6.6-stable review patch. If anyone has any objections, please let me know.
------------------
From: Qu Wenruo wqu@suse.com
[ Upstream commit 313ef70a9f0f637a09d9ef45222f5bdcf30a354b ]
Inside print_data_reloc_error(), if extent_from_logical() failed we return immediately.
However there are the following cases where extent_from_logical() can return error but still holds a path:
- btrfs_search_slot() returned 0
- No backref item found in extent tree
- No flags_ret provided This is not possible in this call site though.
So for the above two cases, we can return without releasing the path, causing extent buffer leaks.
Fixes: b9a9a85059cd ("btrfs: output affected files when relocation fails") Signed-off-by: Qu Wenruo wqu@suse.com Reviewed-by: David Sterba dsterba@suse.com Signed-off-by: David Sterba dsterba@suse.com Signed-off-by: Sasha Levin sashal@kernel.org --- fs/btrfs/inode.c | 1 + 1 file changed, 1 insertion(+)
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index b1be3e0fe7282..68bb5079aef74 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -266,6 +266,7 @@ static void print_data_reloc_error(const struct btrfs_inode *inode, u64 file_off if (ret < 0) { btrfs_err_rl(fs_info, "failed to lookup extent item for logical %llu: %d", logical, ret); + btrfs_release_path(&path); return; } eb = path.nodes[0];