Re: [PATCH 21/24] ASoC: renesas: msiof: Fix msiof_close() being skipped on rollback

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] asoc: msiof: memory leak of prtd in msiof_open() error path
--

commit d7a4f2d9645c8240b2c60d42136e3b2e4694d703
Author: bui duc phuc <phucduc.bui@gmail.com>

ASoC: renesas: msiof: Fix msiof_close() being skipped on rollback

snd_pcm_hw_constraint_integer() may return 1, which is then returned from
msiof_open(). ASoC only marks the component as opened when .open() returns
0, so this patch explicitly returns 0 on success to ensure .close() isn't
skipped on rollback.

> diff --git a/sound/soc/renesas/rcar/msiof.c b/sound/soc/renesas/rcar/msiof.c
> index 128543fc4fc97..9fead9a22cdc4 100644
> --- a/sound/soc/renesas/rcar/msiof.c
> +++ b/sound/soc/renesas/rcar/msiof.c
> @@ -427,10 +427,12 @@ static int msiof_open(struct snd_soc_component *component,
>  	ret = snd_pcm_hw_constraint_integer(substream->runtime, SNDRV_PCM_HW_PARAM_PERIODS);
>  
>  open_err_dma:
> -	if (ret < 0)
> +	if (ret < 0) {
>  		dma_release_channel(chan);
> +		return ret;

[Severity: Medium]
Does this error path leak the dmaengine_pcm_runtime_data (prtd) memory?

In msiof_open(), if snd_dmaengine_pcm_open() succeeds, it allocates a prtd
object. If the subsequent call to snd_pcm_hw_constraint_integer() fails, the
function falls through to this open_err_dma error path.

Since this path calls dma_release_channel(chan) but fails to clean up the
PCM state via snd_dmaengine_pcm_close(substream) or
snd_dmaengine_pcm_close_release_chan(substream), is the prtd memory
permanently leaked when msiof_open() returns an error and the ASoC core
aborts the open sequence?

> +	}
>  
> -	return ret;
> +	return 0;
>  }
>  
>  static int msiof_close(struct snd_soc_component *component,

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261008103831.542533-1-phucduc.bui@gmail.com?part=21