[alsa-devel] [PATCH v1 5/5] ALSA: hda - Add DP-MST support for NVIDIA codecs

[Takashi Iwai]

On Thu, 14 Nov 2019 12:50:29 +0100,
Nikhil Mahale wrote:
> 
> On 11/14/19 4:32 PM, Takashi Iwai wrote:
> > On Thu, 14 Nov 2019 04:37:04 +0100,
> > Nikhil Mahale wrote:
> >>
> >> @@ -3494,11 +3500,86 @@ static const struct drm_audio_component_audio_ops nvhdmi_audio_ops = {
> >>  	.master_unbind = generic_acomp_master_unbind,
> >>  };
> >>  
> >> +static int nvhdmi_find_pcm_slot(struct hdmi_spec *spec,
> >> +				struct hdmi_spec_per_pin *per_pin)
> >> +{
> >> +	int i;
> >> +
> >> +	/*
> >> +	 * generic_hdmi_build_pcms() allocates (num_nids + dev_num - 1)
> >> +	 * number of pcms.
> >> +	 *
> >> +	 * The per_pin of pin_nid_idx=n and dev_id=m prefers to get pcm-n if m==0.
> >> +	 * This guarantees that dynamic pcm assignments are compatible with the
> >> +	 * legacy static per_pin-pmc assignment that existed in the days before
> >> +	 * DP-MST.
> >> +	 *
> >> +	 * per_pin of m!=0 prefers to get pcm=(num_nids + (m - 1)).
> >> +	 */
> >> +	if (per_pin->dev_id == 0 &&
> >> +	    !test_bit(per_pin->pin_nid_idx, &spec->pcm_bitmap))
> >> +		return per_pin->pin_nid_idx;
> >> +
> >> +	if (per_pin->dev_id != 0 &&
> >> +	    !(test_bit(spec->num_nids + (per_pin->dev_id - 1),
> >> +		&spec->pcm_bitmap))) {
> >> +		return spec->num_nids + (per_pin->dev_id - 1);
> >> +	}
> >> +
> >> +	/* have a second try; check the area over num_nids */
> >> +	for (i = spec->num_nids; i < spec->pcm_used; i++) {
> >> +		if (!test_bit(i, &spec->pcm_bitmap))
> >> +			return i;
> >> +	}
> >> +
> >> +	/* the last try; check the empty slots in pins */
> >> +	for (i = 0; i < spec->num_nids; i++) {
> >> +		if (!test_bit(i, &spec->pcm_bitmap))
> >> +			return i;
> >> +	}
> >> +	return -EBUSY;
> >> +}
> > 
> > I think this can be applied for Intel case, too.  No need for creating
> > yet another indirect branch.
> 
> Do you mean I should replace existing logic in hdmi_find_pcm_slot()
> by this new logic?

Yes.  It's a dynamic assignment for DP-MST in anyway, so the
compatibility of the assigned index shouldn't be a big matter.


Takashi