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

Re: cdrecord and "get performance" scsi errors



Hi,

> # test/telltoc --drive /dev/scd1
> Initializing libburnia.pykix.org ...
> Done
> Aquiring drive '/dev/sg4' ...
> <nothing more, doesn't return>

Ahum. 
Obviously the examination of the drive gets stuck.

Do you have rw-permission on /dev/sg4 ?


> ioctl(3, CDROMAUDIOBUFSIZ, 0xbfffa148)  = 0
> close(3)                                = 0
> write(2, "Aquiring drive \'/dev/sg4\' ...\n", 30) = 30

The system call trace is probably following the
wrong thread. At least i am not recognizing much
of libburn's MMC related stuff. For example, that
ioctl CDROMAUDIOBUFSIZ does not come from libburn.

Maybe one could learn more from
  cdrskin/cdrskin -vvv dev=/dev/scd1 -atip

I expect this to get stuck too, but maybe one could
get some hint from the debugging messages.


If you are interested in finding out what's going
on with cdrskin resp. telltoc, then you can make libburn
report its SCSI commands to a log file named
  /tmp/libburn_sg_command_log

For that you would uncomment in
  libburn/sg-linux.c
this macro:
/*
#define Libburn_log_sg_commandS 1
*/

I notice that it would be better to insert 
                        fflush(fp);
before the two occurences of
                        fpcount++;


I understand this stall can be interrupted by Ctrl+C ?


> It does work on the IDE burner (DVD-RAM with ext2 fs on it,
> not ISO, so no TOC):

The TOC reported by telltoc is not the one of the data
image but the eventual one of the media structure. I.e.
the list of tracks and sessions.
DVD-RAM and DVD+RW do not have such a structure. They are
just a string of randomly addressable data blocks.
You would get TOC info if you insert a CD-R[W], DVD-R[W]
or DVD+R.


Have a nice day :)

Thomas



Reply to: