Subject: Re: Floppy - weird behaviour?
To: David Laight <david@l8s.co.uk>
From: hypno <hypno@evilworks.com>
List: port-i386
Date: 02/05/2003 23:17:51
On Wed, 5 Feb 2003, David Laight wrote:

> > ! This is probably something David Laight could do as he is
> > ! fixing the boot blocks up.
> >
> > Yes, correct.
> >
>
> It is a bit problematical though.  You don't want to stop the motor
> if the drive is going to be used again anytime soon.  Otherwise
> you have to wait (a bit) for spin up.
>
> All the early disk accesses are also done via bios calls.  Just
> writing to the floppy device registers could confuse the system
> BIOS into thinking the motor was running when it isn't.
> It isn't until the kernel initialises its own floppy driver
> that any netbsd code that knows how to drive a floppy is loaded.
>
> There is probably a bios call to stop the motor, but I don't
> know whether it will restart automatically.  My system has too
> many fans in it to let you hear the floppy going around.

bios int 0x15, function 0x1 ?

or maybe,

bios int 0x13, function 0x16 ?

> Also the boot code will be passed device 0x0 when booting a CD
> - the BIOS makes it look like a floppy (actually you read a
> floppy image from one of the files on the CD).  If there are
> bios calls to read the real CD I don't know what they are...
> (and you would need the rockridge code for a unix system).

as far as i know, there are no such bios function(s) (neither do my 80386
hardware reference manual know any).

>
> 	David
>
> --
> David Laight: david@l8s.co.uk
>

bios seems to save info about floppy at 0x40:0x3e - 0x40:0x42, maybe it
could be useful?

cheers,
// hypno