All of lore.kernel.org
 help / color / mirror / Atom feed
From: Grant Likely <grant.likely@secretlab.ca>
To: Jon Smirl <jonsmirl@gmail.com>
Cc: linuxppc-dev@ozlabs.org, alsa-devel@alsa-project.org
Subject: Re: [PATCH 2/2] Allow a custom ASOC machine driver with	soc-of-simple
Date: Sat, 26 Jul 2008 19:12:06 -0600	[thread overview]
Message-ID: <20080727011206.GD12191@secretlab.ca> (raw)
In-Reply-To: <20080722235351.11648.56387.stgit@terra>

On Tue, Jul 22, 2008 at 07:53:51PM -0400, Jon Smirl wrote:
> Allow a custom ASOC machine driver with soc-of-simple
> 
> Signed-off-by: Jon Smirl <jonsmirl@gmail.com>

Sorry I didn't respond earlier.  OLS kept me pretty distracted.

Need a more detailed comment block about how your changing things and
why.

> ---
> 
>  include/sound/soc-of-simple.h |    2 ++
>  sound/soc/fsl/soc-of-simple.c |   26 +++++++++++++++++++++-----
>  2 files changed, 23 insertions(+), 5 deletions(-)
> 
> diff --git a/include/sound/soc-of-simple.h b/include/sound/soc-of-simple.h
> index 696fc51..1e83f2f 100644
> --- a/include/sound/soc-of-simple.h
> +++ b/include/sound/soc-of-simple.h
> @@ -18,4 +18,6 @@ int of_snd_soc_register_platform(struct snd_soc_platform *platform,
>  				 struct device_node *node,
>  				 struct snd_soc_dai *cpu_dai);
>  
> +void of_snd_soc_register_machine(char *name, struct snd_soc_ops *ops);
> +
>  #endif /* _INCLUDE_SOC_OF_H_ */
> diff --git a/sound/soc/fsl/soc-of-simple.c b/sound/soc/fsl/soc-of-simple.c
> index 0382fda..dd2fa23 100644
> --- a/sound/soc/fsl/soc-of-simple.c
> +++ b/sound/soc/fsl/soc-of-simple.c
> @@ -38,8 +38,8 @@ struct of_snd_soc_device {
>  	struct device_node *codec_node;
>  };
>  
> -static struct snd_soc_ops of_snd_soc_ops = {
> -};
> +static struct snd_soc_ops *machine_ops = NULL;
> +static char *machine_name = NULL;

Doing this prevents multiple instances of this machine driver (which is
exactly what I have on my board).  To register a machine driver it
creates a 3-way bind condition instead of the existing 2-way one.  Right
now it is easy to match up codec and platform drivers because a common
key is available in the device tree.

Alternately, it might be okay to only allow for a single machine driver
that is able to create multiple sound card instances, but this current
code just uses the same name and ops for each registered device.

>  
>  static struct of_snd_soc_device *
>  of_snd_soc_get_device(struct device_node *codec_node)
> @@ -61,7 +61,7 @@ of_snd_soc_get_device(struct device_node *codec_node)
>  	of_soc->machine.dai_link = &of_soc->dai_link;
>  	of_soc->machine.num_links = 1;
>  	of_soc->device.machine = &of_soc->machine;
> -	of_soc->dai_link.ops = &of_snd_soc_ops;
> +	of_soc->dai_link.ops = machine_ops;
>  	list_add(&of_soc->list, &of_snd_soc_device_list);
>  
>  	return of_soc;
> @@ -74,7 +74,7 @@ static void of_snd_soc_register_device(struct of_snd_soc_device *of_soc)
>  
>  	/* Only register the device if both the codec and platform have
>  	 * been registered */
> -	if ((!of_soc->device.codec_data) || (!of_soc->platform_node))
> +	if ((!of_soc->device.codec_data) || (!of_soc->platform_node) || !machine_name)

I'm not thrilled with the hard requirement for a machine driver, but I
see what you're trying to do.  I want to find a clean way to trigger
this behaviour in the device tree without resorting to encoding linux
internal details into the data.  Need to think about this more.

>  		return;
>  
>  	pr_info("platform<-->codec match achieved; registering machine\n");
> @@ -159,7 +159,7 @@ int of_snd_soc_register_platform(struct snd_soc_platform *platform,
>  	of_soc->platform_node = node;
>  	of_soc->dai_link.cpu_dai = cpu_dai;
>  	of_soc->device.platform = platform;
> -	of_soc->machine.name = of_soc->dai_link.cpu_dai->name;
> +	of_soc->machine.name = machine_name;

As mentioned above, either there needs to be multiple machine drivers
or the ability to change the name for each platform--codec pair.

>  
>  	/* Now try to register the SoC device */
>  	of_snd_soc_register_device(of_soc);
> @@ -169,3 +169,19 @@ int of_snd_soc_register_platform(struct snd_soc_platform *platform,
>  	return rc;
>  }
>  EXPORT_SYMBOL_GPL(of_snd_soc_register_platform);
> +
> +void of_snd_soc_register_machine(char *name, struct snd_soc_ops *ops)
> +{
> +	struct of_snd_soc_device *of_soc;
> +	
> +	machine_name = name;
> +	machine_ops = ops;
> +	
> +	list_for_each_entry(of_soc, &of_snd_soc_device_list, list) {
> +		of_soc->dai_link.ops = machine_ops;
> +		of_soc->machine.name = machine_name;
> +		of_snd_soc_register_device(of_soc);
> +	}
> +
> +}

You need to hold the mutex when manipulating the list.

WARNING: multiple messages have this Message-ID (diff)
From: Grant Likely <grant.likely@secretlab.ca>
To: Jon Smirl <jonsmirl@gmail.com>
Cc: linuxppc-dev@ozlabs.org, alsa-devel@alsa-project.org
Subject: Re: [PATCH 2/2] Allow a custom ASOC machine driver with soc-of-simple
Date: Sat, 26 Jul 2008 19:12:06 -0600	[thread overview]
Message-ID: <20080727011206.GD12191@secretlab.ca> (raw)
In-Reply-To: <20080722235351.11648.56387.stgit@terra>

On Tue, Jul 22, 2008 at 07:53:51PM -0400, Jon Smirl wrote:
> Allow a custom ASOC machine driver with soc-of-simple
> 
> Signed-off-by: Jon Smirl <jonsmirl@gmail.com>

Sorry I didn't respond earlier.  OLS kept me pretty distracted.

Need a more detailed comment block about how your changing things and
why.

> ---
> 
>  include/sound/soc-of-simple.h |    2 ++
>  sound/soc/fsl/soc-of-simple.c |   26 +++++++++++++++++++++-----
>  2 files changed, 23 insertions(+), 5 deletions(-)
> 
> diff --git a/include/sound/soc-of-simple.h b/include/sound/soc-of-simple.h
> index 696fc51..1e83f2f 100644
> --- a/include/sound/soc-of-simple.h
> +++ b/include/sound/soc-of-simple.h
> @@ -18,4 +18,6 @@ int of_snd_soc_register_platform(struct snd_soc_platform *platform,
>  				 struct device_node *node,
>  				 struct snd_soc_dai *cpu_dai);
>  
> +void of_snd_soc_register_machine(char *name, struct snd_soc_ops *ops);
> +
>  #endif /* _INCLUDE_SOC_OF_H_ */
> diff --git a/sound/soc/fsl/soc-of-simple.c b/sound/soc/fsl/soc-of-simple.c
> index 0382fda..dd2fa23 100644
> --- a/sound/soc/fsl/soc-of-simple.c
> +++ b/sound/soc/fsl/soc-of-simple.c
> @@ -38,8 +38,8 @@ struct of_snd_soc_device {
>  	struct device_node *codec_node;
>  };
>  
> -static struct snd_soc_ops of_snd_soc_ops = {
> -};
> +static struct snd_soc_ops *machine_ops = NULL;
> +static char *machine_name = NULL;

Doing this prevents multiple instances of this machine driver (which is
exactly what I have on my board).  To register a machine driver it
creates a 3-way bind condition instead of the existing 2-way one.  Right
now it is easy to match up codec and platform drivers because a common
key is available in the device tree.

Alternately, it might be okay to only allow for a single machine driver
that is able to create multiple sound card instances, but this current
code just uses the same name and ops for each registered device.

>  
>  static struct of_snd_soc_device *
>  of_snd_soc_get_device(struct device_node *codec_node)
> @@ -61,7 +61,7 @@ of_snd_soc_get_device(struct device_node *codec_node)
>  	of_soc->machine.dai_link = &of_soc->dai_link;
>  	of_soc->machine.num_links = 1;
>  	of_soc->device.machine = &of_soc->machine;
> -	of_soc->dai_link.ops = &of_snd_soc_ops;
> +	of_soc->dai_link.ops = machine_ops;
>  	list_add(&of_soc->list, &of_snd_soc_device_list);
>  
>  	return of_soc;
> @@ -74,7 +74,7 @@ static void of_snd_soc_register_device(struct of_snd_soc_device *of_soc)
>  
>  	/* Only register the device if both the codec and platform have
>  	 * been registered */
> -	if ((!of_soc->device.codec_data) || (!of_soc->platform_node))
> +	if ((!of_soc->device.codec_data) || (!of_soc->platform_node) || !machine_name)

I'm not thrilled with the hard requirement for a machine driver, but I
see what you're trying to do.  I want to find a clean way to trigger
this behaviour in the device tree without resorting to encoding linux
internal details into the data.  Need to think about this more.

>  		return;
>  
>  	pr_info("platform<-->codec match achieved; registering machine\n");
> @@ -159,7 +159,7 @@ int of_snd_soc_register_platform(struct snd_soc_platform *platform,
>  	of_soc->platform_node = node;
>  	of_soc->dai_link.cpu_dai = cpu_dai;
>  	of_soc->device.platform = platform;
> -	of_soc->machine.name = of_soc->dai_link.cpu_dai->name;
> +	of_soc->machine.name = machine_name;

As mentioned above, either there needs to be multiple machine drivers
or the ability to change the name for each platform--codec pair.

>  
>  	/* Now try to register the SoC device */
>  	of_snd_soc_register_device(of_soc);
> @@ -169,3 +169,19 @@ int of_snd_soc_register_platform(struct snd_soc_platform *platform,
>  	return rc;
>  }
>  EXPORT_SYMBOL_GPL(of_snd_soc_register_platform);
> +
> +void of_snd_soc_register_machine(char *name, struct snd_soc_ops *ops)
> +{
> +	struct of_snd_soc_device *of_soc;
> +	
> +	machine_name = name;
> +	machine_ops = ops;
> +	
> +	list_for_each_entry(of_soc, &of_snd_soc_device_list, list) {
> +		of_soc->dai_link.ops = machine_ops;
> +		of_soc->machine.name = machine_name;
> +		of_snd_soc_register_device(of_soc);
> +	}
> +
> +}

You need to hold the mutex when manipulating the list.

  parent reply	other threads:[~2008-07-27  4:04 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-07-22 23:53 [PATCH 1/2] Support internal I2S clock sources on MPC5200 Jon Smirl
2008-07-22 23:53 ` [PATCH 2/2] Allow a custom ASOC machine driver with soc-of-simple Jon Smirl
2008-07-23  5:30   ` Stephen Rothwell
2008-07-23  5:30     ` Stephen Rothwell
2008-07-23 10:53   ` Mark Brown
2008-07-23 10:53     ` [alsa-devel] " Mark Brown
2008-07-23 14:09     ` Jon Smirl
2008-07-23 14:09       ` [alsa-devel] " Jon Smirl
2008-07-23 15:14       ` Mark Brown
2008-07-23 15:14         ` [alsa-devel] " Mark Brown
2008-07-23 15:16         ` Jon Smirl
2008-07-23 15:16           ` [alsa-devel] " Jon Smirl
2008-07-23 15:36           ` Mark Brown
2008-07-23 15:36             ` [alsa-devel] " Mark Brown
2008-07-27  1:14         ` Grant Likely
2008-07-27  1:14           ` [alsa-devel] " Grant Likely
2008-07-24  6:20       ` filling dummy struct snd_dma_buffer dinesh
2008-07-24  6:20         ` dinesh
2008-07-27  1:12   ` Grant Likely [this message]
2008-07-27  1:12     ` [PATCH 2/2] Allow a custom ASOC machine driver with soc-of-simple Grant Likely
2008-07-27  4:44     ` Jon Smirl
2008-07-27  5:29       ` Grant Likely
2008-07-27 14:07         ` Jon Smirl
2008-07-23 10:36 ` [PATCH 1/2] Support internal I2S clock sources on MPC5200 Mark Brown
2008-07-23 10:36   ` [alsa-devel] " Mark Brown
2008-07-27  2:05 ` Grant Likely
2008-07-27  4:48   ` Jon Smirl

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=20080727011206.GD12191@secretlab.ca \
    --to=grant.likely@secretlab.ca \
    --cc=alsa-devel@alsa-project.org \
    --cc=jonsmirl@gmail.com \
    --cc=linuxppc-dev@ozlabs.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 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.