==================================
Release Notes for Ext2Fsd V0.61
==================================

Features implemented or bugfix since V0.53:

    1, Ext4 extent writing support (developed by ngkaho1234 @ gmail.com).
       See his repository for more details:
       http://github.com/ngkaho1234/Ext3Fsd.git
    2, Size manipulation and i/o path revised for support of both indirect and
       extent-based mappings
    3, 16T volume supported (tested with 4T volume / 4K block-size)
       Version 0.53 and previous versions only support up to 2T volume
       ticket: https://sourceforge.net/p/ext2fsd/bugs/134/
    4, Deletion acceleration: truncation optimized for indirect-based files
    5, Size-expanding disabled in IRP_MJ_CREATE for better performance.
       Windows sometimes would truncate file size back to 0
    6, Support Volume exclusive-open for e2fsprogs utils. Any attempts
       will be denied while EXT4 volume is being checked (e2fsck).
    7, FIXME:  Possible zero-content returned (requesting data from file tail)
    8, FIXME:  File ByteOffset updated for FO_SYNCHRONOUS_IO open (async i/o)
       ticket: https://sourceforge.net/p/ext2fsd/bugs/133/
    9, FIXME:  don't do byte or range round up for paging i/o requests
       ticket: https://sourceforge.net/p/ext2fsd/bugs/129/
   10, FIXME:  occasional block bitmap difference issue, cased by checksum
               caculation in uninit_bg group initializing
   11, FIXME: Ext4 unwritten-extent supported to avoid possible data corruption
              MSB of ee_len field of unwritten extent is used as only a flag. It
              was mistakenly treated as extent length (block numbers) in V0.60
   12, FIXME: Manually clear file to zero for indirect-block file that is newly
              created but not yet written
   13, Renaming supported: with names only diffs in lower or upper cases
   14, I/o path revised to correctly handle ValidDataLength & unwritten holes
   15, e2fsprogs utils for Windows (e2fsck, mke2fs, tune2fs, debugfs) included


Supported Ext3/4 features by 0.61:

    1, flexible inode size: > 128 bytes, up to block size
    2, dir_index:    htree directory index
    3, filetype:     extra file mode in dentry
    4, large_file:   > 4G files supported
    5, sparse_super: super block backup in group descriptor
    6, uninit_bg:    fast fsck and group checksum
    7, extent:       full support with extending and shrinking.
    8, journal:      only support replay for internal journal
    9, flex_bg:      first flexible metadata group

Unsupported Ext3/4 features:

    1, journal: log-based operations, external journal
    2, EA (extended attributes), ACL support
    3, symlink creation

Features to be implemented in next release:

    1, Reparse point support (hardlink, symbolic link etc)
    2, mounted as given user: privilege checking with given uid & gid

WARNINGS:
 
    The driver may crash your system and ruin your data unexpectedly,
    since there might be software conflicts and I could only test it
    on some of the popular platforms. You should use it with care and
    use it at your own risk!

Matt <mattwu@163.com>
http://www.ext2fsd.com