Michael Smith wrote:
> I've been trying to get cdrdao to read a cd (so as to be able to do
> "cdrdao copy", for example). Writing seems to work fine, but reading fails
> After apparently reading the toc fine (read-toc gives some of these errors,
> but the resultant file looks correct), the following error gets displayed
> repeatedly (at high speed, many times per second). The output file is still
> created, but the file contents is all zeroed ("cmp datafile /dev/zero" finds
> EOF on the data file).
> So, here's the error. Each time it comes up the line beginning with CDB
> changes slightly - one part of it is growing continually. The rest of the
> message is unchanged:
> ?: Input/output error. : scsi sendcmd: no error
> CDB: BE 00 00 00 00 00 00 00 14 F8 01 00
> status: 0x2 (CHECK CONDITION)
> Sense Bytes: 70 00 05 00 00 00 00 0A 00 00 00 00 24 00 00 00
> Sense Key: 0x5 Illegal Request, Segment 0
> Sense Code: 0x24 Qual 0x00 (invalid field in cdb) Fru 0x0
> Sense flags: Blk 0 (not valid)
> cmd finished after 0.001s timeout 20s
> Other information of note: cdparanoia has no troubles with the drive, in
> either native IDE mode or using scsi emulation (ide-scsi). I'm running
> linux 2.4.10 (debian/woody system). I've tried both the latest release
> (1.1.5) and current cvs, no difference noted.
> The drive itself is an ACER CRW 1208A (Which is a 32/12/8 CDRW drive, IDE).
> I've tried using the generic-mmc driver, as well as generic-mmc-raw, also
> no difference. I've also tried setting various driver flags which looked
> like they might potentially be relevent. Again, no difference.
> Advice or ideas would be greatly appreciated.
Please try one of these options:
1. --driver generic-mmc:0x1
2. --driver generic-mmc:0x3
3. --driver generic-mmc:0x20
4. --driver generic-mmc:0x20 --fast-toc
The difference between 1. and 2. is very subtle. Either both option work
without an error message or both options result in error messages. If
no error messages occur you will have to watch the time count that is
displayed during audio extraction. For one of the options the time count
will be jumpy - select the other option then.
If only 3. work the pre-gap length and index marks will be extracted using
a very unreliable algorithm which also takes a lot of time. Since the
pre-gap lengths and index marks are not very important - at least for the
audible part of a CD - you can simply switch this extraction totally off
by choosing option 4.
Andreas Mueller Tel: +49 89 67808848
Ramsmeierstr. 1 Email: andreas@...
85579 Neubiberg, Germany