Re: PATCH: cdrecord: avoiding scsi device numbering for ide devices

From: Jens Axboe
Date: Thu Aug 05 2004 - 00:49:04 EST


On Thu, Aug 05 2004, H.Rosmanith (Kernel Mailing List) wrote:
> > On Wed, Aug 04 2004, Jens Axboe wrote:
> > > > + * Sat Jun 12 12:48:12 CEST 2004 herp - Herbert Rosmanith
> > > > + * Force ATAPI driver if dev= starts with /dev/hd and device
> > > > + * is present in /proc/ide/hdX
> > > > + *
> > >
> > > That's an extremely bad idea, you want to force ATA driver in either
> > > case.
> >
> > Which, happily, is what already happens and why it works fine when you
>
> okay - my last email in this matter to LKML, but: it seems to only work
> fine if you use ide-scsi and configure it acordingly. on our system, where
> I have disabled scsi completely (ide-scsi doesnt work at all for certain
> tasks, and beside from that, I need scsi), cdrecord/cdrtools will
> terminate with "Cannot open /dev/hdX. Cannot open SCSI driver".
>
> this is the reason why the patch forces the ata (atapi?) driver. no
> SCSI driver or configuring of ide-scsi required.

Maybe newer version broke then. Until very recently, cdrecord worked
just fine as-is and used SG_IO access method when you used open by
device name. Which was just the way we wanted it.

If that doesn't work now, I suggest you take it up with Joerg. It's a
problem with his program.

> > just do -dev=/dev/hdX. What should be removed is the warning that
> > cdrecord spits out when you do this, and the whole ATAPI thing
> > should just mirror ATA and scsi-linux-ata be killed completely.
> >
> > So I suggest you do that instead and send it to Joerg,
> > cdrecord/cdrtool
>
> well, sigh .... been there, done that, but emails to Joerg seem to
> have a long RTT. therefore, LKML. sorry for the inconvenience :->

Is there no cdrecord list? lkml surely isn't appropriate.

--
Jens Axboe

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