changelog

linux 内核源代码

	  there would be no way to distinguish between normal attribute inodes
	  and index inodes.  The function to obtain an index inode is
	  ntfs_index_iget() and it uses the helper function
	  ntfs_read_locked_index_inode().  Note, we do not overload
	  ntfs_attr_iget() as indices consist of multiple attributes so using
	  ntfs_attr_iget() to obtain an index inode would be confusing.
	- Ensure that there is no overflow when doing page->index <<
	  PAGE_CACHE_SHIFT by casting page->index to s64 in fs/ntfs/aops.c.
	- Use atomic kmap instead of kmap() in fs/ntfs/aops.c::ntfs_read_page()
	  and ntfs_read_block().
	- Use case sensitive attribute lookups instead of case insensitive ones.
	- Lock all page cache pages belonging to mst protected attributes while
	  accessing them to ensure we never see corrupt data while the page is
	  under writeout.
	- Add framework for generic ntfs collation (fs/ntfs/collation.[hc]).
	  We have ntfs_is_collation_rule_supported() to check if the collation
	  rule you want to use is supported and ntfs_collation() which actually
	  collates two data items.  We currently only support COLLATION_BINARY
	  and COLLATION_NTOFS_ULONG but support for other collation rules will
	  be added as the need arises.
	- Add a new type, ntfs_index_context, to allow retrieval of an index
	  entry using the corresponding index key.  To get an index context,
	  use ntfs_index_ctx_get() and to release it, use ntfs_index_ctx_put().
	  This also adds a new slab cache for the index contexts.  To lookup a
	  key in an index inode, use ntfs_index_lookup().  After modifying an
	  index entry, call ntfs_index_entry_flush_dcache_page() followed by
	  ntfs_index_entry_mark_dirty() to ensure the changes are written out
	  to disk.  For details see fs/ntfs/index.[hc].  Note, at present, if
	  an index entry is in the index allocation attribute rather than the
	  index root attribute it will not be written out (you will get a
	  warning message about discarded changes instead).
	- Load the quota file ($Quota) and check if quota tracking is enabled
	  and if so, mark the quotas out of date.  This causes windows to
	  rescan the volume on boot and update all quota entries.
	- Add a set_page_dirty address space operation for ntfs_m[fs]t_aops.
	  It is simply set to __set_page_dirty_nobuffers() to make sure that
	  running set_page_dirty() on a page containing mft/ntfs records will
	  not affect the dirty state of the page buffers.
	- Add fs/ntfs/index.c::__ntfs_index_entry_mark_dirty() which sets all
	  buffers that are inside the ntfs record in the page dirty after which
	  it sets the page dirty.  This allows ->writepage to only write the
	  dirty index records rather than having to write all the records in
	  the page.  Modify fs/ntfs/index.h::ntfs_index_entry_mark_dirty() to
	  use this rather than __set_page_dirty_nobuffers().
	- Implement fs/ntfs/aops.c::ntfs_write_mst_block() which enables the
	  writing of page cache pages belonging to mst protected attributes
	  like the index allocation attribute in directory indices and other
	  indices like $Quota/$Q, etc.  This means that the quota is now marked
	  out of date on all volumes rather than only on ones where the quota
	  defaults entry is in the index root attribute of the $Quota/$Q index.

2.1.14 - Fix an NFSd caused deadlock reported by several users.

	- Modify fs/ntfs/ntfs_readdir() to copy the index root attribute value
	  to a buffer so that we can put the search context and unmap the mft
	  record before calling the filldir() callback.  We need to do this
	  because of NFSd which calls ->lookup() from its filldir callback()
	  and this causes NTFS to deadlock as ntfs_lookup() maps the mft record
	  of the directory and since ntfs_readdir() has got it mapped already
	  ntfs_lookup() deadlocks.

2.1.13 - Enable overwriting of resident files and housekeeping of system files.

	- Implement writing of mft records (fs/ntfs/mft.[hc]), which includes
	  keeping the mft mirror in sync with the mft when mirrored mft records
	  are written.  The functions are write_mft_record{,_nolock}().  The
	  implementation is quite rudimentary for now with lots of things not
	  implemented yet but I am not sure any of them can actually occur so
	  I will wait for people to hit each one and only then implement it.
	- Commit open system inodes at umount time.  This should make it
	  virtually impossible for sync_mft_mirror_umount() to ever be needed.
	- Implement ->write_inode (fs/ntfs/inode.c::ntfs_write_inode()) for the
	  ntfs super operations.  This gives us inode writing via the VFS inode
	  dirty code paths.  Note:  Access time updates are not implemented yet.
	- Implement fs/ntfs/mft.[hc]::{,__}mark_mft_record_dirty() and make
	  fs/ntfs/aops.c::ntfs_writepage() and ntfs_commit_write() use it, thus
	  finally enabling resident file overwrite!  (-8  This also includes a
	  placeholder for ->writepage (ntfs_mft_writepage()), which for now
	  just redirties the page and returns.  Also, at umount time, we for
	  now throw away all mft data page cache pages after the last call to
	  ntfs_commit_inode() in the hope that all inodes will have been
	  written out by then and hence no dirty (meta)data will be lost.  We
	  also check for this case and emit an error message telling the user
	  to run chkdsk.
	- Use set_page_writeback() and end_page_writeback() in the resident
	  attribute code path of fs/ntfs/aops.c::ntfs_writepage() otherwise
	  the radix-tree tag PAGECACHE_TAG_DIRTY remains set even though the
	  page is clean.
	- Implement ntfs_mft_writepage() so it now checks if any of the mft
	  records in the page are dirty and if so redirties the page and
	  returns.  Otherwise it just returns (after doing set_page_writeback(),
	  unlock_page(), end_page_writeback() or the radix-tree tag
	  PAGECACHE_TAG_DIRTY remains set even though the page is clean), thus
	  alowing the VM to do with the page as it pleases.  Also, at umount
	  time, now only throw away dirty mft (meta)data pages if dirty inodes
	  are present and ask the user to email us if they see this happening.
	- Add functions ntfs_{clear,set}_volume_flags(), to modify the volume
	  information flags (fs/ntfs/super.c).
	- Mark the volume dirty when (re)mounting read-write and mark it clean
	  when unmounting or remounting read-only.  If any volume errors are
	  found, the volume is left marked dirty to force chkdsk to run.
	- Add code to set the NT4 compatibility flag when (re)mounting
	  read-write for newer NTFS versions but leave it commented out for now
	  since we do not make any modifications that are NTFS 1.2 specific yet
	  and since setting this flag breaks Captive-NTFS which is not nice.
	  This code must be enabled once we start writing NTFS 1.2 specific
	  changes otherwise Windows NTFS driver might crash / cause corruption.

2.1.12 - Fix the second fix to the decompression engine and some cleanups.

	- Add a new address space operations struct, ntfs_mst_aops, for mst
	  protected attributes.  This is because the default ntfs_aops do not
	  make sense with mst protected data and were they to write anything to
	  such an attribute they would cause data corruption so we provide
	  ntfs_mst_aops which does not have any write related operations set.
	- Cleanup dirty ntfs inode handling (fs/ntfs/inode.[hc]) which also
	  includes an adapted ntfs_commit_inode() and an implementation of
	  ntfs_write_inode() which for now just cleans dirty inodes without
	  writing them (it does emit a warning that this is happening).
	- Undo the second decompression engine fix (see 2.1.9 release ChangeLog
	  entry) as it was only fixing a theoretical bug but at the same time
	  it badly broke the handling of sparse and uncompressed compression
	  blocks.

2.1.11 - Driver internal cleanups.

	- Only build logfile.o if building the driver with read-write support.
	- Really final white space cleanups.
	- Use generic_ffs() instead of ffs() in logfile.c which allows the
	  log_page_size variable to be optimized by gcc into a constant.
	- Rename uchar_t to ntfschar everywhere as uchar_t is unsigned 1-byte
	  char as defined by POSIX and as found on some systems.

2.1.10 - Force read-only (re)mounting of volumes with unsupported volume flags.

	- Finish off the white space cleanups (remove trailing spaces, etc).
	- Clean up ntfs_fill_super() and ntfs_read_inode_mount() by removing
	  the kludges around the first iget().  Instead of (re)setting ->s_op
	  we have the $MFT inode set up by explicit new_inode() / set ->i_ino /
	  insert_inode_hash() / call ntfs_read_inode_mount() directly.  This
	  kills the need for second super_operations and allows to return error
	  from ntfs_read_inode_mount() without resorting to ugly "poisoning"
	  tricks.  (Al Viro)
	- Force read-only (re)mounting if any of the following bits are set in
	  the volume information flags:
	  	VOLUME_IS_DIRTY, VOLUME_RESIZE_LOG_FILE,
		VOLUME_UPGRADE_ON_MOUNT, VOLUME_DELETE_USN_UNDERWAY,
		VOLUME_REPAIR_OBJECT_ID, VOLUME_MODIFIED_BY_CHKDSK
	  To make this easier we define VOLUME_MUST_MOUNT_RO_MASK with all the
	  above bits set so the test is made easy.

2.1.9 - Fix two bugs in decompression engine.

	- Fix a bug where we would not always detect that we have reached the
	  end of a compression block because we were ending at minus one byte
	  which is effectively the same as being at the end.  The fix is to
	  check whether the uncompressed buffer has been fully filled and if so
	  we assume we have reached the end of the compression block.  A big
	  thank you to Marcin Gibuła for the bug report, the assistance in
	  tracking down the bug and testing the fix.
	- Fix a possible bug where when a compressed read is truncated to the
	  end of the file, the offset inside the last page was not truncated.

2.1.8 - Handle $MFT mirror and $LogFile, improve time handling, and cleanups.

	- Use get_bh() instead of manual atomic_inc() in fs/ntfs/compress.c.
	- Modify fs/ntfs/time.c::ntfs2utc(), get_current_ntfs_time(), and
	  utc2ntfs() to work with struct timespec instead of time_t on the
	  Linux UTC time side thus preserving the full precision of the NTFS
	  time and only loosing up to 99 nano-seconds in the Linux UTC time.
	- Move fs/ntfs/time.c to fs/ntfs/time.h and make the time functions
	  static inline.
	- Remove unused ntfs_dirty_inode().
	- Cleanup super operations declaration in fs/ntfs/super.c.
	- Wrap flush_dcache_mft_record_page() in #ifdef NTFS_RW.
	- Add NInoTestSetFoo() and NInoTestClearFoo() macro magic to
	  fs/ntfs/inode.h and use it to declare NInoTest{Set,Clear}Dirty.
	- Move typedefs for ntfs_attr and test_t from fs/ntfs/inode.c to
	  fs/ntfs/inode.h so they can be used elsewhere.
	- Determine the mft mirror size as the number of mirrored mft records
	  and store it in ntfs_volume->mftmirr_size (fs/ntfs/super.c).
	- Load the mft mirror at mount time and compare the mft records stored
	  in it to the ones in the mft.  Force a read-only mount if the two do
	  not match (fs/ntfs/super.c).
	- Fix type casting related warnings on 64-bit architectures.  Thanks
	  to Meelis Roos for reporting them.
	- Move %L to %ll as %L is floating point and %ll is integer which is
	  what we want.
	- Read the journal ($LogFile) and determine if the volume has been
	  shutdown cleanly and force a read-only mount if not (fs/ntfs/super.c
	  and fs/ntfs/logfile.c).  This is a little bit of a crude check in
	  that we only look at the restart areas and not at the actual log
	  records so that there will be a very small number of cases where we
	  think that a volume is dirty when in fact it is clean.  This should
	  only affect volumes that have not been shutdown cleanly and did not
	  have any pending, non-check-pointed i/o.
	- If the $LogFile indicates a clean shutdown and a read-write (re)mount
	  is requested, empty $LogFile by overwriting it with 0xff bytes to
	  ensure that Windows cannot cause data corruption by replaying a stale
	  journal after Linux has written to the volume.

2.1.7 - Enable NFS exporting of mounted NTFS volumes.

	- Set i_generation in the VFS inode from the seq_no of the NTFS inode.
	- Make ntfs_lookup() NFS export safe, i.e. use d_splice_alias(), etc.
	- Implement ->get_dentry() in fs/ntfs/namei.c::ntfs_get_dentry() as the
	  default doesn't allow inode number 0 which is a valid inode on NTFS
	  and even if it did allow that it uses iget() instead of ntfs_iget()
	  which makes it useless for us.
	- Implement ->get_parent() in fs/ntfs/namei.c::ntfs_get_parent() as the
	  default just returns -EACCES which is not very useful.
	- Define export operations (->s_export_op) for NTFS (ntfs_export_ops)
	  and set them up in the super block at mount time (super.c) this
	  allows mounted NTFS volumes to be exported via NFS.
	- Add missing return -EOPNOTSUPP; in
	  fs/ntfs/aops.c::ntfs_commit_nonresident_write().
	- Enforce no atime and no dir atime updates at mount/remount time as
	  they are not implemented yet anyway.
	- Move a few assignments in fs/ntfs/attrib.c::load_attribute_list() to
	  after a NULL check.  Thanks to Dave Jones for pointing this out.

2.1.6 - Fix minor bug in handling of compressed directories.

	- Fix bug in handling of compressed directories.  A compressed
	  directory is not really compressed so when we set the ->i_blocks
	  field of a compressed directory inode we were setting it from the
	  non-existing field ni->itype.compressed.size which gave random
	  results...  For directories we now always use ni->allocated_size.

2.1.5 - Fix minor bug in attribute list attribute handling.