Skip to contentSkip to conversation

hfs: fix oops on mount with corrupted btree extent records

PATCH In review Jeff Mahoney 2 participants 3 messages Oct 13, 2009 HFS filesystem

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 hfs: fix oops on mount with corrupted btree extent records

A particular fsfuzzer run caused an hfs file system to crash on mount. This is due to a corrupted MDB extent record causing a miscalculation of HFS_I(inode)->first_blocks for the extent tree. If the extent records are zereod out, it won't trigger the first_blocks special case. Instead it falls through to the extent code which we're still in the middle of initializing.

This patch catches the 0 size extent records, reports the corruption, and fails the mount.

---
 fs/hfs/btree.c |    5 +++++
 1 file changed, 5 insertions(+)
fs/hfs/btree.c +5 -0
@@ -58,6 +58,11 @@ struct hfs_btree *hfs_btree_open(struct
 	}
 	unlock_new_inode(tree->inode);
 
+	if (!HFS_I(tree->inode)->first_blocks) {
+		printk(KERN_ERR "hfs: invalid btree extent records (0 size).\n");
+		goto free_inode;
+	}
+
 	mapping = tree->inode->i_mapping;
 	page = read_mapping_page(mapping, 0, NULL);
 	if (IS_ERR(page))
Signature · 2 lines
Jeff Mahoney
SUSE Labs
#1 lore raw
Quoted 9 lines · Jeff Mahoney

A particular fsfuzzer run caused an hfs file system to crash on mount. This is due to a corrupted MDB extent record causing a miscalculation of HFS_I(inode)->first_blocks for the extent tree. If the extent records are zereod out, it won't trigger the first_blocks special case. Instead it falls through to the extent code which we're still in the middle of initializing.

This patch catches the 0 size extent records, reports the corruption, and fails the mount.

Just for completeness - does the current fsck.hfs already know how to fix it, or at least heave stuff over the side so the filesystem is self-consistent again? Or will a user who hits this need a new fsck.hfs to get it back to mountable?

#2 lore raw
Quoted 8 lines · On 10/13/2009 12:12 AM, [email protected] wrote:
Quoted 7 lines · Jeff Mahoney

A particular fsfuzzer run caused an hfs file system to crash on mount. This is due to a corrupted MDB extent record causing a miscalculation of HFS_I(inode)->first_blocks for the extent tree. If the extent records are zereod out, it won't trigger the first_blocks special case. Instead it falls through to the extent code which we're still in the middle of initializing.

Quoted 6 lines
Quoted 2 lines

This patch catches the 0 size extent records, reports the corruption, and fails the mount.

Just for completeness - does the current fsck.hfs already know how to fix it, or at least heave stuff over the side so the filesystem is self-consistent again? Or will a user who hits this need a new fsck.hfs to get it back to mountable?

Sorry, I don't actually know. We don't ship a fsck.hfs so I haven't tested it.

-Jeff

Signature · 2 lines
Jeff Mahoney
SUSE Labs