[PATCH] ASoC: Writing register default value for the reset register

[PATCH] ASoC: Writing register default value for the reset register

[Axel Lin]

The WM8983 can be reset by performing a write of any value to
the software reset register.
To avoid writing to the software reset register while resume,
we should write the same value in wm8983_reg_defs to software
reset register in wm8983_probe().
The write to the reset register is suppressed by the cache
restore code when it skips writes of default registers.

Signed-off-by: Axel Lin <axel.lin@gmail.com>
---
 sound/soc/codecs/wm8983.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/sound/soc/codecs/wm8983.c b/sound/soc/codecs/wm8983.c
index 17f04ec..93ee284 100644
--- a/sound/soc/codecs/wm8983.c
+++ b/sound/soc/codecs/wm8983.c
@@ -1007,7 +1007,7 @@ static int wm8983_probe(struct snd_soc_codec *codec)
 		return ret;
 	}
 
-	ret = snd_soc_write(codec, WM8983_SOFTWARE_RESET, 0x8983);
+	ret = snd_soc_write(codec, WM8983_SOFTWARE_RESET, 0);
 	if (ret < 0) {
 		dev_err(codec->dev, "Failed to issue reset: %d\n", ret);
 		return ret;
-- 
1.7.4.1

Re: [PATCH] ASoC: Writing register default value for the reset register

[Dimitris Papastamos]

On Tue, Oct 11, 2011 at 05:55:00PM +0800, Axel Lin wrote:
> The WM8983 can be reset by performing a write of any value to
> the software reset register.
> To avoid writing to the software reset register while resume,
> we should write the same value in wm8983_reg_defs to software
> reset register in wm8983_probe().
> The write to the reset register is suppressed by the cache
> restore code when it skips writes of default registers.
> 
> Signed-off-by: Axel Lin <axel.lin@gmail.com>
> ---
>  sound/soc/codecs/wm8983.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/sound/soc/codecs/wm8983.c b/sound/soc/codecs/wm8983.c
> index 17f04ec..93ee284 100644
> --- a/sound/soc/codecs/wm8983.c
> +++ b/sound/soc/codecs/wm8983.c
> @@ -1007,7 +1007,7 @@ static int wm8983_probe(struct snd_soc_codec *codec)
>  		return ret;
>  	}
>  
> -	ret = snd_soc_write(codec, WM8983_SOFTWARE_RESET, 0x8983);
> +	ret = snd_soc_write(codec, WM8983_SOFTWARE_RESET, 0);
>  	if (ret < 0) {
>  		dev_err(codec->dev, "Failed to issue reset: %d\n", ret);
>  		return ret;

We should probably just handle the reset register specially during
resume.

Thanks,
Dimitris

Re: [PATCH] ASoC: Writing register default value for the reset register

[Mark Brown]

On Tue, Oct 11, 2011 at 01:33:05PM +0100, Dimitris Papastamos wrote:
> On Tue, Oct 11, 2011 at 05:55:00PM +0800, Axel Lin wrote:
> > The WM8983 can be reset by performing a write of any value to
> > the software reset register.

> > -	ret = snd_soc_write(codec, WM8983_SOFTWARE_RESET, 0x8983);
> > +	ret = snd_soc_write(codec, WM8983_SOFTWARE_RESET, 0);

> We should probably just handle the reset register specially during
> resume.

I really don't think it's worth the code complexity to add special
handling like that, this is a trivial patch which avoids the need for
any extra infrastructure.

Re: [PATCH] ASoC: Writing register default value for the reset register

[Mark Brown]

On Tue, Oct 11, 2011 at 05:55:00PM +0800, Axel Lin wrote:
> The WM8983 can be reset by performing a write of any value to
> the software reset register.

Applied, thanks.

Re: [PATCH] ASoC: Writing register default value for the reset register

[Dimitris Papastamos]

On Tue, Oct 11, 2011 at 02:31:18PM +0100, Mark Brown wrote:
> On Tue, Oct 11, 2011 at 01:33:05PM +0100, Dimitris Papastamos wrote:
> > On Tue, Oct 11, 2011 at 05:55:00PM +0800, Axel Lin wrote:
> > > The WM8983 can be reset by performing a write of any value to
> > > the software reset register.
> 
> > > -	ret = snd_soc_write(codec, WM8983_SOFTWARE_RESET, 0x8983);
> > > +	ret = snd_soc_write(codec, WM8983_SOFTWARE_RESET, 0);
> 
> > We should probably just handle the reset register specially during
> > resume.
> 
> I really don't think it's worth the code complexity to add special
> handling like that, this is a trivial patch which avoids the need for
> any extra infrastructure.

Aw I thought this code was using the old style open coded syncing
functions, which would be trivial and perhaps idiomatic to simply
check for the reset register and continue.

Thanks,
Dimitris