Re: [PATCH] erofs: ensure m_llen is reset to 0 if metadata is invalid

From: Gao Xiang
Date: Sat Jun 29 2024 - 22:51:41 EST




On 2024/6/30 02:57, Gao Xiang wrote:
Sometimes, the on-disk metadata might be invalid due to storage
failure or other unknown issues.
user interrupts, storage failures, or other unknown causes.


In that case, z_erofs_map_blocks_iter() may still return a valid
m_llen while other fields remain invalid (e.g., m_plen can be 0).

Due to the return value of z_erofs_scan_folio() in some path will
be ignored on purpose, the following z_erofs_scan_folio() could
then use the invalid value by accident.

Let's reset m_llen to 0 to prevent this.

Signed-off-by: Gao Xiang <hsiangkao@xxxxxxxxxxxxxxxxx>
---
fs/erofs/zmap.c | 2 ++
1 file changed, 2 insertions(+)

diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c
index 9b248ee5fef2..74d3d7bffcf3 100644
--- a/fs/erofs/zmap.c
+++ b/fs/erofs/zmap.c
@@ -711,6 +711,8 @@ int z_erofs_map_blocks_iter(struct inode *inode, struct erofs_map_blocks *map,
err = z_erofs_do_map_blocks(inode, map, flags);
out:
+ if (err)
+ map->m_llen = 0;
trace_z_erofs_map_blocks_iter_exit(inode, map, flags, err);
return err;
}