> Not sure I like that, TBH. We shouldn't go filling up the user's disk with
> caches unless we really need to - I'd much rather use the free swap space
> for this.
>
Your approach seems better to me for floppies, I agree. Unfortunately, if you extend
this to CDROM (650mb...) you can easily blow up the machine by running out of VM.
Many of the machines I run here do not have more than 500mb or so of swap space, and
use very little of that...
> > 2) By putting the floppy disk drive handling code in user space, we can steal
> > ^H^H^H^H^H borrow large pieces of the mtools libs and avoid reinventing the
> > wheel. At least that code is stable.
>
> I suspect the kernel's floppy handling code is stable too ;-)
>
This was not meant as a knock to the kernel FDD code. (I have never had any trouble
with it.) I was simply pointing out that the mtools libs would probably require less
rewrite than grabbing pieces out of the msdos filesystem code in the kernel. I may
be wrong. If so, please correct me.
>
> All we really need to do is add a couple of hooks here and there. Get the
> floppy driver caching as aggressively as possible into VM, and add some
> feedback mechanism to alert a UI daemon to tell the user off if they pull
> the disk out while Linux is still using it, and to umount the disk if it's
> been removed and we aren't still syncing it.
>
> James.
I was thinking of hooks into the vfs layer to work with a user-space deamon. This
way, we can put the necessary code into userland and if it's done carefully a crash
of the userland daemod would be recoverable. This is just conceptual, as I am not
terribly familiar with the vfs code.
--
Jacques
--
The packet goes out the card, into the copper,
out the router, onto the fiber, across the world,
through the copper......
NOTHING BUT NET!
-
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 - 21:00:32 EST