2005-10-15  Markus Klotzbuecher  <mk@localhost.localdomain>

	* Bugfix for a serious memory leak in mini_fo_follow_link.

2005-09-21  Markus Klotzbuecher  <mk@mary>

	* new release 0.6.1

	* fix of a compiler warning due to changes in 2.6.13

2005-09-21  Klaus Wenninger  <klaus.wenninger@siemens.com>

	* file.c: readdir: fix for a bug that caused directory entries
          to show up twice when using storage filesystems such as
          minixfs or pramfs.

2005-06-30  Eric Lammerts <eric@lammerts.org>

	* fix for an oops when overwriting a binary thats beeing
          executed.

2005-06-09    <mk@mary>

	* Renamed overlay to mini_fo-overlay.

	* Added mini_fo-merge script to allow merging of storage and base
	after making modifications.

2005-05-22  root  <mk@mary>

	* Added overlay script that allows to easily mount mini_fo ontop
	of a given base directory

2005-05-10    <mk@mary>

	* inode.c: xattr functions return -EOPNOSUPP instead of
          -ENOSUPP, what confuses "ls -l"

	* Changed license from LGPL to GPL.

2005-05-08  root  <mk@mary>

	* Makefile: clean it up and added make install and make
          uninstall.
	
2005-05-06    <mk@mary>

	* merged devel branch back to main. [v0-6-0-pre3]

	* removed unused files print.c and fist_ioctl. [devel-0-0-18]

	* ioctl: removed fist_ioctl stuff, that is not needed for
          now.

2005-05-03    <mk@mary>

	* file.c: simplified mini_fo_open and mini_fo_setattr using
          new state changing functions. [devel-0-0-17]

	* inode.c: Fixed getattr state bug (see below) in 2.4 function
          mini_fo_inode revalidate.

	* inode.c: found an other bug in mini_fo_getattr. States are not
	  reliable in this function, as a file can be opened, unlinked and
	  the getattr function called. This results in a deleted dentry
	  with an inode. Fix is to ignore states and simply use the inode
	  available.

2005-04-29    <mk@mary>

	* file.c: Bugfix and cleanup in fasync and fsync. [devel-0-0-16]

	* file.c: do not use mini_fo_lock so the generic version is
          used (I guess).

	* inode.c: getattr, never call getattr on lower files, as this
          will cause the inum to change.

	* inode.c: rename_reg_file renamed to rename_nondir, as it
          doesn't matter as long it't not a dir. Removed all
          rename_dev_file etc. 	  

	* tagged as devel-0-0-15

	* inode.c: added support for chosing support for extended
          attrs at compile time by XATTR define in mini_fo.h .

	* inode.c: fixed mini_fo_getattr to use mini_fo inode and not
          lower again, what avoids inode number changes that confused
          rm again. This is the proper solution.

2005-04-24    <mk@mary>

	* all files: updated Copyright notive to 2005. [devel-0-0-14]

	* inode.c: fixed mini_fo_getattr to not change the inode
          number, even if lower files change.

	* super.c: fixed a bug that caused deleted base file to show
          up suddenly after some time, or after creating a special
          file. The problem was that after some time or after special
          file creating sync_sb_inodes is called by the vfs, that
          called our mini_fo_put_inode. There was (wrongly) called
          __meta_put_lists, that nuked the lists, although the inode
          was going to continue its life. Moving __meta_put_lists to
          mini_fo_clear_inode, where an inode is really destroyed,
          solved the problem.


2005-04-23    <mk@mary>

	* state.c, aux.c: more cleaning up and
          simplifications. [devel-0-0-13] 

	* inode.c: implemented mini_fo_getattr, that was required for
          2.6 because inode_revalidate has been remove there, and the
	  old "du" bug returned.


2005-04-20    <mk@mary>

	* aux.c: get_neg_sto_dentry(): allow to be called for dentries
          in state UNMODIFIED, NON_EXISTANT _and_ DELETED.

2005-04-19    <mk@mary>

	* Fixed a bug under 2.6 that caused files deleted via mini_fo
          not to be deleted properly and therefore the fs filled up
          untill no memory was left. [devel-0-0-12]

	* Added basic hard link support. This means that creating
          hardlinks will work, but existing ones will be treated as
          individual files. [devel-0-0-11]

2005-04-17    <mk@mary>

	* Bugfixes

2005-04-13  root  <mk@mary>

	* Added file state.c for the state transition
          functions. Doesn't work very well yet, though...

2005-04-12    <mk@mary>

	* Porting to 2.6 started, which is easier than expected, also
          due to Olivier previous work.

2005-04-08    <mk@mary>

	* Fixed the bug that caused du to return invalid sizes of
          directory trees. The problem was that
          mini_fo_inode_revalidate didn't always copy the attributes
          from the base inode properly.

2005-04-01  Markus Klotzbuecher  <mk@chasey>

	* Merged devel branch back to main trunk and updated the
          RELEASE notes. This will be 0-6-0-pre1.

2005-03-31  Markus Klotzbuecher  <mk@chasey>

	* Fixed some bugs in rename_reg_file, that only showed up in
          the kernel compile test. Kernel compiles cleanly ontop of
          mini_fo, now also make mrproper etc. work. Seems pretty stable.

2005-03-28  Markus Klotzbuecher  <mk@chasey>

	* Many, many directory renaming bugfixes and a lot of other
          cleanup. Dir renaming seems to work relatively stable.

2005-03-22  Markus Klotzbuecher  <mk@chasey>

	* Finished implementing lightweight directory renaming. Some
          basic testing indicates it works fine.
	  Next is to implement testcases for the testsuite and confirm
          everything is really working ok.

2005-03-18  Markus Klotzbuecher  <mk@chasey>

	* Finished implementing meta.c stuff required for directory
          renaming.

2005-03-17  Markus Klotzbuecher  <mk@chasey>

	* Fixed all compile warnings + an extremly old bug that
          somehow crept in while reworking the wol stuff to the META
          system. Turning on -Werror again... :-)

	* Fixed some bugs in the new rename_reg_file function.

	* Rewrote mini_fo rename and split it into several
          subfunctions, that handle the different types
          seperately. Rewrote the regular file function aswell, as it
          was implemented somewhat inefficient. 

2005-03-16  Markus Klotzbuecher  <mk@chasey>

	* Implemented new META subsystem, removed old WOL stuff in favor 
	  if it.

	* After some basic testing everything seems ok...

2005-03-11  Markus Klotzbuecher  <mk@chasey>

	* Renaming a non regular file caused trouble because I always
	  tried to copy the contents. Now I only do this for regular
	  files. mini_fo_rename still isn't implemented properly, renaming
	  of device files, symlinks etc. results in a empty regular file
	  instead of the proper type.
	
	* Directory renaming suddenly works! What a surprise! I guess
          this is because renaming is implemented as making a copy and
          removing the original. Still this might not work
          everywhere...

2005-03-09  Markus Klotzbuecher  <mk@chasey>

	* Bugfix, when a mini_fo directory that exists in storage
  	  (state: MODIFIED, CREATED and DEL_REWRITTEN) is deleted, a
  	  possibly existing WOL file contained in it needs to be
  	  deleted too.

	* Starting cleanup: defined state names in order to get rid of
          the state numbers.

2005-03-08  Markus Klotzbuecher  <mk@chasey>
	
	* Makefile fix, fist_ioctl was built against wrong sources if ARCH=um

	* Fixed a bug in dentry.c, mini_fo_d_hash. In state 4 =
          DEL_REWRITTEN the hash was calculated from the base dentry,
          which was wrong and and caused assertions in
          __mini_fo_hidden_dentry to fail.

2005-02-21    <mk@mary>

	* Implemented directory deleting (inode.c)

	* main.c: made mini_fo_parse_options a little more robust.

2004-12-22    <mk@mary>

	* Makefile cleanup and uml stuff, removed unneccessary files

	* Created a new and hopefully more informative README

	* CHANGELOG: created a new CHANGELOG and added old entries reversely


2004-10-24 Gleb Natapov <gleb@nbase.co.il>

	* Fix: owner and group where not correctly copied from base to
          storage. 


2004-10-05 Gleb Natapov <gleb@nbase.co.il>

	* Implementation of fsync, fasync and lock mini_fo functions.
	

2004-09-29 Bob Lee <bob@pantasys.com>

	* Fix of a serious pointer bug
	

2004-09-28 Gleb Natapov <gleb@nbase.co.il>

	* Implementation of mini_fo_mknod and mini_fo_rename, support
          for device files.
	
