>
>Good. So we need a special kind of mount, that postpone the reading of
>the superblock etc. until some process tries to access the fs.
>
> > Polling the disk every x secs is just stupid, we have no need to mount
>until
> > he tries to access it anyway.
>Agreed.
> >
> > The only good way to handle removed it is like the macitoshes do, make
>you
> > click a button or type a command to eject it.
>
>Keep the light on as long as there are outstanding requests. When there
>are
>none, wait a little and if nothing happens sync the floppy and turn off
>the light. No fs damage if the user removes it at that point.
>
>At the next access, verify that it is the same floppy. If not, remount
>it
>(and invalidate any open file handles for the old floppy. There will
>usually be none.)
>
>If this is somehow hard to do, keep the light on as long as
>the fs is busy, and auto-umount after a timeout when it
>isn't.
>
yeah, i was thinking something like that. Maybe only mount it ro until
something opens a file in rw mode, then do a remount real quick before we
open the file. when there's no file open in rw mode, might as well sync and
remount ro, and umount completely if the user does nothing to it after so
long, depending on how expensive remounts and umoonts are.
also keep in mind the user could take out a vfat floppy and put in a hfs
floppy or a minix floppy. or any other fs.
oh yeah, and one more thing against that copying an image of the floppy to
the harddisk, or even ram, is that some people have old computers with
almost full hard drives, and save stuff to floppies because their harddisk
is full. This person might also only have 8 megs of ram or something. he
doesn't have room on his harddisk for a 1.44 meg floppy, and doesn't need
you taking up that much ram either.
--Tim
________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu Jun 15 2000 - 21:00:36 EST