lkml.org 
[lkml]   [2000]   [Jul]   [24]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Direct access to hardware
Hi Andre!

On Mon, 24 Jul 2000, Andre Hedrick wrote:

> Why is this so wrong to do?
> Will you explain why you all would willfully want,
> : to endanger me personally?

No-one wants that.

> : void product warrenties?

And this is the point where the paths separate from each other. Of
course, driving a car against the tree breaks it. But that's obvious.
Special commands that may break the drive are not obvious, since drives
come without manual, and drives are not operated by users.

So. The user purchases a drive, he has a contract with the vendor,
that's it. No treaties with the manufacturer, in particular.

Now tell me: What has the OS got to do with that? Why do the
manufacturers try to pass the risk on to the OS or users?

Of course, they may void my warranty if I drop the drive. Why don't they
use either cryptography or just a red jumper in a non-standard (i. e.
other than 1/10 or 1/20 inch) pitch that needs to be placed in order to
do harmful actions such as format servo tracks, update firmware sectors,
seek off the disk or whatever.

> : dirty the reputation of this fine product?

The question is if Linux can reasonably and willingly accept that drive
manufacturers try to pass the card on to third parties that are not
involved in those drives.

I'm inclined to believe that the "use of vendor-specific, but otherwise
undefined commands in the ATA command set voids warranty" issues is not
holding in court, even though there are standards and committees that
will be asked.

Note this is not an opposition to having a sysctl, kernel secure level,
capability or whatever that enables IDE command filtering, I just
feel strong objections to the way vendors deal with the standard.

I don't think removing BIOS flash enable jumpers was a good idea either.

It saves $0.02 and may cause great damage that is to the risk of the
user.

This behaviour of vendors should be opposed to. It saves the $0.02 for
the jumper at the wrong end.

-
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/

\
 
 \ /
  Last update: 2005-03-22 13:57    [W:0.266 / U:0.288 seconds]
©2003-2014 Jasper Spaans. Advertise on this site