alsa-devel.alsa-project.org archive mirror
 help / color / mirror / Atom feed
From: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
To: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Cc: Linux-ALSA <alsa-devel@alsa-project.org>,
	Mark Brown <broonie@kernel.org>, Lars-Peter <lars@metafoo.de>,
	Simon <horms@verge.net.au>
Subject: Re: [PATCH 5/6] ASoC: add Component level struct snd_compr_ops
Date: Tue, 20 Dec 2016 10:02:50 +0000	[thread overview]
Message-ID: <20161220100250.GP1867@localhost.localdomain> (raw)
In-Reply-To: <87inqffjkn.wl%kuninori.morimoto.gx@renesas.com>

On Tue, Dec 20, 2016 at 01:59:43AM +0000, Kuninori Morimoto wrote:
> > > +	for_each_component(component, rtd->card) {
> > > +		const struct snd_compr_ops *compr_ops = component->compr_ops;
> > > +
> > > +		if (compr_ops && compr_ops->open) {
> > > +			int _ret = compr_ops->open(cstream);
> > > +
> > > +			if (_ret < 0) {
> > > +				pr_err("compress asoc: can't open component %s\n",
> > > +				       component->name);
> > > +				ret |= _ret;
> > > +			}
> > 
> > Apologies if I am missing something but this really doesn't look
> > like equivalent code? The old code calls the ops for the platform
> > associated with the stream, the new code looks like it will call
> > the ops for every single component in the system that has
> > compressed ops, which would definitely cause issue. Am I missing
> > something here?
> 
> Basically in *current* existing code, platform only has compr_ops.
> So, converting existing code to new code should be OK.
> 
> Current ALSA SoC clearly separating CPU/Codec/Platform, but in these days,
> new sound device has many features. Thus, it becaming difficult to
> say "it is XXX" today. For example, some Codec has CPU feature,
> some CPU has Platform feature etc.
> So, our (= Me/Lars) plan is that remove current CPU/Codec/Platform separation,
> and all will be Component connection.
> 
> Above code will call every component's compr_ops,
> currently, it should be platform now.
> But in the future device, Codec and/or CPU can have compr_ops.
> 
> If you want, currently, it can call 1st found compr_ops (= should be platform)
> only with FIXME comment at this point.
> 
> Does this clear answer for you ?
> 
> 	for_each_component(component, rtd->card) {
> 		const struct snd_compr_ops *compr_ops = component->compr_ops;
> 
> 		/* FIXME: 1st compr_ops only at this point */
> 		if (compr_ops && compr_ops->open) {

But how do you know this is the correct compressed open here? The
system could have multiple platforms providing compressed ops and
you have just picked the first one in the card list. I think you
are making the assumption that there is only one platform
providing compressed ops and that seems like a very dangerous
assumption to me. Our CODECs provide some compressed features as
do many applications processors, which would easily give you at
two platforms. But I could even imagine APs registering multiple
compressed platforms for different DSPs or some such.

Thanks,
Charles

> 			ret = compr_ops->open(cstream);
> 			if (ret < 0) {
> 				pr_err("compress asoc: can't open component %s\n",
> 				       component->name);
> 			}
> 			break;
> 		}
> 	}

  reply	other threads:[~2016-12-20 10:01 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-12-19  7:33 [PATCH 0/6] ASoC: platform cleanup Kuninori Morimoto
2016-12-19  7:36 ` [PATCH 1/6] ASoC: remove .delay from snd_soc_platform_driver Kuninori Morimoto
2017-01-09 20:01   ` Applied "ASoC: remove .delay from snd_soc_platform_driver" to the asoc tree Mark Brown
2016-12-19  7:37 ` [PATCH 2/6] ASoC: remove .bespoke_trigger from snd_soc_platform_driver Kuninori Morimoto
2017-01-09 20:01   ` Applied "ASoC: remove .bespoke_trigger from snd_soc_platform_driver" to the asoc tree Mark Brown
2016-12-19  7:37 ` [PATCH 3/6] ASoC: remove snd_soc_platform_trigger() Kuninori Morimoto
2017-01-09 20:01   ` Applied "ASoC: remove snd_soc_platform_trigger()" to the asoc tree Mark Brown
2016-12-19  7:37 ` [PATCH 4/6] ASoC: add for_each_component{_safe} Kuninori Morimoto
2017-01-09 20:01   ` Applied "ASoC: add for_each_component{_safe}" to the asoc tree Mark Brown
2017-01-10 12:13     ` Mark Brown
2017-01-11  6:46       ` Kuninori Morimoto
2016-12-19  7:38 ` [PATCH 5/6] ASoC: add Component level struct snd_compr_ops Kuninori Morimoto
2016-12-19 11:12   ` Charles Keepax
2016-12-20  1:59     ` Kuninori Morimoto
2016-12-20 10:02       ` Charles Keepax [this message]
2016-12-21  0:42         ` Kuninori Morimoto
2016-12-21  9:29           ` Charles Keepax
2016-12-21 23:45             ` Kuninori Morimoto
2016-12-19  7:38 ` [PATCH 6/6] ASoC: add Component level struct snd_pcm_ops Kuninori Morimoto

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=20161220100250.GP1867@localhost.localdomain \
    --to=ckeepax@opensource.wolfsonmicro.com \
    --cc=alsa-devel@alsa-project.org \
    --cc=broonie@kernel.org \
    --cc=horms@verge.net.au \
    --cc=kuninori.morimoto.gx@renesas.com \
    --cc=lars@metafoo.de \
    /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).