Re: [PATCH 2/4] ALSA: hda/senary: Implement proper resume callback

[Takashi Iwai]

On Thu, 05 Mar 2026 04:27:28 +0100,
wangdich9700@xxxxxxx wrote:
> 
> From: wangdicheng <wangdicheng@xxxxxxxxxx>
> 
> The SN6186 codec requires register re-synchronization after resume from
> S3 state.

When no resume callback is set, regcache_sync() is called via
snd_hdac_regmap_sync(), but after snd_hda_codec_init().

So this looks superfluous to me.  Please check whether something
really missing or not.


thanks,

Takashi


> The current driver lacks a resume callback, which may lead to
> loss of hardware state or audio malfunction after system resume.
> 
> This patch adds a resume callback that:
> 1. Syncs the regcache to restore register values.
> 2. Re-applies the vendor-specific hardware init verbs.
> 3. Reports jack status to update the audio path.
> 
> Signed-off-by: wangdicheng <wangdicheng@xxxxxxxxxx>
> ---
>  sound/hda/codecs/senarytech.c | 13 +++++++++++++
>  1 file changed, 13 insertions(+)
> 
> diff --git a/sound/hda/codecs/senarytech.c b/sound/hda/codecs/senarytech.c
> index f9a389df3a17..8ebd974817ac 100644
> --- a/sound/hda/codecs/senarytech.c
> +++ b/sound/hda/codecs/senarytech.c
> @@ -205,6 +205,18 @@ static int senary_suspend(struct hda_codec *codec)
>  	return 0;
>  }
>  
> +static int senary_resume(struct hda_codec *codec)
> +{
> +	/* Re-sync registers */
> +	regcache_sync(codec->core.regmap);
> +
> +	/* Re-run init verbs to restore hardware state */
> +	senary_init_verb(codec);
> +
> +	snd_hda_jack_report_sync(codec);
> +	return 0;
> +}
> +
>  static int senary_probe(struct hda_codec *codec, const struct hda_device_id *id)
>  {
>  	struct senary_spec *spec;
> @@ -286,6 +298,7 @@ static const struct hda_codec_ops senary_codec_ops = {
>  	.init = senary_init,
>  	.unsol_event = snd_hda_jack_unsol_event,
>  	.suspend = senary_suspend,
> +	.resume = senary_resume,
>  	.check_power_status = snd_hda_gen_check_power_status,
>  	.stream_pm = snd_hda_gen_stream_pm,
>  };
> -- 
> 2.25.1
>