[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: lots of c2scan errors without ide-scsi



Christoph Heine <Christoph.Heine@HaDiKo.DE> wrote:

> Hi,
>
> On Sat, Sep 11, 2004 at 03:36:04PM +0200, Joerg Schilling wrote:
> > > Is this a bug in the kernel or in readcd? 
> > >
> > > I'm using Debian unstable, readcd from the cdrecord 2.0+a38-1 package,
> > > kernel 2.6.8.1, and a Plextor PX-W1210A with firmware 1.10. 
> > 
> > The kernel you are using is defective.
>
> After trying several -rc versions, I could identify the guilty patch:
>
> http://linux.bkbits.net:8080/linux-2.5/diffs/drivers/block/scsi_ioctl.c@1.49?nav=index.html|src/|src/drivers|src/drivers/block|hist/drivers/block/scsi_ioctl.c
>
> --- 1.48/drivers/block/scsi_ioctl.c	2004-09-11 10:51:59 -07:00
> +++ 1.49/drivers/block/scsi_ioctl.c	2004-09-11 10:51:59 -07:00
> @@ -90,7 +90,7 @@
>  	if (size < 0)
>  		return -EINVAL;
>  	if (size > (q->max_sectors << 9))
> -		return -EINVAL;
> +		size = q->max_sectors << 9;
>  
>  	q->sg_reserved_size = size;
>  	return 0;
>
>
> After reversing this patch, the c2scan works as expected. 

This is most unlikely.

The patch should make the linux kernel behave as documented.

Readcd just uses the standard code in libscg that fails completely
if you hit the cause fot this patch as libscg just thinks that the driver
is broken and does not continue with the old code.

Try to experiment with readcd's ts= option and make a bug report against 
the Linux kernel.... It looks like there is a DMA implementation bug in
the Linux kernel if the DMA size has effects on the read data.

Jörg

-- 
 EMail:joerg@schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin
       js@cs.tu-berlin.de		(uni)  If you don't have iso-8859-1
       schilling@fokus.fraunhofer.de	(work) chars I am J"org Schilling
 URL:  http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily



Reply to: