Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
btrfs: zoned: use dedicated lock for data relocation
Currently, we use btrfs_inode_{lock,unlock}() to grant an exclusive writeback of the relocation data inode in btrfs_zoned_data_reloc_{lock,unlock}(). However, that can cause a deadlock in the following path. Thread A takes btrfs_inode_lock() and waits for metadata reservation by e.g, waiting for writeback: prealloc_file_extent_cluster() - btrfs_inode_lock(&inode->vfs_inode, 0); - btrfs_prealloc_file_range() ... - btrfs_replace_file_extents() - btrfs_start_transaction ... - btrfs_reserve_metadata_bytes() Thread B (e.g, doing a writeback work) needs to wait for the inode lock to continue writeback process: do_writepages - btrfs_writepages - extent_writpages - btrfs_zoned_data_reloc_lock(BTRFS_I(inode)); - btrfs_inode_lock() The deadlock is caused by relying on the vfs_inode's lock. By using it, we introduced unnecessary exclusion of writeback and btrfs_prealloc_file_range(). Also, the lock at this point is useless as we don't have any dirty pages in the inode yet. Introduce fs_info->zoned_data_reloc_io_lock and use it for the exclusive writeback. Fixes: 35156d8 ("btrfs: zoned: only allow one process to add pages to a relocation inode") CC: stable@vger.kernel.org # 5.16.x: 869f4cd: btrfs: zoned: encapsulate inode locking for zoned relocation CC: stable@vger.kernel.org # 5.16.x CC: stable@vger.kernel.org # 5.17 Cc: Johannes Thumshirn <johannes.thumshirn@wdc.com> Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Signed-off-by: Naohiro Aota <naohiro.aota@wdc.com> Signed-off-by: David Sterba <dsterba@suse.com>
- Loading branch information