lkml.org 
[lkml]   [2004]   [Aug]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: PATCH: cdrecord: avoiding scsi device numbering for ide devices
Date
On Thursday 19 August 2004 19:32, Horst von Brand wrote:
> Joerg Schilling <schilling@fokus.fraunhofer.de> said:
> > Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl> said:
> > >> As a security fix it was sufficiently important that it had to be
> > >> done.
> > >
> > >IMO work-rounding this in kernel is a bad idea and could break a lot of
> > >existing apps (some you even don't know about). Much better way to deal
> > >with this is to create library for handling I/O commands submission and
> > >gradually teach user-space apps to use it.
>
> Nonsense (as I just said in another message).

Please read Mark Lord's mail and my reply.

> > This is exactly what libscg is for......
> > libscg already includes similar support for Solaris 9 & Solaris 10.
>
> OK, their problem.
>
> > Cdrtools is is code freeze state. This is why I say the best idea is to
> > remove this interface change from the current Linux kernel and wait until
> > there will be new cdrtools alpha for 2.02 releases. These alpha could get
> > support for uid switching. If Linux then would again switch the changes
> > on, it makes sense.
>
> Sorry, you have absolutely no say in the development of the kernel
> here. You fix your broken app, code freeze or no code freeze. Or let others
> that fix it alone.
>
> > BTW: it makes absolutely no sense to have a list of "safe" commands in
> > the kernel as the kernel simply cannot know which SCSI commands are
> > "safe" and which not.
>
> "Normal" read/write commands are safe, others are off-limits unless you
> have the required capability (one which allows you to set the device on
> fire at will, that is).
>
> > The list would be if ever subject to changess on a
> > dayly base which is a real bad idea.
>
> Not unless standard SCSI commands change by the day. And I somewhat doubt
> that to be the case.

theory != practice

> > Note that having such a list of aparently safe commands would cause a lot
> > of untracable problems (why does it run for you but not for me....).
>
> Right. But better "Funny, it doesn't work here..." than "Sh*t! Another
> CD/DVD-writer turned into a brick!".

Horst, the fact that Joerg is hard to deal with and usually not right doesn't
mean that he can't be right sometimes. ;-)

Bartlomiej
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 14:05    [W:0.162 / U:6.700 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site