On Thu, Sep 23, 2010 at 11:40 PM, Mark Brown broonie@opensource.wolfsonmicro.com wrote:

At least some of the systems using this device have multiple audio subsystems so provide some guidance to userspace about which one this is.

Signed-off-by: Mark Brown broonie@opensource.wolfsonmicro.com

sound/soc/s3c24xx/smdk_wm9713.c |    2 +-  1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/sound/soc/s3c24xx/smdk_wm9713.c b/sound/soc/s3c24xx/smdk_wm9713.c index 8619a9a..78a4a30 100644 --- a/sound/soc/s3c24xx/smdk_wm9713.c +++ b/sound/soc/s3c24xx/smdk_wm9713.c @@ -52,7 +52,7 @@ static struct snd_soc_dai_link smdk_dai = {  };

static struct snd_soc_card smdk = {

• .name = "SMDK",

• .name = "SMDK WM9713",

.dai_link = &smdk_dai,        .num_links = 1,  };

Acked-by: Jassi Brar jassi.brar@samsung.com

On Thu, Sep 23, 2010 at 11:40 PM, Mark Brown broonie@opensource.wolfsonmicro.com wrote:

Otherwise we try to re-register the CODEC device if the module is reloaded and sysfs becomes miserable.

Signed-off-by: Mark Brown broonie@opensource.wolfsonmicro.com

sound/soc/s3c24xx/smdk_wm9713.c |    1 +  1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/sound/soc/s3c24xx/smdk_wm9713.c b/sound/soc/s3c24xx/smdk_wm9713.c index 78a4a30..33ba8fd 100644 --- a/sound/soc/s3c24xx/smdk_wm9713.c +++ b/sound/soc/s3c24xx/smdk_wm9713.c @@ -95,6 +95,7 @@ err:  static void __exit smdk_exit(void)  {        platform_device_unregister(smdk_snd_ac97_device);

• platform_device_unregister(smdk_snd_wm9713_device);

}

module_init(smdk_init);

I think we need more than this to make module load/unload/load work. Like the soc-audio device added. Wouldn't that spoil the party next time the module is tried to loaded ?

On Fri, Sep 24, 2010 at 9:01 PM, Mark Brown broonie@opensource.wolfsonmicro.com wrote:

On Fri, Sep 24, 2010 at 08:10:50PM +0900, Jassi Brar wrote:

...

On Fri, Sep 24, 2010 at 6:22 PM, Mark Brown

...

...

Could you be more specific about what other changes you believe are needed?

...

IIUIC, the aim of this patch is to be able to reload the module, which isn't going to work simply with this chage - we need to do as well something about the 'soc-audio' platform device added the first time ? So IMHO, this change is a right step but not enough.

As I said in reply to your earlier mail the pre-existing code in the driver already does that:

--- a/sound/soc/s3c24xx/smdk_wm9713.c +++ b/sound/soc/s3c24xx/smdk_wm9713.c @@ -95,6 +95,7 @@ err:  static void __exit smdk_exit(void)  {        platform_device_unregister(smdk_snd_ac97_device);

This preexisting line from the context is cleaning up the soc-audio device.

• platform_device_unregister(smdk_snd_wm9713_device);

This new line cleans up the wm9713-codec device which is also added in the probe function (yay AC'97!).

Reloading the module works, I've been doing that rather a lot.

Ok, I forgot unregister would free the object.