5.10-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kemeng Shi <shikemeng@xxxxxxxxxxxxxxx>
[ Upstream commit 227d31b9214d1b9513383cf6c7180628d4b3b61f ]
If a group is marked EXT4_GROUP_INFO_IBITMAP_CORRUPT after it's inode
bitmap buffer_head was successfully verified, then __ext4_new_inode()
will get a valid inode_bitmap_bh of a corrupted group from
ext4_read_inode_bitmap() in which case inode_bitmap_bh misses a release.
Hnadle "IS_ERR(inode_bitmap_bh)" and group corruption separately like
how ext4_free_inode() does to avoid buffer_head leak.
Fixes: 9008a58e5dce ("ext4: make the bitmap read routines return real error codes")
Signed-off-by: Kemeng Shi <shikemeng@xxxxxxxxxxxxxxx>
Link: https://patch.msgid.link/20240820132234.2759926-3-shikemeng@xxxxxxxxxxxxxxx
Signed-off-by: Theodore Ts'o <tytso@xxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
fs/ext4/ialloc.c | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index 34def2892b838..26ebbb0388cc9 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -1053,12 +1053,13 @@ struct inode *__ext4_new_inode(handle_t *handle, struct inode *dir,
brelse(inode_bitmap_bh);
inode_bitmap_bh = ext4_read_inode_bitmap(sb, group);
/* Skip groups with suspicious inode tables */
- if (((!(sbi->s_mount_state & EXT4_FC_REPLAY))
- && EXT4_MB_GRP_IBITMAP_CORRUPT(grp)) ||
- IS_ERR(inode_bitmap_bh)) {
+ if (IS_ERR(inode_bitmap_bh)) {
inode_bitmap_bh = NULL;
goto next_group;
}
+ if (!(sbi->s_mount_state & EXT4_FC_REPLAY) &&
+ EXT4_MB_GRP_IBITMAP_CORRUPT(grp))
+ goto next_group;
repeat_in_this_group:
ret2 = find_inode_bit(sb, group, inode_bitmap_bh, &ino);
--
2.43.0
