All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jarkko Nikula <jhnikula@gmail.com>
To: Peter Ujfalusi <peter.ujfalusi@nokia.com>
Cc: alsa-devel@alsa-project.org,
	Mark Brown <broonie@opensource.wolfsonmicro.com>,
	Liam Girdwood <lrg@slimlogic.co.uk>
Subject: Re: [PATCH] ASoC: Add optional pointer to machine audio routes to snd_soc_card
Date: Tue, 21 Dec 2010 10:16:59 +0200	[thread overview]
Message-ID: <20101221101659.4fde5a83.jhnikula@gmail.com> (raw)
In-Reply-To: <201012210930.46971.peter.ujfalusi@nokia.com>

On Tue, 21 Dec 2010 09:30:46 +0200
Peter Ujfalusi <peter.ujfalusi@nokia.com> wrote:

> > This is due if there are machines with multiple codecs that are not
> > cross-connected but if they have widgets with a same name and I didn't
> > want to force them to use name prefixing and thus causing
> > userspace-kernel space breakage (kcontrol names changed) compared to
> > current implementation.
> 
> If you have multiple codecs in your device, it is likely that those codecs will 
> have colliding widget names for sure.
> IMHO using prefixes for multi codec cards shall not break the userspace-kernel 
> interface, since with old ASoC we had one codec per one card.
> So if you move your machine to real multi-comp, then you anyway break the 
> userspace-kernel interface by moving the PCMs to one card from the original PCM 
> per card way.
> If you have one codec in your system you do not need to prefix that, so you 
> break nothing.
> I would use prefixes in real multi codec cases...
>
It's not entirely sure. Before cross-device set, i.e. in 2.6.37 it's
possible to build a machine driver with multiple codecs that can have
the same widgets names since the paths and widgets are per codec.
Collision could come only from kcontrol names.

But I agree. Prefixing new machines sounds safer in long run. Then it's
more unlikely that changes in one codec driver causes hard to see
problems to machines with multiple codecs.

> Are we using the prefixes for the non DAPM kcontrols as well? I think we shall, 
> if the prefix is defined.
> 
Yep, prefix is put to all codec kcontrols, widget and route names.

-- 
Jarkko

      reply	other threads:[~2010-12-21  8:15 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-12-17 17:39 [PATCH] ASoC: Add optional pointer to machine audio routes to snd_soc_card Jarkko Nikula
2010-12-17 22:37 ` Mark Brown
2010-12-20  9:24   ` Liam Girdwood
2010-12-20 11:22 ` Peter Ujfalusi
2010-12-20 12:02   ` Mark Brown
2010-12-21  7:42     ` Peter Ujfalusi
2010-12-21 16:29       ` Mark Brown
2010-12-20 12:17   ` Jarkko Nikula
2010-12-21  7:30     ` Peter Ujfalusi
2010-12-21  8:16       ` Jarkko Nikula [this message]

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=20101221101659.4fde5a83.jhnikula@gmail.com \
    --to=jhnikula@gmail.com \
    --cc=alsa-devel@alsa-project.org \
    --cc=broonie@opensource.wolfsonmicro.com \
    --cc=lrg@slimlogic.co.uk \
    --cc=peter.ujfalusi@nokia.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.