TODO - pub/scm/linux/kernel/git/luto/util-linux-playground - Git at Google

 libblkid
 --------

  - add command line interface for blkid_probe_filter_types():

        # blkid -p -o udev --filter-type nofat

  - (?) we need to ignore cache and config files when the files are writable
    for non-root users and the library is linked with suid programs

  - add -<BE|LE> suffix to test images for native-endian filesystems (e.g. swap)
    and add support for such functionality to tests/ts/blkid/low-probe

  - do we really need to depend on libuuid? Maybe the default should be
    use in-tree code rather and link against libuuid -- especially when we
    need it for uuid_unparse() only.

  - consolidate "getsize" stuff (see getsize.c and lib/blkdev.c)

  - cleanup __attribute__ usage

  - add stripe size / alignment retrival support to libbkid for use
    with mkfs.  This would allow to deprecate libdisk from xfsprogs (1)
    in favour of a common libblkid.

    [ -- Christoph Hellwig, 16 Feb 2009 ]
    (1) http://git.kernel.org/?p=fs/xfs/xfsprogs-dev.git;a=tree;f=libdisk;

  - use fstatat() in blkid__scan_dir()

  - add values:

 	FSSIZE   -- filesystem size (klibc requirement)

 	SBOFFSET -- superblock offset; offset where SB was detected
                     (swapon requirement to check details from swap header)

         MAGIC    -- magic string
 	MAGICLEN -- length of magic string

 fdisk(s)
 --------

  * use off_t instead "long long"

  * fdisk/* refactoring

  * add GPT support

 misc
 ----

  * add a new command (unshare, clone, or so...) with all kinds of
    clone(2) options.
    http://thread.gmane.org/gmane.linux.utilities.util-linux-ng/2178

  * add to lib/blkdev.c code for /proc/partitions parsing -- unfortunate we
    duplicate this code in many places. The parser has to support unlimited
    size (or 4096 bytes) of partition name.

  * partx: copy sun.c, mac.c and dash.c from kpartx
    git://git.kernel.org/pub/scm/linux/storage/multipath-tools/.git

  * mkswap: use /proc/sys/kernel/random/uuid as a fallback solution for builds
    without libuuid

  * swapon -s  -- LABELs support

    Although mkswap has recently been -L option to create a label nothing appears to
    have been change to swapon to display said labels. (rh#430386)

  * use TZ=UTC for tests

  * add NLS and err.h stuff to schedutils (chrt.c, taskset.c)

  * add mllockall() and SCHED_FIFO to hwclock,
    see http://lkml.org/lkml/2008/10/12/132

  * use rpmatch() for all Y/N questions

  * mount -a -- reorder fstab entries by paths before mount (just idea only)

  * mount -a  (just idea only)

         Date:   Sun, 3 Jun 2007 18:04:24 +0300 (MET DST)
         From: Szabolcs Szakacsits <szaka@sienet.hu>

 	In the past the right record order could be figured out easily by just
 	checking out fstab (if one knew what to look for) but considering the
 	fastly increasing number of user space file systems and their usage, with
 	their path, library, etc dependencies, it's getting trickier and is a black
 	magic for most users because they simply expect drives to be mounted
 	independently of their order in fstab.

 	One typical, wrongly edited fstab example is:

 	  /dev/hda2 /             ext3      defaults  1 1
 	  /dev/hda1 /mnt/windows  ntfs-3g   defaults  0 0
 	  /dev/hda3 /usr          ext3      defaults  0 0

 	The events:

 	  mount  ->  /sbin/mount.ntfs-3g  ->
 		 ->  resolves to <path1>/ntfs-3g via a symlink ->
 		 ->  ntfs-3g requires at least <path2>/libfuse*

 	There are many potential solutions. For example installing everything on
 	the root file system which may be needed for successful mount. But this
 	is not always feasible or practical since we could end up putting almost
 	everything on the root file system in the end.

 	Another idea is an improved mount strategy:

 	  do {
 		try to mount all unmounted entries
 	  } while (not all mounted && at least one new was successfully mounted)

  * agetty: use nl_langinfo() for days and months rather than hardcoded
    English names.

  * rewrite ipcs to use /proc/sys/kernel rather than unreliable syscalls
    (there are problems with 32bit userspace on 64bit kernel)

  * minix v3

 	From: Matthias Koenig <mkoenig@suse.de>
 	Date:   Tue, 25 Sep 2007 12:00:01 +0200

 	It seems that the kernel has support for minix fs v3
 	(though I have not tried it, just inspected some code when
 	trying to find a mkfs.minix issue).
 	It might be worth a thought implementing v3 support
 	(though I am not really sure how much people us minix fs ;-)
 	This might require some major code cleanup in mkfs.minix.

  * rtcwake does not support wake from S5/off
    http://bugzilla.redhat.com/show_bug.cgi?id=449115

  * add SELinux security contexts support to the 'ipcs' utility
    http://bugzilla.redhat.com/show_bug.cgi?id=225342

    Would be great to list the current system IPC Objects with their respective
    security labels (where allowed) with something like 'ipcs -Z' - following the
    way other tools reports those.
	libblkid
	--------

	- add command line interface for blkid_probe_filter_types():

	# blkid -p -o udev --filter-type nofat

	- (?) we need to ignore cache and config files when the files are writable
	for non-root users and the library is linked with suid programs

	- add -<BE\|LE> suffix to test images for native-endian filesystems (e.g. swap)
	and add support for such functionality to tests/ts/blkid/low-probe

	- do we really need to depend on libuuid? Maybe the default should be
	use in-tree code rather and link against libuuid -- especially when we
	need it for uuid_unparse() only.

	- consolidate "getsize" stuff (see getsize.c and lib/blkdev.c)

	- cleanup __attribute__ usage

	- add stripe size / alignment retrival support to libbkid for use
	with mkfs. This would allow to deprecate libdisk from xfsprogs (1)
	in favour of a common libblkid.

	[ -- Christoph Hellwig, 16 Feb 2009 ]
	(1) http://git.kernel.org/?p=fs/xfs/xfsprogs-dev.git;a=tree;f=libdisk;

	- use fstatat() in blkid__scan_dir()

	- add values:

	FSSIZE -- filesystem size (klibc requirement)

	SBOFFSET -- superblock offset; offset where SB was detected
	(swapon requirement to check details from swap header)

	MAGIC -- magic string
	MAGICLEN -- length of magic string

	fdisk(s)
	--------

	* use off_t instead "long long"

	* fdisk/* refactoring

	* add GPT support

	misc
	----

	* add a new command (unshare, clone, or so...) with all kinds of
	clone(2) options.
	http://thread.gmane.org/gmane.linux.utilities.util-linux-ng/2178

	* add to lib/blkdev.c code for /proc/partitions parsing -- unfortunate we
	duplicate this code in many places. The parser has to support unlimited
	size (or 4096 bytes) of partition name.

	* partx: copy sun.c, mac.c and dash.c from kpartx
	git://git.kernel.org/pub/scm/linux/storage/multipath-tools/.git

	* mkswap: use /proc/sys/kernel/random/uuid as a fallback solution for builds
	without libuuid

	* swapon -s -- LABELs support

	Although mkswap has recently been -L option to create a label nothing appears to
	have been change to swapon to display said labels. (rh#430386)

	* use TZ=UTC for tests

	* add NLS and err.h stuff to schedutils (chrt.c, taskset.c)

	* add mllockall() and SCHED_FIFO to hwclock,
	see http://lkml.org/lkml/2008/10/12/132

	* use rpmatch() for all Y/N questions

	* mount -a -- reorder fstab entries by paths before mount (just idea only)

	* mount -a (just idea only)

	Date: Sun, 3 Jun 2007 18:04:24 +0300 (MET DST)
	From: Szabolcs Szakacsits <szaka@sienet.hu>

	In the past the right record order could be figured out easily by just
	checking out fstab (if one knew what to look for) but considering the
	fastly increasing number of user space file systems and their usage, with
	their path, library, etc dependencies, it's getting trickier and is a black
	magic for most users because they simply expect drives to be mounted
	independently of their order in fstab.

	One typical, wrongly edited fstab example is:

	/dev/hda2 / ext3 defaults 1 1
	/dev/hda1 /mnt/windows ntfs-3g defaults 0 0
	/dev/hda3 /usr ext3 defaults 0 0

	The events:

	mount -> /sbin/mount.ntfs-3g ->
	-> resolves to <path1>/ntfs-3g via a symlink ->
	-> ntfs-3g requires at least <path2>/libfuse*

	There are many potential solutions. For example installing everything on
	the root file system which may be needed for successful mount. But this
	is not always feasible or practical since we could end up putting almost
	everything on the root file system in the end.

	Another idea is an improved mount strategy:

	do {
	try to mount all unmounted entries
	} while (not all mounted && at least one new was successfully mounted)

	* agetty: use nl_langinfo() for days and months rather than hardcoded
	English names.

	* rewrite ipcs to use /proc/sys/kernel rather than unreliable syscalls
	(there are problems with 32bit userspace on 64bit kernel)

	* minix v3

	From: Matthias Koenig <mkoenig@suse.de>
	Date: Tue, 25 Sep 2007 12:00:01 +0200

	It seems that the kernel has support for minix fs v3
	(though I have not tried it, just inspected some code when
	trying to find a mkfs.minix issue).
	It might be worth a thought implementing v3 support
	(though I am not really sure how much people us minix fs ;-)
	This might require some major code cleanup in mkfs.minix.

	* rtcwake does not support wake from S5/off
	http://bugzilla.redhat.com/show_bug.cgi?id=449115

	* add SELinux security contexts support to the 'ipcs' utility
	http://bugzilla.redhat.com/show_bug.cgi?id=225342

	Would be great to list the current system IPC Objects with their respective
	security labels (where allowed) with something like 'ipcs -Z' - following the
	way other tools reports those.