Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 22 Dec 2011 21:46:19 +0200
From:      Andriy Gapon <avg@FreeBSD.org>
To:        Alexander Motin <mav@FreeBSD.org>
Cc:        svn-src-head@FreeBSD.org, svn-src-all@FreeBSD.org, src-committers@FreeBSD.org
Subject:   Re: svn commit: r228808 - head/sys/cam/scsi
Message-ID:  <4EF3890B.6020408@FreeBSD.org>
In-Reply-To: <4EF387F4.9050008@FreeBSD.org>
References:  <201112221640.pBMGeZod025237@svn.freebsd.org> <4EF375E5.2010809@FreeBSD.org> <4EF387F4.9050008@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
on 22/12/2011 21:41 Alexander Motin said the following:
> On 22.12.2011 20:24, Andriy Gapon wrote:
>> on 22/12/2011 18:40 Alexander Motin said the following:
>>> Author: mav
>>> Date: Thu Dec 22 16:40:35 2011
>>> New Revision: 228808
>>> URL: http://svn.freebsd.org/changeset/base/228808
>>>
>>> Log:
>>>    Make cd driver to handle Audio CDs, reporting their 2352 bytes sectors to
>>>    GEOM and using READ CD command for reading data, same as acd driver does.
>>>    Audio CDs identified by checking respective bit of the control field of
>>>    the first track in TOC.
>>>
>>>    This fixes bunch of error messages during boot (GEOM taste) with Audio CD
>>>    inserted and allows to grab Audio CD image using just dd.
>>
>> Thank you!
>> What about CDs with mixed audio and data tracks?
>> I don't ask for any support, just curious if our users could run into any
>> troubles with those.
> 
> If data track is the first (I don't know whether it is mandatory, but wiki tells
> it is usual), the whole disk will be handled as data by the driver and work as
> before (only data part will be readable). If it can be not the first, then I am
> not sure how acd driver works with it, but we can just make check a bit more
> strict, switching to data if found it anywhere on disk. Any way I don't think we
> will see many Mixed Mode CDs in the future, while Audio CDs are still on market.
> 
> If you worry about audio extraction of Mixed Mode CDs, then quick look on
> mplayer and it's underlying cdparanoia and libcdio shows that they are
> extracting audio data talking directly to the drive using SCSI commands, so cd
> driver is irrelevant for them. So this change is mostly about staying consistent
> when possible with a small blood.

Yes, I agree.  As I've said I wasn't asking for any new capabilities, just
wanted to make that no new edge cases are introduced.

BTW, acd(4) ... not sure which tense I should use here ... used to create
individual devices for each track - e.g. acd0t1, acd0t2 - where each track could
have its own properties.


-- 
Andriy Gapon



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4EF3890B.6020408>