Re: [alsa-devel] [PATCH 2/4] drm: bridge: dw-hdmi: Report connector status using callback
From: Cheng-yi Chiang
Date: Sat Jul 06 2019 - 06:18:31 EST
On Sat, Jul 6, 2019 at 1:16 AM Mark Brown <broonie@xxxxxxxxxx> wrote:
>
> On Fri, Jul 05, 2019 at 03:31:24PM +0800, Cheng-yi Chiang wrote:
>
> > It was a long discussion.
> > I think the conclusion was that if we are only talking about
> > hdmi-codec, then we just need to extend the ops exposed in hdmi-codec
> > and don't need to use
> > hdmi-notifier or drm_audio_component.
>
> What I'd picked up from the bits of that discussion that I
> followed was that there was some desire to come up with a unified
> approach to ELD notification rather than having to go through
> this discussion repeatedly? That would certianly seem more
> sensible. Admittedly it was a long thread with lots of enormous
> mails so I didn't follow the whole thing.
Hi Mark, thanks for following the long thread.
The end of the discussion was at
https://lkml.org/lkml/2019/6/20/1397
Quoted from Daniel's suggestion:
"
I need to think about this more, but if all we need to look at is
hdmi_codec, then I think this becomes a lot easier. And we can ignore
drm_audio_component.h completely.
"
My thought is that the codec driver under ASoC are only these two:
hdac_hdmi.c and hdmi-codec.c ( forgive me if I missed others. I just
grep "hdmi" under sound/soc/codecs/ )
hdac_hdmi.c is like a wrapper for HDA and drm_audio_components.
hdmi-codec.c is the only other codec driver that cares HDMI under ASoC.
So adding the jack/eld support at hdmi-codec driver can cover the
existing use cases for HDMI codec driver in ASoC.
That said, adding a new unified approach for Jack/ELD notification
that will only be used by hdmi-codec seems not a high priority.
Hope this explanation helps your decision.
Thanks!