Re: [PATCH v2 3/4] ASoC: Intel: atom: Use scoped_guard() for scoped locking

[Cezary Rojewski <cezary.rojewski@intel.com>]

On 6/18/2026 1:06 PM, phucduc.bui@gmail.com wrote:
> From: bui duc phuc <phucduc.bui@gmail.com>
> 
> Convert mutex-protected sections to scoped_guard() helpers and
> simplify cleanup paths. Also use __free(kfree) for the temporary
> stream pointer in sst_media_open().
> 
> No functional change intended.

...

> diff --git a/sound/soc/intel/atom/sst-mfld-platform-pcm.c b/sound/soc/intel/atom/sst-mfld-platform-pcm.c
> index f074af2499c8..ce2814e206e1 100644
> --- a/sound/soc/intel/atom/sst-mfld-platform-pcm.c
> +++ b/sound/soc/intel/atom/sst-mfld-platform-pcm.c
> @@ -304,7 +304,7 @@ static int sst_media_open(struct snd_pcm_substream *substream,
>   {
>   	int ret_val = 0;
>   	struct snd_pcm_runtime *runtime = substream->runtime;
> -	struct sst_runtime_stream *stream;
> +	struct sst_runtime_stream *stream __free(kfree) = NULL;

No mixing of cleanup classes.  Cleaning locking != cleaning pointers.
One action per patch.

...

> @@ -347,13 +346,10 @@ static int sst_media_open(struct snd_pcm_substream *substream,
>   	snd_pcm_hw_constraint_step(substream->runtime, 0,
>   			   SNDRV_PCM_HW_PARAM_PERIODS, 2);
>   
> +	stream = NULL;
> +

Is this needed?  If you believe so, then I'd rather see it placed 
immediately after any operation that could cause the variable to leave 
the scope.

>   	return snd_pcm_hw_constraint_integer(runtime,
>   			 SNDRV_PCM_HW_PARAM_PERIODS);
> -out_ops:
> -	mutex_unlock(&sst_lock);
> -out_power_up:
> -	kfree(stream);
> -	return ret_val;
>   }
>   
>   static void sst_media_close(struct snd_pcm_substream *substream,