>That sort of flexibility is nice, however, I hope to see a fairly
>general-purpose interface for floppy disk parameter control; i.e.
>one that does not presume much knowledge about the capabilities
>of the particular hardware.  Asking this interface to allow those
>strange sector numbering schemes looks pretty low-level to me.
>
>I might suggest instead that strange formats like that might be
>supported by a different ioctl that makes no attempt to hide the
>attributes of the actual hardware (i.e. NEC 765 or Intel 82072)
>and gives you pretty much raw access to that hardware.
>
>Gordon

Along those lines, shouldn't there really be only two 'standard ioctls':

*	List the available FD formats either as an array of constants or an
array containing both formatting information and a 'cookie' to pass to
>the second ioctl
>*	Format the floppy given the constant or 'cookie'.
>
>This makes the 'fdformat' program completely machine-independent (and tiny)
>and keeps all the gory tables in the driver where they belong.  The previous
>proposal assumes that the strange formats are the common case as opposed to
>the extreme exception.
>
>later,
>Adam