alsa-devel.alsa-project.org archive mirror
 help / color / mirror / Atom feed
From: Anssi Hannula <anssi.hannula@iki.fi>
To: Andrew Eikum <aeikum@codeweavers.com>
Cc: alsa-devel@alsa-project.org, Joerg-Cyril.Hoehle@t-systems.com
Subject: Re: how to enumerate useful ALSA devices?
Date: Wed, 04 Jan 2012 22:47:52 +0200	[thread overview]
Message-ID: <4F04BAF8.6010701@iki.fi> (raw)
In-Reply-To: <20120104194956.GB9868@foghorn.codeweavers.com>

Hi,

I'm not really an expert on ALSA api (having never actually used it),
but I'll try to clarify what I can. I hope someone will correct me if
I'm wrong on some point.

On 04.01.2012 21:49, Andrew Eikum wrote:
>> The 2008 audio guide recommends
>> http://0pointer.de/blog/projects/guide-to-sound-apis.html
>> "Do not use the snd_card_xxx() APIs. For enumerating use
>>  snd_device_name_hint() (and related functions).
>>  snd_card_xxx() is obsolete... Bluetooth audio are not included"
>>
>> Is that correct advice?
>>
>> Wine still uses snd_card, so it knows only about "default" and
>> "plughw:x,y".  I believe that's wrong but don't know what should be
>> used instead in order not to come up with a long list of artificial
>> names mentioned in /usr/share/alsa/alsa.conf such as surround51 when
>> the poor on-board card knows nothing but 16 bit stereo.
>>
>> I'm looking for a list of working devices on this machine, which likely
>> should contain "default" (usually leading to PA on a recent Ubuntu machine),
>> "plug:dmix" (PA getting out of the way when trying to use other devices),
>> "hw:0,0" and presumably not much more.

Probably not "hw:0,0", though, as AFAIK one should prefer the symbolical
names instead (e.g. those in "aplay -L").

> Bump, still hoping for an answer to this.
> 
> I tried experimenting with snd_device_name_hint(). It gave me what
> looked like a promising list, consisting of:
> 
> null
> front:CARD=SB,DEV=0
> surround40:CARD=SB,DEV=0
> surround41:CARD=SB,DEV=0
> surround50:CARD=SB,DEV=0
> surround51:CARD=SB,DEV=0
> surround71:CARD=SB,DEV=0
> iec958:CARD=SB,DEV=0
> 
> This is identical to the list from "aplay -L".
> 
> But, passing these to aplay fails if I have PulseAudio running
> ("device in use").

Correct. You need to use pulse directly or the default/pulse device
which passes audio to pulseaudio.

> If I kill PA, then they don't fail, but give
> warnings about rate conversion and suggest the plug plugin. So I stick
> "plug:" in front of the device name, and they fail again ("unknown
> parameter front:CARD"). Clearly I'm doing something very wrong. But I
> don't know what!

"plug:'surround71:CARD=SB,DEV=0'"

> Then it occurred to me that this list also contains neither "default"
> nor the "pulse" device in my asound.conf[1]. Both of these work when
> passed to aplay.

AFAIK this is because that file is missing the "hint" statement for the
devices, which are present on my system (Mageia), and also in upstream ALSA:
http://git.alsa-project.org/?p=alsa-plugins.git;a=blob;f=pulse/50-pulseaudio.conf;h=dd85dab835168bde84d7b8a3be78fb896ece8ea8;hb=HEAD
http://git.alsa-project.org/?p=alsa-plugins.git;a=blob;f=pulse/99-pulseaudio-default.conf.example;h=4f5885806898a76f5a5c9b3ec76b804fe54e2480;hb=HEAD

> Looking to other projects for help, I find that VLC uses the snd_card_
> functions, just like Wine currently does. Of course, that fails to
> find the "pulse" device from asound.conf. And mplayer's libao2 uses a
> couple of hard-coded device names, which is even worse (right?).
> 
> So I'm really stuck here. Where am I supposed to get a list of devices
> that I can present to the user, including their Bluetooth devices and
> virtual devices from asound.conf?

I guess the preferred one is 'snd_device_name_hint()', but I don't
really know if the caller is supposed to then handle surroundXX stuff
themself.

However, note that on systems using pulseaudio (i.e. the default on most
distros), an automatically (i.e. via bluetooth "add device" wizard of
their desktop) configured bluetooth headset may be handled by pulse
directly and not appear as a distinct device in ALSA at all. At least
this seems to be the case on my system.

> [1] asound.conf from Arch Linux pulseaudio-alsa package:
> http://projects.archlinux.org/svntogit/packages.git/tree/trunk/asound.conf?h=packages/pulseaudio-alsa
> 
> Andrew
> _______________________________________________
> Alsa-devel mailing list
> Alsa-devel@alsa-project.org
> http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
> 


-- 
Anssi Hannula

  reply	other threads:[~2012-01-04 20:47 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-01-04 19:49 how to enumerate useful ALSA devices? Andrew Eikum
2012-01-04 20:47 ` Anssi Hannula [this message]
2012-01-04 21:04   ` Andrew Eikum
2012-01-04 21:16   ` Anssi Hannula
2012-01-05 23:06     ` Raymond Yau
2012-01-06  0:36       ` Anssi Hannula
2012-03-11  3:01         ` Raymond Yau
2012-01-05  5:35   ` Raymond Yau
2012-01-05  9:59 ` Joerg-Cyril.Hoehle
2012-01-05 13:30   ` Raymond Yau
2012-01-05 17:26   ` Anssi Hannula
  -- strict thread matches above, loose matches on Subject: below --
2011-12-21 14:37 Joerg-Cyril.Hoehle
2011-12-22  6:27 ` Raymond Yau

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4F04BAF8.6010701@iki.fi \
    --to=anssi.hannula@iki.fi \
    --cc=Joerg-Cyril.Hoehle@t-systems.com \
    --cc=aeikum@codeweavers.com \
    --cc=alsa-devel@alsa-project.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).