f2fs: fix IS_CHECKPOINTED flag inconsistency issue caused by concurrent atomic commit and checkpoint writes

During SPO tests, when mounting F2FS, an -EINVAL error was returned from
f2fs_recover_inode_page. The issue occurred under the following scenario

Thread A                                     Thread B
f2fs_ioc_commit_atomic_write
 - f2fs_do_sync_file // atomic = true
  - f2fs_fsync_node_pages
    : last_folio = inode folio
    : schedule before folio_lock(last_folio) f2fs_write_checkpoint
                                              - block_operations// writeback last_folio
                                              - schedule before f2fs_flush_nat_entries
    : set_fsync_mark(last_folio, 1)
    : set_dentry_mark(last_folio, 1)
    : folio_mark_dirty(last_folio)
    - __write_node_folio(last_folio)
      : f2fs_down_read(&sbi->node_write)//block
                                              - f2fs_flush_nat_entries
                                                : {struct nat_entry}->flag |= BIT(IS_CHECKPOINTED)
                                              - unblock_operations
                                                : f2fs_up_write(&sbi->node_write)
                                             f2fs_write_checkpoint//return
      : f2fs_do_write_node_page()
f2fs_ioc_commit_atomic_write//return
                                             SPO

Thread A calls f2fs_need_dentry_mark(sbi, ino), and the last_folio has
already been written once. However, the {struct nat_entry}->flag did not
have the IS_CHECKPOINTED set, causing set_dentry_mark(last_folio, 1) and
write last_folio again after Thread B finishes f2fs_write_checkpoint.

After SPO and reboot, it was detected that {struct node_info}->blk_addr
was not NULL_ADDR because Thread B successfully write the checkpoint.

This issue only occurs in atomic write scenarios. For regular file
fsync operations, the folio must be dirty. If
block_operations->f2fs_sync_node_pages successfully submit the folio
write, this path will not be executed. Otherwise, the
f2fs_write_checkpoint will need to wait for the folio write submission
to complete, as sbi->nr_pages[F2FS_DIRTY_NODES] > 0. Therefore, the
situation where f2fs_need_dentry_mark checks that the {struct
nat_entry}->flag /wo the IS_CHECKPOINTED flag, but the folio write has
already been submitted, will not occur.

Therefore, for atomic file fsync, sbi->node_write should be acquired
through __write_node_folio to ensure that the IS_CHECKPOINTED flag
correctly indicates that the checkpoint write has been completed.

Fixes: 608514deba38 ("f2fs: set fsync mark only for the last dnode")
Cc: stable@kernel.org
Signed-off-by: Sheng Yong <shengyong1@xiaomi.com>
Signed-off-by: Jinbao Liu <liujinbao1@xiaomi.com>
Signed-off-by: Yongpeng Yang <yangyongpeng@xiaomi.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>

1 files changed, 10 insertions, 4 deletions

diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
index d378549010e6..99e425e8c00a 100644
--- a/fs/f2fs/node.c
+++ b/fs/f2fs/node.c
@@ -1786,8 +1786,13 @@ static bool __write_node_folio(struct folio *folio, bool atomic, bool *submitted
 		goto redirty_out;
 	}
 
-	if (atomic && !test_opt(sbi, NOBARRIER))
-		fio.op_flags |= REQ_PREFLUSH | REQ_FUA;
+	if (atomic) {
+		if (!test_opt(sbi, NOBARRIER))
+			fio.op_flags |= REQ_PREFLUSH | REQ_FUA;
+		if (IS_INODE(folio))
+			set_dentry_mark(folio,
+				f2fs_need_dentry_mark(sbi, ino_of_node(folio)));
+	}
 
 	/* should add to global list before clearing PAGECACHE status */
 	if (f2fs_in_warm_node_list(sbi, folio)) {
@@ -1928,8 +1933,9 @@ continue_unlock:
 					if (is_inode_flag_set(inode,
 								FI_DIRTY_INODE))
 						f2fs_update_inode(inode, folio);
-					set_dentry_mark(folio,
-						f2fs_need_dentry_mark(sbi, ino));
+					if (!atomic)
+						set_dentry_mark(folio,
+							f2fs_need_dentry_mark(sbi, ino));
 				}
 				/* may be written by other thread */
 				if (!folio_test_dirty(folio))