From 6561a40ceced9082f50c374a22d5966cf9fc5f5c Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Wed, 18 Jun 2025 16:57:07 +0100 Subject: btrfs: fix missing error handling when searching for inode refs during log replay During log replay, at __add_inode_ref(), when we are searching for inode ref keys we totally ignore if btrfs_search_slot() returns an error. This may make a log replay succeed when there was an actual error and leave some metadata inconsistency in a subvolume tree. Fix this by checking if an error was returned from btrfs_search_slot() and if so, return it to the caller. Fixes: e02119d5a7b4 ("Btrfs: Add a write ahead tree log to optimize synchronous operations") Reviewed-by: Johannes Thumshirn Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/tree-log.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs/btrfs/tree-log.c') diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 858b609e292c..8b66173d9023 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -1073,7 +1073,9 @@ again: search_key.type = BTRFS_INODE_REF_KEY; search_key.offset = parent_objectid; ret = btrfs_search_slot(NULL, root, &search_key, path, 0, 0); - if (ret == 0) { + if (ret < 0) { + return ret; + } else if (ret == 0) { struct btrfs_inode_ref *victim_ref; unsigned long ptr; unsigned long ptr_end; -- cgit From 54a7081ed168b72a8a2d6ef4ba3a1259705a2926 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 23 Jun 2025 12:11:58 +0100 Subject: btrfs: fix iteration of extrefs during log replay At __inode_add_ref() when processing extrefs, if we jump into the next label we have an undefined value of victim_name.len, since we haven't initialized it before we did the goto. This results in an invalid memory access in the next iteration of the loop since victim_name.len was not initialized to the length of the name of the current extref. Fix this by initializing victim_name.len with the current extref's name length. Fixes: e43eec81c516 ("btrfs: use struct qstr instead of name and namelen pairs") Reviewed-by: Johannes Thumshirn Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/tree-log.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs/btrfs/tree-log.c') diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 8b66173d9023..d514bf531b3b 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -1148,13 +1148,13 @@ again: struct fscrypt_str victim_name; extref = (struct btrfs_inode_extref *)(base + cur_offset); + victim_name.len = btrfs_inode_extref_name_len(leaf, extref); if (btrfs_inode_extref_parent(leaf, extref) != parent_objectid) goto next; ret = read_alloc_one_name(leaf, &extref->name, - btrfs_inode_extref_name_len(leaf, extref), - &victim_name); + victim_name.len, &victim_name); if (ret) return ret; -- cgit From 5f61b961599acbd2bed028d3089105a1f7d224b8 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Wed, 18 Jun 2025 15:58:31 +0100 Subject: btrfs: fix inode lookup error handling during log replay When replaying log trees we use read_one_inode() to get an inode, which is just a wrapper around btrfs_iget_logging(), which in turn is a wrapper for btrfs_iget(). But read_one_inode() always returns NULL for any error that btrfs_iget_logging() / btrfs_iget() may return and this is a problem because: 1) In many callers of read_one_inode() we convert the NULL into -EIO, which is not accurate since btrfs_iget() may return -ENOMEM and -ENOENT for example, besides -EIO and other errors. So during log replay we may end up reporting a false -EIO, which is confusing since we may not have had any IO error at all; 2) When replaying directory deletes, at replay_dir_deletes(), we assume the NULL returned from read_one_inode() means that the inode doesn't exist and then proceed as if no error had happened. This is wrong because unless btrfs_iget() returned ERR_PTR(-ENOENT), we had an actual error and the target inode may exist in the target subvolume root - this may later result in the log replay code failing at a later stage (if we are "lucky") or succeed but leaving some inconsistency in the filesystem. So fix this by not ignoring errors from btrfs_iget_logging() and as a consequence remove the read_one_inode() wrapper and just use btrfs_iget_logging() directly. Also since btrfs_iget_logging() is supposed to be called only against subvolume roots, just like read_one_inode() which had a comment about it, add an assertion to btrfs_iget_logging() to check that the target root corresponds to a subvolume root. Fixes: 5d4f98a28c7d ("Btrfs: Mixed back reference (FORWARD ROLLING FORMAT CHANGE)") Reviewed-by: Johannes Thumshirn Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/tree-log.c | 127 +++++++++++++++++++++++++--------------------------- 1 file changed, 62 insertions(+), 65 deletions(-) (limited to 'fs/btrfs/tree-log.c') diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index d514bf531b3b..8cf5e5ae593c 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -143,6 +143,9 @@ static struct btrfs_inode *btrfs_iget_logging(u64 objectid, struct btrfs_root *r unsigned int nofs_flag; struct btrfs_inode *inode; + /* Only meant to be called for subvolume roots and not for log roots. */ + ASSERT(is_fstree(btrfs_root_id(root))); + /* * We're holding a transaction handle whether we are logging or * replaying a log tree, so we must make sure NOFS semantics apply @@ -604,21 +607,6 @@ static int read_alloc_one_name(struct extent_buffer *eb, void *start, int len, return 0; } -/* - * simple helper to read an inode off the disk from a given root - * This can only be called for subvolume roots and not for the log - */ -static noinline struct btrfs_inode *read_one_inode(struct btrfs_root *root, - u64 objectid) -{ - struct btrfs_inode *inode; - - inode = btrfs_iget_logging(objectid, root); - if (IS_ERR(inode)) - return NULL; - return inode; -} - /* replays a single extent in 'eb' at 'slot' with 'key' into the * subvolume 'root'. path is released on entry and should be released * on exit. @@ -674,9 +662,9 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans, return -EUCLEAN; } - inode = read_one_inode(root, key->objectid); - if (!inode) - return -EIO; + inode = btrfs_iget_logging(key->objectid, root); + if (IS_ERR(inode)) + return PTR_ERR(inode); /* * first check to see if we already have this extent in the @@ -948,9 +936,10 @@ static noinline int drop_one_dir_item(struct btrfs_trans_handle *trans, btrfs_release_path(path); - inode = read_one_inode(root, location.objectid); - if (!inode) { - ret = -EIO; + inode = btrfs_iget_logging(location.objectid, root); + if (IS_ERR(inode)) { + ret = PTR_ERR(inode); + inode = NULL; goto out; } @@ -1169,10 +1158,10 @@ again: kfree(victim_name.name); return ret; } else if (!ret) { - ret = -ENOENT; - victim_parent = read_one_inode(root, - parent_objectid); - if (victim_parent) { + victim_parent = btrfs_iget_logging(parent_objectid, root); + if (IS_ERR(victim_parent)) { + ret = PTR_ERR(victim_parent); + } else { inc_nlink(&inode->vfs_inode); btrfs_release_path(path); @@ -1317,9 +1306,9 @@ again: struct btrfs_inode *dir; btrfs_release_path(path); - dir = read_one_inode(root, parent_id); - if (!dir) { - ret = -ENOENT; + dir = btrfs_iget_logging(parent_id, root); + if (IS_ERR(dir)) { + ret = PTR_ERR(dir); kfree(name.name); goto out; } @@ -1391,15 +1380,17 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans, * copy the back ref in. The link count fixup code will take * care of the rest */ - dir = read_one_inode(root, parent_objectid); - if (!dir) { - ret = -ENOENT; + dir = btrfs_iget_logging(parent_objectid, root); + if (IS_ERR(dir)) { + ret = PTR_ERR(dir); + dir = NULL; goto out; } - inode = read_one_inode(root, inode_objectid); - if (!inode) { - ret = -EIO; + inode = btrfs_iget_logging(inode_objectid, root); + if (IS_ERR(inode)) { + ret = PTR_ERR(inode); + inode = NULL; goto out; } @@ -1411,11 +1402,13 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans, * parent object can change from one array * item to another. */ - if (!dir) - dir = read_one_inode(root, parent_objectid); if (!dir) { - ret = -ENOENT; - goto out; + dir = btrfs_iget_logging(parent_objectid, root); + if (IS_ERR(dir)) { + ret = PTR_ERR(dir); + dir = NULL; + goto out; + } } } else { ret = ref_get_fields(eb, ref_ptr, &name, &ref_index); @@ -1684,9 +1677,9 @@ static noinline int fixup_inode_link_counts(struct btrfs_trans_handle *trans, break; btrfs_release_path(path); - inode = read_one_inode(root, key.offset); - if (!inode) { - ret = -EIO; + inode = btrfs_iget_logging(key.offset, root); + if (IS_ERR(inode)) { + ret = PTR_ERR(inode); break; } @@ -1722,9 +1715,9 @@ static noinline int link_to_fixup_dir(struct btrfs_trans_handle *trans, struct btrfs_inode *inode; struct inode *vfs_inode; - inode = read_one_inode(root, objectid); - if (!inode) - return -EIO; + inode = btrfs_iget_logging(objectid, root); + if (IS_ERR(inode)) + return PTR_ERR(inode); vfs_inode = &inode->vfs_inode; key.objectid = BTRFS_TREE_LOG_FIXUP_OBJECTID; @@ -1763,14 +1756,14 @@ static noinline int insert_one_name(struct btrfs_trans_handle *trans, struct btrfs_inode *dir; int ret; - inode = read_one_inode(root, location->objectid); - if (!inode) - return -ENOENT; + inode = btrfs_iget_logging(location->objectid, root); + if (IS_ERR(inode)) + return PTR_ERR(inode); - dir = read_one_inode(root, dirid); - if (!dir) { + dir = btrfs_iget_logging(dirid, root); + if (IS_ERR(dir)) { iput(&inode->vfs_inode); - return -EIO; + return PTR_ERR(dir); } ret = btrfs_add_link(trans, dir, inode, name, 1, index); @@ -1847,9 +1840,9 @@ static noinline int replay_one_name(struct btrfs_trans_handle *trans, bool update_size = true; bool name_added = false; - dir = read_one_inode(root, key->objectid); - if (!dir) - return -EIO; + dir = btrfs_iget_logging(key->objectid, root); + if (IS_ERR(dir)) + return PTR_ERR(dir); ret = read_alloc_one_name(eb, di + 1, btrfs_dir_name_len(eb, di), &name); if (ret) @@ -2149,9 +2142,10 @@ static noinline int check_item_in_log(struct btrfs_trans_handle *trans, btrfs_dir_item_key_to_cpu(eb, di, &location); btrfs_release_path(path); btrfs_release_path(log_path); - inode = read_one_inode(root, location.objectid); - if (!inode) { - ret = -EIO; + inode = btrfs_iget_logging(location.objectid, root); + if (IS_ERR(inode)) { + ret = PTR_ERR(inode); + inode = NULL; goto out; } @@ -2303,14 +2297,17 @@ static noinline int replay_dir_deletes(struct btrfs_trans_handle *trans, if (!log_path) return -ENOMEM; - dir = read_one_inode(root, dirid); - /* it isn't an error if the inode isn't there, that can happen - * because we replay the deletes before we copy in the inode item - * from the log + dir = btrfs_iget_logging(dirid, root); + /* + * It isn't an error if the inode isn't there, that can happen because + * we replay the deletes before we copy in the inode item from the log. */ - if (!dir) { + if (IS_ERR(dir)) { btrfs_free_path(log_path); - return 0; + ret = PTR_ERR(dir); + if (ret == -ENOENT) + ret = 0; + return ret; } range_start = 0; @@ -2469,9 +2466,9 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb, struct btrfs_inode *inode; u64 from; - inode = read_one_inode(root, key.objectid); - if (!inode) { - ret = -EIO; + inode = btrfs_iget_logging(key.objectid, root); + if (IS_ERR(inode)) { + ret = PTR_ERR(inode); break; } from = ALIGN(i_size_read(&inode->vfs_inode), -- cgit From bf5bcf9a6fa070ec8a725b08db63fb1318f77366 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 19 Jun 2025 13:13:38 +0100 Subject: btrfs: record new subvolume in parent dir earlier to avoid dir logging races Instead of recording that a new subvolume was created in a directory after we add the entry do the directory, record it before adding the entry. This is to avoid races where after creating the entry and before recording the new subvolume in the directory (the call to btrfs_record_new_subvolume()), another task logs the directory, so we end up with a log tree where we logged a directory that has an entry pointing to a root that was not yet committed, resulting in an invalid entry if the log is persisted and replayed later due to a power failure or crash. Also state this requirement in the function comment for btrfs_record_new_subvolume(), similar to what we do for the btrfs_record_unlink_dir() and btrfs_record_snapshot_destroy(). Fixes: 45c4102f0d82 ("btrfs: avoid transaction commit on any fsync after subvolume creation") Reviewed-by: Johannes Thumshirn Signed-off-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/ioctl.c | 4 ++-- fs/btrfs/tree-log.c | 2 ++ 2 files changed, 4 insertions(+), 2 deletions(-) (limited to 'fs/btrfs/tree-log.c') diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 1e8f7082239c..e6f7d24e3b47 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -666,14 +666,14 @@ static noinline int create_subvol(struct mnt_idmap *idmap, goto out; } + btrfs_record_new_subvolume(trans, BTRFS_I(dir)); + ret = btrfs_create_new_inode(trans, &new_inode_args); if (ret) { btrfs_abort_transaction(trans, ret); goto out; } - btrfs_record_new_subvolume(trans, BTRFS_I(dir)); - d_instantiate_new(dentry, new_inode_args.inode); new_inode_args.inode = NULL; diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 8cf5e5ae593c..cea8a7e9d6d3 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -7447,6 +7447,8 @@ void btrfs_record_snapshot_destroy(struct btrfs_trans_handle *trans, * full log sync. * Also we don't need to worry with renames, since btrfs_rename() marks the log * for full commit when renaming a subvolume. + * + * Must be called before creating the subvolume entry in its parent directory. */ void btrfs_record_new_subvolume(const struct btrfs_trans_handle *trans, struct btrfs_inode *dir) -- cgit