Question regarding HDMI Audio support on i.MX6 with vanilla kernel

[Russell King - ARM Linux]

Hi Luís,

On Tue, Nov 07, 2017 at 04:53:02PM +0000, Luís Mendes wrote:
> Hi Russell,
> 
> I've just tested my Wandboard Rev.D1 with a Samsung monitor PX2370, model:
> LS23WHEKFV/EN which has HDMI with audio (has an audio pass-through with an
> optical IEC958 connector and an analog stereo audio jack which can be
> connected to an amplifier, despite the monitor has no speakers embedded to
> it) that provides no EDID information, thus I had to force the EDID data
> from my home TV for the audio to work.

What do you mean "provides no EDID information" - you mean there's no
EDID at all from the monitor, or it's just missing the HDMI vendor
block?  I'd find it hard to accept that there's no EDID information at
all from the monitor.

Can you provide me with whatever EDID information the monitor does give
please, preferably in binary format.  Thanks.

> This made me think the kernel option has_audio={-1,0,1} would be
> appropriate for these cases. I know that you don't find this idea so good,
> but for these cases would be nice. If you want I can provide a patch for
> this.
> What do you say?

I disagree - if the monitor supports audio, but doesn't advertise in
the EDID that it supports audio, it isn't a source device problem,
and it isn't a source driver problem either.  It's a generic problem.

Consider if you connect this monitor to some other device, such as a
PC running Intel graphics - should the i915 driver also spring up a
"has_audio" option as well?

Would it not be more sensible for this to be handled generically
within DRM?

Regards,

-- 
RMK's Patch system: http://www.armlinux.org.uk/developer/patches/
FTTC broadband for 0.8mile line in suburbia: sync at 8.8Mbps down 630kbps up
According to speedtest.net: 8.21Mbps down 510kbps up