Re: [RFC PATCH 4/5] ASoC: SOF: Intel: hda: add SoundWire stream config/free callbacks

[Pierre-Louis Bossart]

Thanks for the review Guennadi

> > +static int sdw_config_stream(void *arg, void *s, void *dai,
> > +			     void *params, int link_id, int alh_stream_id)
>
> I realise, that these function prototypes aren't being introduced by these
> patches, but just wondering whether such overly generic prototype is really
> a good idea here, whether some of those "void *" pointers could be given
> real types. The first one could be "struct device *" etc.

In this case the 'arg' parameter is actually a private 'struct 
snd_sof_dev', as shown below [1]. We probably want to keep this 
relatively opaque, this is a context that doesn't need to be exposed to 
the SoundWire code.

The dai and params are indeed cases where we could use stronger types, 
they are snd_soc_dai and hw_params respectively. I don't recall why the 
existing code is this way, Vinod and Sanyog may have the history of this.

> > +{
> > +	struct snd_sof_dev *sdev = arg;
> > +	struct snd_soc_dai *d = dai;
[1]

> > +	struct sof_ipc_dai_config config;
> > +	struct sof_ipc_reply reply;
> > +	int ret;
> > +	u32 size = sizeof(config);
> > +
> > +	memset(&config, 0, size);
> > +	config.hdr.size = size;
> > +	config.hdr.cmd = SOF_IPC_GLB_DAI_MSG | SOF_IPC_DAI_CONFIG;
> > +	config.type = SOF_DAI_INTEL_ALH;
> > +	config.dai_index = (link_id << 8) | (d->id);
> > +	config.alh.stream_id = alh_stream_id;
>
> Entirely up to you, in such cases I usually do something like
>
> +	struct sof_ipc_dai_config config = {
> +		.type = SOF_DAI_INTEL_ALH,
> +		.hre = {
> +			.size = sizeof(config),
> +			.cmd = SOF_IPC_GLB_DAI_MSG | SOF_IPC_DAI_CONFIG,
> +			...
>
> which then also avoids a memset(). But that's mostly a matter of personal
> preference, since this is on stack, the compiler would probably internally
> anyway translate the above initialisation to a memset() with all the
> following assignments.

I have no preference, so in this case I will go with consistency with 
existing code, which uses the suggested style for all IPCs.

> > +
> > +	/* send message to DSP */
> > +	ret = sof_ipc_tx_message(sdev->ipc,
> > +				 config.hdr.cmd, &config, size, &reply,
> > +				 sizeof(reply));
> > +	if (ret < 0) {
> > +		dev_err(sdev->dev,
> > +			"error: failed to set DAI hw_params for link %d dai->id %d ALH %d\n",
>
> Are readers really expected to understand what "dai->id" means? Wouldn't
> "DAI ID" be friendlier, although I understand you - who might not know
> what "x->y" stands for?.. ;-)

I was trying to avoid a confusion here, we have config->dai_index which 
are shared concepts between topology and firmware, and dai->id which is 
shared between topology and machine driver (which refers to the dai in 
the dai_link which has its own .id). In topology files we have the three 
indices and of course after a couple of weeks I can't recall which one 
maps to what.
I am afraid DAI ID might be confused with dai_index. If there are 
suggestions on this I am all ears, all I care about is avoiding 
ambiguity and having to ask Ranjani what index this really is :-)