NTFS: Fix a bug in fs/ntfs/inode.c::ntfs_read_locked_index_inode() where we
forgot to update a temporary variable so loading index inodes which have an index allocation attribute failed. Signed-off-by: Anton Altaparmakov <aia21@cantab.net>
This commit is contained in:
@@ -651,10 +651,7 @@ err_out:
|
||||
* fs/ntfs/aops.c::mark_ntfs_record_dirty().
|
||||
*
|
||||
* On success, clean the mft record and return 0. On error, leave the mft
|
||||
* record dirty and return -errno. The caller should call make_bad_inode() on
|
||||
* the base inode to ensure no more access happens to this inode. We do not do
|
||||
* it here as the caller may want to finish writing other extent mft records
|
||||
* first to minimize on-disk metadata inconsistencies.
|
||||
* record dirty and return -errno.
|
||||
*
|
||||
* NOTE: We always perform synchronous i/o and ignore the @sync parameter.
|
||||
* However, if the mft record has a counterpart in the mft mirror and @sync is
|
||||
|
Reference in New Issue
Block a user