[Debburn-devel] Re: Bug#413960: wodim: fails to burn with dev='0, 0, 0' but burns with dev='/dev/sr0'

[Eduard Bloch]

#include <hallo.h>
* Vladimir Nadvornik [Wed, Mar 14 2007, 10:53:01AM]:
> On Wednesday 14 March 2007 01:58, Eduard Bloch wrote:
> >
> > This just does not add up. The only difference is IMO the driver used by
> > /dev/sgX vs. /dev/sr0 (sg vs. sg_io). sg_io works pretty well, while sg
> > does weird things: after few data writes it starts returning false sense
> > codes indicating that the drive is busy (04/08, long drive in progress).
> > So wodim keeps retrying, which is the correct way to do in order to not
> > come to late. And after few dozens of retry it returns weird stuff and
> > the process breaks.
> >
> 
> This seems to be the same problem as here:
> https://bugzilla.novell.com/show_bug.cgi?id=226019
> 
> There is no shared lock between sg and sr, some other program (HAL)
> accesses the device in parallel and LG burners don't like it.

OMG. This is naturally a kernel issue. It allows access through two
interfaces simultaneosly, both interfaces are known and in wide use.
And it is pretty naive to think that all people and all software will
start using /dev/srX and only srX overnight just become somebody said
so, and it is not very nice to abandon those users to their faith.

And following Lars' words consequently means that the sg driver better
should not be able to load or allocate the same devices. Not the most
convinient but a safe solution. Sharing resources requires
synchronization addons in all conflicting spaces to work properly.
That's life. And currently the kernel is broken in this respect.

Eduard.
-- 
* Joey darf www.debian.org kaputt machen?  Hmm, interessantes Privileg :)
<h01ger> .oO( alle duerfen planet. und joey halt auch www. :)