Re: [PATCH 04/23] ASoC: soc-compress: Implement trigger FE-BE sequencing as with normal PCMs

[Mark Brown <broonie@kernel.org>]

[-- Attachment #1: Type: text/plain, Size: 1233 bytes --]

On Fri, Sep 04, 2026 at 10:31:15AM +0300, Peter Ujfalusi wrote:

> The FE-BE trigger sequence should be dynamic, similarly how soc-pcm.c
> dpcm_fe_dai_do_trigger() does it.

> +static int soc_compr_trigger_fe(struct snd_compr_stream *cstream, int cmd)
> +{
> +	struct snd_soc_pcm_runtime *fe = cstream->private_data;
>  	int stream = cstream->direction; /* SND_COMPRESS_xxx is same as SNDRV_PCM_STREAM_xxx */
> +	bool fe_first;
>  	int ret;
>  
>  	if (cmd == SND_COMPR_TRIGGER_PARTIAL_DRAIN ||
>  	    cmd == SND_COMPR_TRIGGER_DRAIN)
>  		return snd_soc_component_compr_trigger(cstream, cmd);

What about SND_COMPR_TRIGGER_NEXT_TRACK...

> +	switch (cmd) {
> +	case SNDRV_PCM_TRIGGER_START:
> +	case SNDRV_PCM_TRIGGER_RESUME:
> +	case SNDRV_PCM_TRIGGER_PAUSE_RELEASE:
> +		ret = soc_compr_trigger_fe_be(cstream, cmd, fe_first);
> +		break;
> +	case SNDRV_PCM_TRIGGER_STOP:
> +	case SNDRV_PCM_TRIGGER_SUSPEND:
> +	case SNDRV_PCM_TRIGGER_PAUSE_PUSH:
> +		ret = soc_compr_trigger_fe_be(cstream, cmd, !fe_first);
> +		break;
> +	default:
> +		ret = -EINVAL;
> +		break;

...otherwise it'll end up with -EINVAL instead of getting passed
through.  Also a switch statement would be a bit more idiomatic.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]