src/patches/suse-2.6.27.39/patches.fixes/hfs-corrupt-extent-tree-mount-oops-fix

   1 From: Jeff Mahoney <jeffm@suse.com>
   2 Subject: [PATCH] hfs: fix oops on mount with corrupted btree extent records
   3 References: bnc#476582 bko#13083
   4
   5  A particular fsfuzzer run caused an hfs file system to crash on mount. This
   6  is due to a corrupted MDB extent record causing a miscalculation of
   7  HFS_I(inode)->first_blocks for the extent tree. If the extent records
   8  are zereod out, then it won't trigger the first_blocks special case and
   9  instead falls through to the extent code, which we're in the middle
  10  of initializing.
  11
  12  This patch catches the 0 size extent records, reports the corruption,
  13  and fails the mount.
  14
  15 Reported-by: Ramon de Carvalho Valle <rcvalle@linux.vnet.ibm.com>
  16 Signed-off-by: Jeff Mahoney <jeffm@suse.com>
  17 ---
  18  fs/hfs/btree.c |    5 +++++
  19  1 file changed, 5 insertions(+)
  20
  21 --- a/fs/hfs/btree.c
  22 +++ b/fs/hfs/btree.c
  23 @@ -58,6 +58,11 @@ struct hfs_btree *hfs_btree_open(struct
  24         }
  25         unlock_new_inode(tree->inode);
  26
  27 +       if (!HFS_I(tree->inode)->first_blocks) {
  28 +               printk(KERN_ERR "hfs: invalid btree extent records (0 size).\n");
  29 +               goto free_inode;
  30 +       }
  31 +
  32         mapping = tree->inode->i_mapping;
  33         page = read_mapping_page(mapping, 0, NULL);
  34         if (IS_ERR(page))