From: rohrer@fncrd6.fnal.gov (Keith Rohrer) Subject: Re: Diskette Changeline (was: Need Help: Mtools) Date: 21 Aug 1992 14:29:26 GMT
In article <1992Aug21.023532.28214@athena.mit.edu> tytso@ATHENA.MIT.EDU (Theodore Ts'o) writes:
> From: jhuang@Chaos.mcs.kent.edu (Jing Huang)
> Date: Thu, 20 Aug 1992 20:39:10 GMT
[Possible typical diskette changeline problem deleted]
To be certain, you ought to try this under MS-DOG (or DR-DOS, take your pick,
even OS/2 ought to do it; if you're considering testing this with the Windows
NT pre-release developer's kit, Begone Evil One, though that might work too...)
just to make sure; I've got some nasty hard drive problems that don't
show up under MS- or DR- DOS (or OS/2 2.0) at all...
>It sounds like your floppy drive doesn't have its "diskette change line"
>working correctly --- apparently that's a common problem. The meaning
>of that line (I think it's pin 34 on the ribbon connector) changed from
If I remember my lore correctly, not so much changed as became useful.
>XT floppy dirves to AT floppy drives. On AT systems, that line is
>supposed to signal to computer that someone has opened the driver door
>(on a 5" disk) or ejected the floppy (on a 3" disk), so the computer
>should flush its disk buffers.
>On some diskette drives, there is a jumper you can set to enable/disable
>that line. If you have the documentation for your drive, you might look
This line is (1) not present on most, if not all, 360K drives, and (2)
usually selectable between Diskette Changeline and Standard Ready. Do
*not* specify standard ready! That signals when there's a disk loaded
properly in the drive, and the computer interprets that as "diskette
unchanged"....
>for such a jumper, or see if there's an obvious jumper whose trace goes
>to that pin on the edge connector.
>Even if you can fix your hardware, it sounds like, we need an ioctl in
>the floppy disk driver to artificially simulate a disk change event, for
>people who can't. It wouldn't be hard to do, and it would probably be a
>good exercise for somone who wants to start learning how a Linux device
>driver works. (Hint: look at the variable changed_floppy.)
Don't we need to have diskette changeline support disablable already, or has
nobody tested Linux with a 360K drive?
>If no one gets around to posting a patch in a week or two, though, I'll
>make the changes when I can find some time.
> - Ted
Keith
(rohrer@fncrd0.fnal.gov)
-- Disclaimer: None of Grinnell College, URA, Fermilab, and any other affiliated persons or orginizations have licensed my ideas or opinions, and thus are not entitled to any which may appear above.