Btrfs: fix fsync after succession of renames and unlink/rmdir
After a succession of renames operations of different files and unlinking
one of them, if we fsync one of the renamed files we can end up with a
log that will either fail to replay at mount time or result in a filesystem
that is in an inconsistent state. One example scenario:
$ mkfs.btrfs -f /dev/sdb
$ mount /dev/sdb /mnt
$ mkdir /mnt/testdir
$ touch /mnt/testdir/fname1
$ touch /mnt/testdir/fname2
$ sync
$ mv /mnt/testdir/fname1 /mnt/testdir/fname3
$ rm -f /mnt/testdir/fname2
$ ln /mnt/testdir/fname3 /mnt/testdir/fname2
$ touch /mnt/testdir/fname1
$ xfs_io -c "fsync" /mnt/testdir/fname1
<power failure>
$ mount /dev/sdb /mnt
$ umount /mnt
$ btrfs check /dev/sdb
[1/7] checking root items
[2/7] checking extents
[3/7] checking free space cache
[4/7] checking fs roots
root 5 inode 259 errors 2, no orphan item
ERROR: errors found in fs roots
Opening filesystem to check...
Checking filesystem on /dev/sdc
UUID: 20e4abb8-5a19-4492-8bb4-6084125c2d0d
found 393216 bytes used, error(s) found
total csum bytes: 0
total tree bytes: 131072
total fs tree bytes: 32768
total extent tree bytes: 16384
btree space waste bytes: 122986
file data blocks allocated: 262144
referenced 262144
On a kernel without the first patch in this series, titled
"[PATCH] Btrfs: fix fsync after succession of renames of different files",
we get instead an error when mounting the filesystem due to failure of
replaying the log:
$ mount /dev/sdb /mnt
mount: mount /dev/sdb on /mnt failed: File exists
Fix this by logging the parent directory of an inode whenever we find an
inode that no longer exists (was unlinked in the current transaction),
during the procedure which finds inodes that have old names that collide
with new names of other inodes.
A test case for fstests follows soon.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 70d41f6..e76345f 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -27,6 +27,7 @@
#define LOG_INODE_ALL 0
#define LOG_INODE_EXISTS 1
#define LOG_OTHER_INODE 2
+#define LOG_OTHER_INODE_ALL 3
/*
* directory trouble cases
@@ -4777,7 +4778,7 @@
const int slot,
const struct btrfs_key *key,
struct btrfs_inode *inode,
- u64 *other_ino)
+ u64 *other_ino, u64 *other_parent)
{
int ret;
struct btrfs_path *search_path;
@@ -4843,6 +4844,7 @@
if (di_key.objectid != key->objectid) {
ret = 1;
*other_ino = di_key.objectid;
+ *other_parent = parent;
} else {
ret = 0;
}
@@ -4867,6 +4869,7 @@
struct btrfs_ino_list {
u64 ino;
+ u64 parent;
struct list_head list;
};
@@ -4874,7 +4877,7 @@
struct btrfs_root *root,
struct btrfs_path *path,
struct btrfs_log_ctx *ctx,
- u64 ino)
+ u64 ino, u64 parent)
{
struct btrfs_ino_list *ino_elem;
LIST_HEAD(inode_list);
@@ -4884,6 +4887,7 @@
if (!ino_elem)
return -ENOMEM;
ino_elem->ino = ino;
+ ino_elem->parent = parent;
list_add_tail(&ino_elem->list, &inode_list);
while (!list_empty(&inode_list)) {
@@ -4894,6 +4898,7 @@
ino_elem = list_first_entry(&inode_list, struct btrfs_ino_list,
list);
ino = ino_elem->ino;
+ parent = ino_elem->parent;
list_del(&ino_elem->list);
kfree(ino_elem);
if (ret)
@@ -4907,13 +4912,25 @@
inode = btrfs_iget(fs_info->sb, &key, root, NULL);
/*
* If the other inode that had a conflicting dir entry was
- * deleted in the current transaction, we don't need to do more
- * work nor fallback to a transaction commit.
+ * deleted in the current transaction, we need to log its parent
+ * directory.
*/
if (IS_ERR(inode)) {
ret = PTR_ERR(inode);
- if (ret == -ENOENT)
- ret = 0;
+ if (ret == -ENOENT) {
+ key.objectid = parent;
+ inode = btrfs_iget(fs_info->sb, &key, root,
+ NULL);
+ if (IS_ERR(inode)) {
+ ret = PTR_ERR(inode);
+ } else {
+ ret = btrfs_log_inode(trans, root,
+ BTRFS_I(inode),
+ LOG_OTHER_INODE_ALL,
+ 0, LLONG_MAX, ctx);
+ iput(inode);
+ }
+ }
continue;
}
/*
@@ -4943,6 +4960,7 @@
struct extent_buffer *leaf = path->nodes[0];
int slot = path->slots[0];
u64 other_ino = 0;
+ u64 other_parent = 0;
if (slot >= btrfs_header_nritems(leaf)) {
ret = btrfs_next_leaf(root, path);
@@ -4964,7 +4982,8 @@
}
ret = btrfs_check_ref_name_override(leaf, slot, &key,
- BTRFS_I(inode), &other_ino);
+ BTRFS_I(inode), &other_ino,
+ &other_parent);
if (ret < 0)
break;
if (ret > 0) {
@@ -4974,6 +4993,7 @@
break;
}
ino_elem->ino = other_ino;
+ ino_elem->parent = other_parent;
list_add_tail(&ino_elem->list, &inode_list);
ret = 0;
}
@@ -5024,7 +5044,7 @@
u64 logged_isize = 0;
bool need_log_inode_item = true;
bool xattrs_logged = false;
- bool recursive_logging = (inode_only == LOG_OTHER_INODE);
+ bool recursive_logging = false;
path = btrfs_alloc_path();
if (!path)
@@ -5070,8 +5090,12 @@
return ret;
}
- if (inode_only == LOG_OTHER_INODE) {
- inode_only = LOG_INODE_EXISTS;
+ if (inode_only == LOG_OTHER_INODE || inode_only == LOG_OTHER_INODE_ALL) {
+ recursive_logging = true;
+ if (inode_only == LOG_OTHER_INODE)
+ inode_only = LOG_INODE_EXISTS;
+ else
+ inode_only = LOG_INODE_ALL;
mutex_lock_nested(&inode->log_mutex, SINGLE_DEPTH_NESTING);
} else {
mutex_lock(&inode->log_mutex);
@@ -5169,10 +5193,11 @@
inode->generation == trans->transid &&
!recursive_logging) {
u64 other_ino = 0;
+ u64 other_parent = 0;
ret = btrfs_check_ref_name_override(path->nodes[0],
path->slots[0], &min_key, inode,
- &other_ino);
+ &other_ino, &other_parent);
if (ret < 0) {
err = ret;
goto out_unlock;
@@ -5195,7 +5220,7 @@
ins_nr = 0;
err = log_conflicting_inodes(trans, root, path,
- ctx, other_ino);
+ ctx, other_ino, other_parent);
if (err)
goto out_unlock;
btrfs_release_path(path);