1 files changed, 189 insertions, 170 deletions

diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c
index ffb1036ef10d..6a7e297ab0a7 100644
--- a/fs/btrfs/root-tree.c
+++ b/fs/btrfs/root-tree.c
@@ -1,26 +1,20 @@
+// SPDX-License-Identifier: GPL-2.0
 /*
  * Copyright (C) 2007 Oracle.  All rights reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public
- * License v2 as published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public
- * License along with this program; if not, write to the
- * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
- * Boston, MA 021110-1307, USA.
  */
 
+#include <linux/err.h>
 #include <linux/uuid.h>
 #include "ctree.h"
+#include "fs.h"
+#include "messages.h"
 #include "transaction.h"
 #include "disk-io.h"
-#include "print-tree.h"
+#include "qgroup.h"
+#include "space-info.h"
+#include "accessors.h"
+#include "root-tree.h"
+#include "orphan.h"
 
 /*
  * Read a root item from the tree. In case we detect a root item smaller then
@@ -29,55 +23,48 @@
  * generation numbers as then we know the root was once mounted with an older
  * kernel that was not aware of the root item structure change.
  */
-void btrfs_read_root_item(struct extent_buffer *eb, int slot,
-			  struct btrfs_root_item *item)
+static void btrfs_read_root_item(struct extent_buffer *eb, int slot,
+				struct btrfs_root_item *item)
 {
-	uuid_le uuid;
-	int len;
+	u32 len;
 	int need_reset = 0;
 
-	len = btrfs_item_size_nr(eb, slot);
+	len = btrfs_item_size(eb, slot);
 	read_extent_buffer(eb, item, btrfs_item_ptr_offset(eb, slot),
-			min_t(int, len, (int)sizeof(*item)));
+			   min_t(u32, len, sizeof(*item)));
 	if (len < sizeof(*item))
 		need_reset = 1;
 	if (!need_reset && btrfs_root_generation(item)
 		!= btrfs_root_generation_v2(item)) {
 		if (btrfs_root_generation_v2(item) != 0) {
-			printk(KERN_WARNING "btrfs: mismatching "
-					"generation and generation_v2 "
-					"found in root item. This root "
-					"was probably mounted with an "
-					"older kernel. Resetting all "
-					"new fields.\n");
+			btrfs_warn(eb->fs_info,
+					"mismatching generation and generation_v2 found in root item. This root was probably mounted with an older kernel. Resetting all new fields.");
 		}
 		need_reset = 1;
 	}
 	if (need_reset) {
-		memset(&item->generation_v2, 0,
-			sizeof(*item) - offsetof(struct btrfs_root_item,
-					generation_v2));
-
-		uuid_le_gen(&uuid);
-		memcpy(item->uuid, uuid.b, BTRFS_UUID_SIZE);
+		/* Clear all members from generation_v2 onwards. */
+		memset_startat(item, 0, generation_v2);
+		generate_random_guid(item->uuid);
 	}
 }
 
 /*
- * btrfs_find_root - lookup the root by the key.
+ * Lookup the root by the key.
+ *
  * root: the root of the root tree
  * search_key: the key to search
  * path: the path we search
  * root_item: the root item of the tree we look for
- * root_key: the reak key of the tree we look for
+ * root_key: the root key of the tree we look for
  *
- * If ->offset of 'seach_key' is -1ULL, it means we are not sure the offset
+ * If ->offset of 'search_key' is -1ULL, it means we are not sure the offset
  * of the search key, just lookup the root with the highest offset for a
  * given objectid.
  *
  * If we find something return 0, otherwise > 0, < 0 on error.
  */
-int btrfs_find_root(struct btrfs_root *root, struct btrfs_key *search_key,
+int btrfs_find_root(struct btrfs_root *root, const struct btrfs_key *search_key,
 		    struct btrfs_path *path, struct btrfs_root_item *root_item,
 		    struct btrfs_key *root_key)
 {
@@ -94,7 +81,14 @@ int btrfs_find_root(struct btrfs_root *root, struct btrfs_key *search_key,
 		if (ret > 0)
 			goto out;
 	} else {
-		BUG_ON(ret == 0);		/* Logical error */
+		/*
+		 * Key with offset -1 found, there would have to exist a root
+		 * with such id, but this is out of the valid range.
+		 */
+		if (unlikely(ret == 0)) {
+			ret = -EUCLEAN;
+			goto out;
+		}
 		if (path->slots[0] == 0)
 			goto out;
 		path->slots[0]--;
@@ -135,35 +129,35 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root
 		      *root, struct btrfs_key *key, struct btrfs_root_item
 		      *item)
 {
-	struct btrfs_path *path;
+	struct btrfs_fs_info *fs_info = root->fs_info;
+	BTRFS_PATH_AUTO_FREE(path);
 	struct extent_buffer *l;
 	int ret;
 	int slot;
 	unsigned long ptr;
-	int old_len;
+	u32 old_len;
 
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
 
 	ret = btrfs_search_slot(trans, root, key, path, 0, 1);
-	if (ret < 0) {
-		btrfs_abort_transaction(trans, root, ret);
-		goto out;
-	}
+	if (ret < 0)
+		return ret;
 
-	if (ret != 0) {
-		btrfs_print_leaf(root, path->nodes[0]);
-		printk(KERN_CRIT "unable to update root key %llu %u %llu\n",
-		       (unsigned long long)key->objectid, key->type,
-		       (unsigned long long)key->offset);
-		BUG_ON(1);
+	if (unlikely(ret > 0)) {
+		btrfs_crit(fs_info,
+			   "unable to find root key " BTRFS_KEY_FMT " in tree %llu",
+			   BTRFS_KEY_FMT_VALUE(key), btrfs_root_id(root));
+		ret = -EUCLEAN;
+		btrfs_abort_transaction(trans, ret);
+		return ret;
 	}
 
 	l = path->nodes[0];
 	slot = path->slots[0];
 	ptr = btrfs_item_ptr_offset(l, slot);
-	old_len = btrfs_item_size_nr(l, slot);
+	old_len = btrfs_item_size(l, slot);
 
 	/*
 	 * If this is the first time we update the root item which originated
@@ -174,22 +168,22 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root
 		btrfs_release_path(path);
 		ret = btrfs_search_slot(trans, root, key, path,
 				-1, 1);
-		if (ret < 0) {
-			btrfs_abort_transaction(trans, root, ret);
-			goto out;
+		if (unlikely(ret < 0)) {
+			btrfs_abort_transaction(trans, ret);
+			return ret;
 		}
 
 		ret = btrfs_del_item(trans, root, path);
-		if (ret < 0) {
-			btrfs_abort_transaction(trans, root, ret);
-			goto out;
+		if (unlikely(ret < 0)) {
+			btrfs_abort_transaction(trans, ret);
+			return ret;
 		}
 		btrfs_release_path(path);
 		ret = btrfs_insert_empty_item(trans, root, path,
 				key, sizeof(*item));
-		if (ret < 0) {
-			btrfs_abort_transaction(trans, root, ret);
-			goto out;
+		if (unlikely(ret < 0)) {
+			btrfs_abort_transaction(trans, ret);
+			return ret;
 		}
 		l = path->nodes[0];
 		slot = path->slots[0];
@@ -203,14 +197,11 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root
 	btrfs_set_root_generation_v2(item, btrfs_root_generation(item));
 
 	write_extent_buffer(l, item, ptr, sizeof(*item));
-	btrfs_mark_buffer_dirty(path->nodes[0]);
-out:
-	btrfs_free_path(path);
 	return ret;
 }
 
 int btrfs_insert_root(struct btrfs_trans_handle *trans, struct btrfs_root *root,
-		      struct btrfs_key *key, struct btrfs_root_item *item)
+		      const struct btrfs_key *key, struct btrfs_root_item *item)
 {
 	/*
 	 * Make sure generation v1 and v2 match. See update_root for details.
@@ -219,19 +210,15 @@ int btrfs_insert_root(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 	return btrfs_insert_item(trans, root, key, item, sizeof(*item));
 }
 
-int btrfs_find_orphan_roots(struct btrfs_root *tree_root)
+int btrfs_find_orphan_roots(struct btrfs_fs_info *fs_info)
 {
+	struct btrfs_root *tree_root = fs_info->tree_root;
 	struct extent_buffer *leaf;
-	struct btrfs_path *path;
+	BTRFS_PATH_AUTO_FREE(path);
 	struct btrfs_key key;
-	struct btrfs_key root_key;
 	struct btrfs_root *root;
 	int err = 0;
 	int ret;
-	bool can_recover = true;
-
-	if (tree_root->fs_info->sb->s_flags & MS_RDONLY)
-		can_recover = false;
 
 	path = btrfs_alloc_path();
 	if (!path)
@@ -241,10 +228,9 @@ int btrfs_find_orphan_roots(struct btrfs_root *tree_root)
 	key.type = BTRFS_ORPHAN_ITEM_KEY;
 	key.offset = 0;
 
-	root_key.type = BTRFS_ROOT_ITEM_KEY;
-	root_key.offset = (u64)-1;
-
 	while (1) {
+		u64 root_objectid;
+
 		ret = btrfs_search_slot(NULL, tree_root, &key, path, 0, 0);
 		if (ret < 0) {
 			err = ret;
@@ -268,11 +254,11 @@ int btrfs_find_orphan_roots(struct btrfs_root *tree_root)
 		    key.type != BTRFS_ORPHAN_ITEM_KEY)
 			break;
 
-		root_key.objectid = key.offset;
+		root_objectid = key.offset;
 		key.offset++;
 
-		root = btrfs_read_fs_root(tree_root, &root_key);
-		err = PTR_RET(root);
+		root = btrfs_get_fs_root(fs_info, root_objectid, false);
+		err = PTR_ERR_OR_ZERO(root);
 		if (err && err != -ENOENT) {
 			break;
 		} else if (err == -ENOENT) {
@@ -283,53 +269,53 @@ int btrfs_find_orphan_roots(struct btrfs_root *tree_root)
 			trans = btrfs_join_transaction(tree_root);
 			if (IS_ERR(trans)) {
 				err = PTR_ERR(trans);
-				btrfs_error(tree_root->fs_info, err,
-					    "Failed to start trans to delete "
-					    "orphan item");
+				btrfs_handle_fs_error(fs_info, err,
+					    "Failed to start trans to delete orphan item");
 				break;
 			}
 			err = btrfs_del_orphan_item(trans, tree_root,
-						    root_key.objectid);
-			btrfs_end_transaction(trans, tree_root);
+						    root_objectid);
+			btrfs_end_transaction(trans);
 			if (err) {
-				btrfs_error(tree_root->fs_info, err,
-					    "Failed to delete root orphan "
-					    "item");
+				btrfs_handle_fs_error(fs_info, err,
+					    "Failed to delete root orphan item");
 				break;
 			}
 			continue;
 		}
 
+		WARN_ON(!test_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state));
 		if (btrfs_root_refs(&root->root_item) == 0) {
-			btrfs_add_dead_root(root);
-			continue;
-		}
-
-		err = btrfs_init_fs_root(root);
-		if (err) {
-			btrfs_free_fs_root(root);
-			break;
-		}
-
-		root->orphan_item_inserted = 1;
+			struct btrfs_key drop_key;
+
+			btrfs_disk_key_to_cpu(&drop_key, &root->root_item.drop_progress);
+			/*
+			 * If we have a non-zero drop_progress then we know we
+			 * made it partly through deleting this snapshot, and
+			 * thus we need to make sure we block any balance from
+			 * happening until this snapshot is completely dropped.
+			 */
+			if (drop_key.objectid != 0 || drop_key.type != 0 ||
+			    drop_key.offset != 0) {
+				set_bit(BTRFS_FS_UNFINISHED_DROPS, &fs_info->flags);
+				set_bit(BTRFS_ROOT_UNFINISHED_DROP, &root->state);
+			}
 
-		err = btrfs_insert_fs_root(root->fs_info, root);
-		if (err) {
-			BUG_ON(err == -EEXIST);
-			btrfs_free_fs_root(root);
-			break;
+			set_bit(BTRFS_ROOT_DEAD_TREE, &root->state);
+			btrfs_add_dead_root(root);
 		}
+		btrfs_put_root(root);
 	}
 
-	btrfs_free_path(path);
 	return err;
 }
 
-/* drop the root item for 'key' from 'root' */
-int btrfs_del_root(struct btrfs_trans_handle *trans, struct btrfs_root *root,
-		   struct btrfs_key *key)
+/* drop the root item for 'key' from the tree root */
+int btrfs_del_root(struct btrfs_trans_handle *trans,
+		   const struct btrfs_key *key)
 {
-	struct btrfs_path *path;
+	struct btrfs_root *root = trans->fs_info->tree_root;
+	BTRFS_PATH_AUTO_FREE(path);
 	int ret;
 
 	path = btrfs_alloc_path();
@@ -337,28 +323,24 @@ int btrfs_del_root(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		return -ENOMEM;
 	ret = btrfs_search_slot(trans, root, key, path, -1, 1);
 	if (ret < 0)
-		goto out;
-
-	BUG_ON(ret != 0);
+		return ret;
+	if (unlikely(ret > 0))
+		/* The root must exist but we did not find it by the key. */
+		return -EUCLEAN;
 
-	ret = btrfs_del_item(trans, root, path);
-out:
-	btrfs_free_path(path);
-	return ret;
+	return btrfs_del_item(trans, root, path);
 }
 
-int btrfs_del_root_ref(struct btrfs_trans_handle *trans,
-		       struct btrfs_root *tree_root,
-		       u64 root_id, u64 ref_id, u64 dirid, u64 *sequence,
-		       const char *name, int name_len)
-
+int btrfs_del_root_ref(struct btrfs_trans_handle *trans, u64 root_id,
+		       u64 ref_id, u64 dirid, u64 *sequence,
+		       const struct fscrypt_str *name)
 {
-	struct btrfs_path *path;
+	struct btrfs_root *tree_root = trans->fs_info->tree_root;
+	BTRFS_PATH_AUTO_FREE(path);
 	struct btrfs_root_ref *ref;
 	struct extent_buffer *leaf;
 	struct btrfs_key key;
 	unsigned long ptr;
-	int err = 0;
 	int ret;
 
 	path = btrfs_alloc_path();
@@ -370,25 +352,26 @@ int btrfs_del_root_ref(struct btrfs_trans_handle *trans,
 	key.offset = ref_id;
 again:
 	ret = btrfs_search_slot(trans, tree_root, &key, path, -1, 1);
-	BUG_ON(ret < 0);
-	if (ret == 0) {
+	if (ret < 0) {
+		return ret;
+	} else if (ret == 0) {
 		leaf = path->nodes[0];
 		ref = btrfs_item_ptr(leaf, path->slots[0],
 				     struct btrfs_root_ref);
-
-		WARN_ON(btrfs_root_ref_dirid(leaf, ref) != dirid);
-		WARN_ON(btrfs_root_ref_name_len(leaf, ref) != name_len);
 		ptr = (unsigned long)(ref + 1);
-		WARN_ON(memcmp_extent_buffer(leaf, name, ptr, name_len));
+		if ((btrfs_root_ref_dirid(leaf, ref) != dirid) ||
+		    (btrfs_root_ref_name_len(leaf, ref) != name->len) ||
+		    memcmp_extent_buffer(leaf, name->name, ptr, name->len))
+			return -ENOENT;
+
 		*sequence = btrfs_root_ref_sequence(leaf, ref);
 
 		ret = btrfs_del_item(trans, tree_root, path);
-		if (ret) {
-			err = ret;
-			goto out;
-		}
-	} else
-		err = -ENOENT;
+		if (ret)
+			return ret;
+	} else {
+		return -ENOENT;
+	}
 
 	if (key.type == BTRFS_ROOT_BACKREF_KEY) {
 		btrfs_release_path(path);
@@ -398,23 +381,6 @@ again:
 		goto again;
 	}
 
-out:
-	btrfs_free_path(path);
-	return err;
-}
-
-int btrfs_find_root_ref(struct btrfs_root *tree_root,
-		   struct btrfs_path *path,
-		   u64 root_id, u64 ref_id)
-{
-	struct btrfs_key key;
-	int ret;
-
-	key.objectid = root_id;
-	key.type = BTRFS_ROOT_REF_KEY;
-	key.offset = ref_id;
-
-	ret = btrfs_search_slot(NULL, tree_root, &key, path, 0, 0);
 	return ret;
 }
 
@@ -433,14 +399,14 @@ int btrfs_find_root_ref(struct btrfs_root *tree_root,
  *
  * Will return 0, -ENOMEM, or anything from the CoW path
  */
-int btrfs_add_root_ref(struct btrfs_trans_handle *trans,
-		       struct btrfs_root *tree_root,
-		       u64 root_id, u64 ref_id, u64 dirid, u64 sequence,
-		       const char *name, int name_len)
+int btrfs_add_root_ref(struct btrfs_trans_handle *trans, u64 root_id,
+		       u64 ref_id, u64 dirid, u64 sequence,
+		       const struct fscrypt_str *name)
 {
+	struct btrfs_root *tree_root = trans->fs_info->tree_root;
 	struct btrfs_key key;
 	int ret;
-	struct btrfs_path *path;
+	BTRFS_PATH_AUTO_FREE(path);
 	struct btrfs_root_ref *ref;
 	struct extent_buffer *leaf;
 	unsigned long ptr;
@@ -454,10 +420,9 @@ int btrfs_add_root_ref(struct btrfs_trans_handle *trans,
 	key.offset = ref_id;
 again:
 	ret = btrfs_insert_empty_item(trans, tree_root, path, &key,
-				      sizeof(*ref) + name_len);
-	if (ret) {
-		btrfs_abort_transaction(trans, tree_root, ret);
-		btrfs_free_path(path);
+				      sizeof(*ref) + name->len);
+	if (unlikely(ret)) {
+		btrfs_abort_transaction(trans, ret);
 		return ret;
 	}
 
@@ -465,10 +430,9 @@ again:
 	ref = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_root_ref);
 	btrfs_set_root_ref_dirid(leaf, ref, dirid);
 	btrfs_set_root_ref_sequence(leaf, ref, sequence);
-	btrfs_set_root_ref_name_len(leaf, ref, name_len);
+	btrfs_set_root_ref_name_len(leaf, ref, name->len);
 	ptr = (unsigned long)(ref + 1);
-	write_extent_buffer(leaf, name, ptr, name_len);
-	btrfs_mark_buffer_dirty(leaf);
+	write_extent_buffer(leaf, name->name, ptr, name->len);
 
 	if (key.type == BTRFS_ROOT_BACKREF_KEY) {
 		btrfs_release_path(path);
@@ -478,7 +442,6 @@ again:
 		goto again;
 	}
 
-	btrfs_free_path(path);
 	return 0;
 }
 
@@ -490,13 +453,13 @@ again:
  */
 void btrfs_check_and_init_root_item(struct btrfs_root_item *root_item)
 {
-	u64 inode_flags = le64_to_cpu(root_item->inode.flags);
+	u64 inode_flags = btrfs_stack_inode_flags(&root_item->inode);
 
 	if (!(inode_flags & BTRFS_INODE_ROOT_ITEM_INIT)) {
 		inode_flags |= BTRFS_INODE_ROOT_ITEM_INIT;
-		root_item->inode.flags = cpu_to_le64(inode_flags);
-		root_item->flags = 0;
-		root_item->byte_limit = 0;
+		btrfs_set_stack_inode_flags(&root_item->inode, inode_flags);
+		btrfs_set_root_flags(root_item, 0);
+		btrfs_set_root_limit(root_item, 0);
 	}
 }
 
@@ -504,11 +467,67 @@ void btrfs_update_root_times(struct btrfs_trans_handle *trans,
 			     struct btrfs_root *root)
 {
 	struct btrfs_root_item *item = &root->root_item;
-	struct timespec ct = CURRENT_TIME;
+	struct timespec64 ct;
 
+	ktime_get_real_ts64(&ct);
 	spin_lock(&root->root_item_lock);
-	item->ctransid = cpu_to_le64(trans->transid);
-	item->ctime.sec = cpu_to_le64(ct.tv_sec);
-	item->ctime.nsec = cpu_to_le32(ct.tv_nsec);
+	btrfs_set_root_ctransid(item, trans->transid);
+	btrfs_set_stack_timespec_sec(&item->ctime, ct.tv_sec);
+	btrfs_set_stack_timespec_nsec(&item->ctime, ct.tv_nsec);
 	spin_unlock(&root->root_item_lock);
 }
+
+/*
+ * Reserve space for subvolume operation.
+ *
+ * root: the root of the parent directory
+ * rsv: block reservation
+ * items: the number of items that we need do reservation
+ * use_global_rsv: allow fallback to the global block reservation
+ *
+ * This function is used to reserve the space for snapshot/subvolume
+ * creation and deletion. Those operations are different with the
+ * common file/directory operations, they change two fs/file trees
+ * and root tree, the number of items that the qgroup reserves is
+ * different with the free space reservation. So we can not use
+ * the space reservation mechanism in start_transaction().
+ */
+int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
+				     struct btrfs_block_rsv *rsv, int items,
+				     bool use_global_rsv)
+{
+	u64 qgroup_num_bytes = 0;
+	u64 num_bytes;
+	int ret;
+	struct btrfs_fs_info *fs_info = root->fs_info;
+	struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv;
+
+	if (btrfs_qgroup_enabled(fs_info)) {
+		/* One for parent inode, two for dir entries */
+		qgroup_num_bytes = 3 * fs_info->nodesize;
+		ret = btrfs_qgroup_reserve_meta_prealloc(root,
+							 qgroup_num_bytes, true,
+							 false);
+		if (ret)
+			return ret;
+	}
+
+	num_bytes = btrfs_calc_insert_metadata_size(fs_info, items);
+	rsv->space_info = btrfs_find_space_info(fs_info,
+					    BTRFS_BLOCK_GROUP_METADATA);
+	ret = btrfs_block_rsv_add(fs_info, rsv, num_bytes,
+				  BTRFS_RESERVE_FLUSH_ALL);
+
+	if (ret == -ENOSPC && use_global_rsv)
+		ret = btrfs_block_rsv_migrate(global_rsv, rsv, num_bytes, true);
+
+	if (ret && qgroup_num_bytes)
+		btrfs_qgroup_free_meta_prealloc(root, qgroup_num_bytes);
+
+	if (!ret) {
+		spin_lock(&rsv->lock);
+		rsv->qgroup_rsv_reserved += qgroup_num_bytes;
+		spin_unlock(&rsv->lock);
+	}
+	return ret;
+}