kern/60163
Søren Schmidt
sos at deepcore.dk
Tue Feb 14 04:17:53 PST 2006
Ariff Abdullah wrote:
> On Mon, 24 Feb 2003 00:26:11 -0800
> soralx at cydem.org wrote:
>>>> Hmm, could it be that the data track is always the last track ?
>>> Yes.
>> so that CD players would be able to play the disks? or does not
>> matter?
>>
> Correct. Most hardware cd player are designed to play only the first
> session on multisession disc, where in this case (CDEXTRA), the data
> track will be ignored. Unlike mixed-mode cd, the data track (on
> first track/session) usually result in silence playback.
>
> In case somebody might interested, I have a patch for (unfortunately)
> RELENG_5 for atapicd multiblock access. Among other things, it
> also fix multisession/dao writing of burncd(8). With this, you can
> have conccurent access with varying blocksize on atapicd.
>
> http://people.freebsd.org/~ariff/misc/releng5_ata.diff
>
> It's a bit ugly. Somebody with deeper knowledge on ata/GEOM probably
> will have better solution for this.
Thats closer to a real solution. The problem is that for a CD EXTRA the
offsets embedded in the iso is absolute to the start of media, if we
address those directly as a track the offsets are off.
This cannot coexist with CD containing multiple tracks with relative to
the track start iso's on each track.
We need a solution to this that does the right thing but its certainly
not trivial, the above does contain some of the solution (from an
eyeball review of the code), but more needs to be done..
I'll look into this when more important issues has been dealt with for
the 6.1 release, possibly together with a burncd replacement I'm working
on when time permits...
-Søren
More information about the freebsd-hackers
mailing list