Hey, Emacs, we're -*-Text-*- mode!

Known Bugs:

* proc files system
  The procfs entry causes kernel Oopses when the ftape module is
  removed while somebody accesses the /proc/ftape entry. Unluckily
  this can't be solved from inside ftape. The problem is that the
  current procfs implementation provides no means to determine whether
  the procfs entry is in use.

  Solved for kernel versions > 2.1.58

* llseek
  Trying to seek the device special files will not return an error
  although llseek isn't implemented. This is much the same with st.c
  and tpqic.c and ide-tape.c
  Should this be fixed? Maybe supply zftape with a dummy llseek() that
  simply returns -ESPIPE.

* formatting with Colorado Jumbo 350
  This doesn't seem to work. I don't think that the kernel driver
  needs to be changed; it should be possible to fix this in user
  space, modifying ftformat.

* You can't use the Iomega Ditto MAX drives with any other device than
  the first one, i.e. with /dev/nqft0, /dev/qft0 etc. This seems to be
  a peculiarity of the Ditto MAX, i.e. a hardware bug of either the
  Ditto Max drive or the Ditto EZ controller.

* The Ditto EZ 4Mbps controller will break the standard floppy disk
  controller when configured to use DMA channel 2 (the one used by the
  standard FDC). Apparently it doesn't help to try to disable the DMA
  gate of the Ditto EZ controller. I think this is a hardware bug of
  the EZ controller.

FIXME!

Claus-Justus Heine <heine@math1.rwth-aachen.de>

-------------------------------------------------------------------------------
 LocalWords:  procfs ftape llseek tpqic zftape ESPIPE ftformat Iomega
 LocalWords:  dev nqft qft FIXME Justus claus momo rwth aachen
 LocalWords:  LocalWords EZ Mbps FDC tjones estinc Linux SMP
 LocalWords:  sigpending fttools README heine