Re: [PATCH] hda: Add support for 92HD73xxx codecs

[Takashi Iwai <tiwai@suse.de>]

At Fri, 13 Jun 2008 10:16:45 -0400,
Matthew Ranostay wrote:
> 
> Takashi Iwai wrote:
> > At Thu, 12 Jun 2008 15:19:31 -0400,
> > Matthew Ranostay wrote:
> >> @@ -2633,7 +2639,7 @@ static int stac92xx_auto_create_dmic_input_ctls(struct
> >> hda_codec *codec,
> >>  					      0,
> >>  					      AC_VERB_GET_CONFIG_DEFAULT,
> >>  					      0);
> >> -		if (get_defcfg_connect(def_conf) == AC_JACK_PORT_NONE)
> >> +		if (def_conf && (def_conf == AC_JACK_PORT_NONE))
> >>  			continue;
> > 
> > I guess the line def_conf = get_defcfg_connect(def_conf) is missing.
> > But, still I don't see why to change to that.
> > 
> Ok I meant this for the power management block(see below).
> Any port that has a jack should be included (settings 0 and 3) in the
> power management by nid.
> 
> > 
> >> @@ -3614,12 +3621,39 @@ static int patch_stac92hd71bxx(struct hda_codec *codec)
> > (snip)
> >> +	if (codec->vendor_id != 0x111d7608)
> >> +		snd_hda_sequence_write(codec, stac92hd71bxx_enable_portf);
> >> +	else {
> >> +		/* skip the VSW on port f */
> >> +		spec->num_pins--;
> >> +		stac92xx_set_config_reg(codec, 0xf, 0x40f000f0);
> >> +	}
> > 
> > These sequence wouldn't be executed in the resume.
> > snd_hda_sequence_write() should be replaced with *_cache().
> > stac92xx_set_config_reg() is a bit problem...
> > 
> Ok I totally forgot about suspend/resume issues.
> I'm not sure if the method I have below is acceptable to ALSA
> coding standards, but I'm sure you'll let me know :).

This looks good to me.  Thanks!


Takashi