[Paranoia] copy protection

[Peter Jones]

On Wed, 2005-01-19 at 08:10 -0700, Aaron Birenboim wrote:
> On Tue, 2005-01-18 at 15:14, Peter Jones wrote:
[...]
> fairly typical music CD's.
> (Details sent in private response)
> 
> They have pops and some repeats (like a skipping CD)
> on the rip.
> paranoia reports V's and !'s.

OK, I'm pretty sure this isn't a copy protection problem.

> > > xmms seems to be able to play them.
> > >
> > > I am running 2.6 kernel, and using the IDE-ATAPI device.
> > > 
> > > Will it help me to go to SCSI-generic emulation on
> > > the 2.6 kernel?  (I thought 2.6 improved IDE, so that
> > > was not necessary)
> > > Do I really have to remove IDE-ATAPI from 2.6 kernel
> > > to get this to work? (Can I just load a SCSI-generic module
> > > and use /dev/sg2?)
> > 
> > At the moment, it unfortunately depends on where you got cdparanoia.
> 
> I was running from svn to xiph.org.
> I think i updated and re-tried yesterday morning, similar results.

Yeah, the SG_IO patch in the RH package is a bit of a kludge, so I
haven't added it to the svn tree yet.  You might want to try the FC3
package and see if it helps.

If it does, it'll more likely be because it tries to use much smaller
buffers during the reads.  I'm pretty sure what you're seeing is a
kernel bug I've seen before, but can't consistently reproduce.

> > Unfortunately, I haven't merged that code into the svn repo for
> > cdparanoia yet, or done a release with it, so I don't think other
> > distros are using that yet.
> 
> I'm on debian, so it never occurred to me to get a redhat FC3 package.
> I can try.
> 
> Similar results (one may have reported all zeros though...)
> on both my ancient CD-ROM, and fairly new (but very cheap)
> DVD-RW.

Yeah, I'm pretty sure you're seeing this kernel problem I've spoken of.

> Wasn't there a command to probe the PCI/IDE bus and get some
> sort of vendor code from the hardware?
> If so...  I wouldn't mind being reminded, then I can positively
> ID the hardware.

cdparanoia -vQ

> I have ripped other CD's from this drive.

That's a bit puzzling; I wouldn't have expected it'd vary by the CD.  I
guess it could be an under-run problem, so that'd have results like
this...

> > In general, sg and ide-scsi are the way of the past, and need to be done
> > away with in a very serious way.

OK, so that comment probably didn't go over so well with Doug.  Sorry
Doug, I didn't mean this as broadly as I said it.  I don't think
cdparanoia should be using these; there are clearly some cases where
they're still a good solution to real situations.

> I will be happy to send any output you want.
> Just let me know if you have any extra flags you want (like -v...?)
> Also let me know if I should get the latest FC3 RPM, or if
> you have more updates since the release... perhaps we
> could synch through xiph.org or was there a version at sourceforge?

Try the source tree at
http://people.redhat.com/pjones/cdparanoia/cdparanoia-III-alpha9.8.tar.gz

And be sure /dev/cdrom is a link to /dev/hda (that was your cd device,
right?  if not, /dev/hdc, or whichever block device is appropriate on
your machine)
-- 
        Peter