Re: [PATCH 2/2] ALSA: hda - Add driver for Tegra SoC HDA

[Thierry Reding <thierry.reding-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>]

[-- Attachment #1: Type: text/plain, Size: 1317 bytes --]

On Wed, Apr 09, 2014 at 11:40:34AM +0200, Takashi Iwai wrote:
> At Wed, 9 Apr 2014 11:22:13 +0200,
> Thierry Reding wrote:
> > 
> > > +
> > > +static int probe_mask[SNDRV_CARDS] = {[0 ... (SNDRV_CARDS-1)] = -1};
> > > +module_param_array(probe_mask, int, NULL, 0444);
> > > +MODULE_PARM_DESC(probe_mask, "Bitmask to probe codecs (default = -1).");
> > 
> > Is this really necessary? Do we need this with Tegra? Can't we simply
> > assume that there's always only one codec? Or always use a probe_mask of
> > -1? Alternatively, wouldn't this be better off in DT?
> 
> This is an option rather for debugging.  Usually the codec is
> recognized automatically in the case of HD-audio.
> 
> But, another point is, suppose this is really specific to Tegra and
> nothing else, whether we need multiple instances.  If there is only a
> single controller, we can omit the arrays.

Yes, I think it's reasonable to assume that there will only ever be a
single HDA controller on Tegra SoCs.

One other thought: I can see how this option would be useful for
debugging things on a PC for instance, where the configurations can vary
fairly drastically, but I'm having a hard time to imagine this ever
being necessary for Tegra.

But if people prefer to keep this I have no strong objections.

Thierry

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]