PROTO(rmw_helper);
+BTRFS_WORK_HELPER_PROTO(endio_write_helper);
+BTRFS_WORK_HELPER_PROTO(freespace_write_helper);
+BTRFS_WORK_HELPER_PROTO(delayed_meta_helper);
+BTRFS_WORK_HELPER_PROTO(readahead_helper);
+BTRFS_WORK_HELPER_PROTO(qgroup_rescan_helper);
+BTRFS_WORK_HELPER_PROTO(extent_refs_helper);
+BTRFS_WORK_HELPER_PROTO(scrub_helper);
+BTRFS_WORK_HELPER_PROTO(scrubwrc_helper);
+BTRFS_WORK_HELPER_PROTO(scrubnc_helper);
+
 struct btrfs_workqueue *btrfs_alloc_workqueue(const char *name,
 					      int flags,
 					      int max_active,
 					      int thresh);
-void btrfs_init_work(struct btrfs_work *work,
+void btrfs_init_work(struct btrfs_work *work, btrfs_work_func_t helper,
 		     btrfs_func_t func,
 		     btrfs_func_t ordered_func,
 		     btrfs_func_t ordered_free);
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index da775bfdebc9..a2e90f855d7d 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -1395,8 +1395,8 @@ static int btrfs_wq_run_delayed_node(struct btrfs_delayed_root *delayed_root,
 		return -ENOMEM;
 
 	async_work->delayed_root = delayed_root;
-	btrfs_init_work(&async_work->work, btrfs_async_run_delayed_root,
-			NULL, NULL);
+	btrfs_init_work(&async_work->work, btrfs_delayed_meta_helper,
+			btrfs_async_run_delayed_root, NULL, NULL);
 	async_work->nr = nr;
 
 	btrfs_queue_work(root->fs_info->delayed_workers, &async_work->work);
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index c99a414813c1..a1d36e62179c 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -39,7 +39,6 @@
 #include "btrfs_inode.h"
 #include "volumes.h"
 #include "print-tree.h"
-#include "async-thread.h"
 #include "locking.h"
 #include "tree-log.h"
 #include "free-space-cache.h"
@@ -693,35 +692,41 @@ static void end_workqueue_bio(struct bio *bio, int err)
 {
 	struct end_io_wq *end_io_wq = bio->bi_private;
 	struct btrfs_fs_info *fs_info;
+	struct btrfs_workqueue *wq;
+	btrfs_work_func_t func;
 
 	fs_info = end_io_wq->info;
 	end_io_wq->error = err;
-	btrfs_init_work(&end_io_wq->work, end_workqueue_fn, NULL, NULL);
 
 	if (bio->bi_rw & REQ_WRITE) {
-		if (end_io_wq->metadata == BTRFS_WQ_ENDIO_METADATA)
-			btrfs_queue_work(fs_info->endio_meta_write_workers,
-					 &end_io_wq->work);
-		else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_FREE_SPACE)
-			btrfs_queue_work(fs_info->endio_freespace_worker,
-					 &end_io_wq->work);
-		else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
-			btrfs_queue_work(fs_info->endio_raid56_workers,
-					 &end_io_wq->work);
-		else
-			btrfs_queue_work(fs_info->endio_write_workers,
-					 &end_io_wq->work);
+		if (end_io_wq->metadata == BTRFS_WQ_ENDIO_METADATA) {
+			wq = fs_info->endio_meta_write_workers;
+			func = btrfs_endio_meta_write_helper;
+		} else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_FREE_SPACE) {
+			wq = fs_info->endio_freespace_worker;
+			func = btrfs_freespace_write_helper;
+		} else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56) {
+			wq = fs_info->endio_raid56_workers;
+			func = btrfs_endio_raid56_helper;
+		} else {
+			wq = fs_info->endio_write_workers;
+			func = btrfs_endio_write_helper;
+		}
 	} else {
-		if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
-			btrfs_queue_work(fs_info->endio_raid56_workers,
-					 &end_io_wq->work);
-		else if (end_io_wq->metadata)
-			btrfs_queue_work(fs_info->endio_meta_workers,
-					 &end_io_wq->work);
-		else
-			btrfs_queue_work(fs_info->endio_workers,
-					 &end_io_wq->work);
+		if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56) {
+			wq = fs_info->endio_raid56_workers;
+			func = btrfs_endio_raid56_helper;
+		} else if (end_io_wq->metadata) {
+			wq = fs_info->endio_meta_workers;
+			func = btrfs_endio_meta_helper;
+		} else {
+			wq = fs_info->endio_workers;
+			func = btrfs_endio_helper;
+		}
 	}
+
+	btrfs_init_work(&end_io_wq->work, func, end_workqueue_fn, NULL, NULL);
+	btrfs_queue_work(wq, &end_io_wq->work);
 }
 
 /*
@@ -828,7 +833,7 @@ int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode,
 	async->submit_bio_start = submit_bio_start;
 	async->submit_bio_done = submit_bio_done;
 
-	btrfs_init_work(&async->work, run_one_async_start,
+	btrfs_init_work(&async->work, btrfs_worker_helper, run_one_async_start,
 			run_one_async_done, run_one_async_free);
 
 	async->bio_flags = bio_flags;
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 5524434da059..3efe1c3877bf 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -552,7 +552,8 @@ static int cache_block_group(struct btrfs_block_group_cache *cache,
 	caching_ctl->block_group = cache;
 	caching_ctl->progress = cache->key.objectid;
 	atomic_set(&caching_ctl->count, 1);
-	btrfs_init_work(&caching_ctl->work, caching_thread, NULL, NULL);
+	btrfs_init_work(&caching_ctl->work, btrfs_cache_helper,
+			caching_thread, NULL, NULL);
 
 	spin_lock(&cache->lock);
 	/*
@@ -2749,8 +2750,8 @@ int btrfs_async_run_delayed_refs(struct btrfs_root *root,
 		async->sync = 0;
 	init_completion(&async->wait);
 
-	btrfs_init_work(&async->work, delayed_ref_async_start,
-			NULL, NULL);
+	btrfs_init_work(&async->work, btrfs_extent_refs_helper,
+			delayed_ref_async_start, NULL, NULL);
 
 	btrfs_queue_work(root->fs_info->extent_workers, &async->work);
 
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index ae98df67950f..3d020d6d9ace 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -1096,8 +1096,10 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page,
 		async_cow->end = cur_end;
 		INIT_LIST_HEAD(&async_cow->extents);
 
-		btrfs_init_work(&async_cow->work, async_cow_start,
-				async_cow_submit, async_cow_free);
+		btrfs_init_work(&async_cow->work,
+				btrfs_delalloc_helper,
+				async_cow_start, async_cow_submit,
+				async_cow_free);
 
 		nr_pages = (cur_end - start + PAGE_CACHE_SIZE) >>
 			PAGE_CACHE_SHIFT;
@@ -1881,7 +1883,8 @@ static int btrfs_writepage_start_hook(struct page *page, u64 start, u64 end)
 
 	SetPageChecked(page);
 	page_cache_get(page);
-	btrfs_init_work(&fixup->work, btrfs_writepage_fixup_worker, NULL, NULL);
+	btrfs_init_work(&fixup->work, btrfs_fixup_helper,
+			btrfs_writepage_fixup_worker, NULL, NULL);
 	fixup->page = page;
 	btrfs_queue_work(root->fs_info->fixup_workers, &fixup->work);
 	return -EBUSY;
@@ -2822,7 +2825,8 @@ static int btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end,
 	struct inode *inode = page->mapping->host;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_ordered_extent *ordered_extent = NULL;
-	struct btrfs_workqueue *workers;
+	struct btrfs_workqueue *wq;
+	btrfs_work_func_t func;
 
 	trace_btrfs_writepage_end_io_hook(page, start, end, uptodate);
 
@@ -2831,13 +2835,17 @@ static int btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end,
 					    end - start + 1, uptodate))
 		return 0;
 
-	btrfs_init_work(&ordered_extent->work, finish_ordered_fn, NULL, NULL);
+	if (btrfs_is_free_space_inode(inode)) {
+		wq = root->fs_info->endio_freespace_worker;
+		func = btrfs_freespace_write_helper;
+	} else {
+		wq = root->fs_info->endio_write_workers;
+		func = btrfs_endio_write_helper;
+	}
 
-	if (btrfs_is_free_space_inode(inode))
-		workers = root->fs_info->endio_freespace_worker;
-	else
-		workers = root->fs_info->endio_write_workers;
-	btrfs_queue_work(workers, &ordered_extent->work);
+	btrfs_init_work(&ordered_extent->work, func, finish_ordered_fn, NULL,
+			NULL);
+	btrfs_queue_work(wq, &ordered_extent->work);
 
 	return 0;
 }
@@ -7208,7 +7216,8 @@ again:
 	if (!ret)
 		goto out_test;
 
-	btrfs_init_work(&ordered->work, finish_ordered_fn, NULL, NULL);
+	btrfs_init_work(&ordered->work, btrfs_endio_write_helper,
+			finish_ordered_fn, NULL, NULL);
 	btrfs_queue_work(root->fs_info->endio_write_workers,
 			 &ordered->work);
 out_test:
@@ -8535,7 +8544,9 @@ struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode,
 	work->inode = inode;
 	work->wait = wait;
 	work->delay_iput = delay_iput;
-	btrfs_init_work(&work->work, btrfs_run_delalloc_work, NULL, NULL);
+	WARN_ON_ONCE(!inode);
+	btrfs_init_work(&work->work, btrfs_flush_delalloc_helper,
+			btrfs_run_delalloc_work, NULL, NULL);
 
 	return work;
 }
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index 963895c1f801..ac734ec4cc20 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -615,6 +615,7 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr)
 		spin_unlock(&root->ordered_extent_lock);
 
 		btrfs_init_work(&ordered->flush_work,
+				btrfs_flush_delalloc_helper,
 				btrfs_run_ordered_extent_work, NULL, NULL);
 		list_add_tail(&ordered->work_list, &works);
 		btrfs_queue_work(root->fs_info->flush_workers,
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index 8abe45524de9..ded5c601d916 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -2720,6 +2720,7 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid,
 	memset(&fs_info->qgroup_rescan_work, 0,
 	       sizeof(fs_info->qgroup_rescan_work));
 	btrfs_init_work(&fs_info->qgroup_rescan_work,
+			btrfs_qgroup_rescan_helper,
 			btrfs_qgroup_rescan_worker, NULL, NULL);
 
 	if (ret) {
diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c
index 4a88f073fdd7..0a6b6e4bcbb9 100644
--- a/fs/btrfs/raid56.c
+++ b/fs/btrfs/raid56.c
@@ -1416,7 +1416,8 @@ cleanup:
 
 static void async_rmw_stripe(struct btrfs_raid_bio *rbio)
 {
-	btrfs_init_work(&rbio->work, rmw_work, NULL, NULL);
+	btrfs_init_work(&rbio->work, btrfs_rmw_helper,
+			rmw_work, NULL, NULL);
 
 	btrfs_queue_work(rbio->fs_info->rmw_workers,
 			 &rbio->work);
@@ -1424,7 +1425,8 @@ static void async_rmw_stripe(struct btrfs_raid_bio *rbio)
 
 static void async_read_rebuild(struct btrfs_raid_bio *rbio)
 {
-	btrfs_init_work(&rbio->work, read_rebuild_work, NULL, NULL);
+	btrfs_init_work(&rbio->work, btrfs_rmw_helper,
+			read_rebuild_work, NULL, NULL);
 
 	btrfs_queue_work(rbio->fs_info->rmw_workers,
 			 &rbio->work);
@@ -1665,7 +1667,8 @@ static void btrfs_raid_unplug(struct blk_plug_cb *cb, bool from_schedule)
 	plug = container_of(cb, struct btrfs_plug_cb, cb);
 
 	if (from_schedule) {
-		btrfs_init_work(&plug->work, unplug_work, NULL, NULL);
+		btrfs_init_work(&plug->work, btrfs_rmw_helper,
+				unplug_work, NULL, NULL);
 		btrfs_queue_work(plug->info->rmw_workers,
 				 &plug->work);
 		return;
diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c
index 09230cf3a244..20408c6b665a 100644
--- a/fs/btrfs/reada.c
+++ b/fs/btrfs/reada.c
@@ -798,7 +798,8 @@ static void reada_start_machine(struct btrfs_fs_info *fs_info)
 		/* FIXME we cannot handle this properly right now */
 		BUG();
 	}
-	btrfs_init_work(&rmw->work, reada_start_machine_worker, NULL, NULL);
+	btrfs_init_work(&rmw->work, btrfs_readahead_helper,
+			reada_start_machine_worker, NULL, NULL);
 	rmw->fs_info = fs_info;
 
 	btrfs_queue_work(fs_info->readahead_workers, &rmw->work);
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index 23d3f6e6a482..f4a41f37be22 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -428,8 +428,8 @@ struct scrub_ctx *scrub_setup_ctx(struct btrfs_device *dev, int is_dev_replace)
 		sbio->index = i;
 		sbio->sctx = sctx;
 		sbio->page_count = 0;
-		btrfs_init_work(&sbio->work, scrub_bio_end_io_worker,
-				NULL, NULL);
+		btrfs_init_work(&sbio->work, btrfs_scrub_helper,
+				scrub_bio_end_io_worker, NULL, NULL);
 
 		if (i != SCRUB_BIOS_PER_SCTX - 1)
 			sctx->bios[i]->next_free = i + 1;
@@ -999,8 +999,8 @@ nodatasum_case:
 		fixup_nodatasum->root = fs_info->extent_root;
 		fixup_nodatasum->mirror_num = failed_mirror_index + 1;
 		scrub_pending_trans_workers_inc(sctx);
-		btrfs_init_work(&fixup_nodatasum->work, scrub_fixup_nodatasum,
-				NULL, NULL);
+		btrfs_init_work(&fixup_nodatasum->work, btrfs_scrub_helper,
+				scrub_fixup_nodatasum, NULL, NULL);
 		btrfs_queue_work(fs_info->scrub_workers,
 				 &fixup_nodatasum->work);
 		goto out;
@@ -1616,7 +1616,8 @@ static void scrub_wr_bio_end_io(struct bio *bio, int err)
 	sbio->err = err;
 	sbio->bio = bio;
 
-	btrfs_init_work(&sbio->work, scrub_wr_bio_end_io_worker, NULL, NULL);
+	btrfs_init_work(&sbio->work, btrfs_scrubwrc_helper,
+			 scrub_wr_bio_end_io_worker, NULL, NULL);
 	btrfs_queue_work(fs_info->scrub_wr_completion_workers, &sbio->work);
 }
 
@@ -3214,7 +3215,8 @@ static int copy_nocow_pages(struct scrub_ctx *sctx, u64 logical, u64 len,
 	nocow_ctx->len = len;
 	nocow_ctx->mirror_num = mirror_num;
 	nocow_ctx->physical_for_dev_replace = physical_for_dev_replace;
-	btrfs_init_work(&nocow_ctx->work, copy_nocow_pages_worker, NULL, NULL);
+	btrfs_init_work(&nocow_ctx->work, btrfs_scrubnc_helper,
+			copy_nocow_pages_worker, NULL, NULL);
 	INIT_LIST_HEAD(&nocow_ctx->inodes);
 	btrfs_queue_work(fs_info->scrub_nocow_workers,
 			 &nocow_ctx->work);
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 9d4ce53d7569..340a92d08e84 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -5854,7 +5854,8 @@ struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
 	else
 		generate_random_uuid(dev->uuid);
 
-	btrfs_init_work(&dev->work, pending_bios_fn, NULL, NULL);
+	btrfs_init_work(&dev->work, btrfs_submit_helper,
+			pending_bios_fn, NULL, NULL);
 
 	return dev;
 }
-- 
cgit v1.2.3