Re: [PATCH 2/4] ALSA: hda - add DP mst verb support

[Takashi Iwai <tiwai@suse.de>]

On Tue, 08 Mar 2016 07:54:44 +0100,
Yang, Libin wrote:
> 
> Hi Takashi,
> 
> > -----Original Message-----
> > From: Takashi Iwai [mailto:tiwai@suse.de]
> > Sent: Tuesday, March 08, 2016 12:46 AM
> > To: libin.yang@linux.intel.com
> > Cc: alsa-devel@alsa-project.org; Lin, Mengdong; Yang, Libin
> > Subject: Re: [alsa-devel] [PATCH 2/4] ALSA: hda - add DP mst verb
> > support
> > 
> > On Mon, 07 Mar 2016 15:57:44 +0100,
> > libin.yang@linux.intel.com wrote:
> > >
> > > +int snd_hda_set_dev_select(struct hda_codec *codec, hda_nid_t nid,
> > int dev_id)
> > > +{
> > > +	int ret, dev;
> > > +	unsigned long timeout;
> > > +
> > > +	/* not support dp_mst will always return 0, using first dev_entry
> > */
> > > +	if (!codec->dp_mst)
> > > +		return 0;
> > > +
> > > +	/* If Device List Length is 0, the pin is not multi stream capable.
> > > +	 */
> > > +	if ((snd_hda_get_num_devices(codec, nid) + 1) == 0)
> > > +		return 0;
> > 
> > I don't understand what this code is for.  Why +1?
> 
> Sorry, it should be if ((snd_hda_get_num_devices(codec, nid)) == 0).
> 0 means not MST capable. Thanks for point it out. 
> 
> > 
> > > +
> > > +	/* Behavior of setting index being equal to or greater than
> > > +	 * Device List Length is not predictable
> > > +	 */
> > > +	if ((snd_hda_get_num_devices(codec, nid) + 1) <= dev_id)
> > > +		return 0;
> > 
> > Ditto.  And, calling the same function twice?
> 
> (snd_hda_get_num_devices(codec, nid) + 1) means the device entry number.
> If the (snd_hda_get_num_devices(codec, nid) + 1) == 3, it means there are
> only 3 device entries. And dev_id being 0, 1, 2 is legal. It dev_id is 3 or greater,
> it is illegal. 

So, shouldn't it be like below?
    if (dev_id >= num_devices)
        return 0;

Also my point is that you don't need to call snd_hda_get_num_devices()
twice.  It can be stored locally.


Takashi