*) Add toggle inode/block in use/bad flag.  Also a reread tables from
   disk command in case the user edits the tables.

*) Update the tty stuff.  Inodes aren't fully dumped (just info for 
   a minix inode is displayed -- but it will work on any fs).  Use
   Ted's debugfs if you want a tty driven ext2 hacker.
 
*) If you edit a block with inodes on it in block_mode, 
   then view it under inode mode, the inode will be incorrect
   until you view another inode and come back to this one.
   (This is only a problem with ext2fs -- see EXT2_read_inode).
   (Actually, it is a bigger problem with the other fs's which 
   store the inode tables in memory.)

*) Something is strange under xiafs:

   dd if=/dev/zero of=test.xiafs bs=1024 count=100 
   mkxfs test.xiafs 100
   mount -t xiafs -o loop test.xiafs /mnt 
   cp /usr/src/lde/* /mnt
   umount /mnt
   lde test.xiafs

   There's nothing there.  A few files make it and show up if you do a
   dir while test.xiafs is mounted, but I can't find the inode
   entries.  The disk is only 100 blocks long, but I don't see them.
   If I copy one file at a time, things are fine, it's there and lde
   is happy.  Is there some feature of xiafs that would cause it to
   relocate its inode table or overwrite it when the disk is full.

   After repeating the exercise, it works now! I did this 3 or for
   times in a row and had it fail on me.  Then tried it on minix and
   ext2 which worked fine,, but now xiafs is fine!!!!!

*) if I could suppress the hex/and or the ascii output, not just toggle
   which is active for edits

*) Make scrolling work a little better someday.  i.e. if we
   are at the end of a block and hit down arrow, go to the next block
   rather than stopping.  Display end of last block and start of
   this block on the same screen. 

*) Put up more super block info.  It was fine for Minix, but
   ext2 has a lot more "junk".

*) Throw in more inode info.  What are the inode.flags for
   ext2 anyways?
