Re: [PATCH 2/3] block: fail SCSI passthrough ioctls on partition devices

[Daniel Barkalow <barkalow@iabervon.org>]

On Fri, 23 Dec 2011, Willy Tarreau wrote:

> On Thu, Dec 22, 2011 at 04:07:46PM -0800, Linus Torvalds wrote:
> > For example, I just traced it, and "eject /dev/sdb1" does a CDROMEJECT
> > ioctl when used as the root user. I haven't tested the patch, but just
> > reading it, I'd expect it to break that.
> > 
> > And that's the *natural* way to eject a mounted device. Look at the
> > USB memory sticks you have. They are almost all partitioned to have
> > one partition, and that one partition doesn't cover the whole device.
> > And it's that one partition you use to interact with it - it's what
> > you mount, and what you eject.
> 
> Call me dumb, but why would someone use "eject" on a non-physically
> ejectable device such as a memory stick ? I use it on CDs, I've used
> it on Sun floppy drives, but USB memory stick ??? After the umount,
> I just have to pull it from the plug and that's all. I don't catch
> what an eject command can bring me on top of that :-/

I use "eject" on my (old) ipod in order to get it to stop telling me not 
to unplug it. The device doesn't actually have any problems if it just 
gets yanked while it's neither mounted nor ejected, but it acts unhappy 
through its UI, since it doesn't know the computer's state. (And I pass 
"eject" the mountpoint, because that's short and tab-completes and 
"eject" translates the mountpoint into a device node with fstab so it 
works.)

	-Daniel
*This .sig left intentionally blank*