Re: [PATCH] arm64: dts: imx8mp: Enable HDMI on i.MX8MP DHCOM PDK2 and PDK3

[Alexander Stein]

Hi everyone,

Am Donnerstag, 16. Mai 2024, 13:46:33 CEST schrieb Ahmad Fatoum:
> On 16.05.24 13:44, Marek Vasut wrote:
> > On 5/16/24 1:36 PM, Ahmad Fatoum wrote:
> >> Hello Marek,
> > 
> > Hi,
> > 
> >> On 16.05.24 13:14, Marek Vasut wrote:
> >>> On 5/16/24 10:00 AM, Francesco Dolcini wrote:
> >>>> Hello Marek,
> >>>
> >>> Hi,
> >>>
> >>>> On Tue, May 14, 2024 at 03:06:42AM +0200, Marek Vasut wrote:
> >>>>> Enable HDMI output on i.MX8MP DHCOM PDK2 and PDK3. The I2C5 on PDK2 and
> >>>>> I2C mux port 1 on PDK3 respectively are used in regular I2C mode instead
> >>>>> of HDMI DDC mode to permit connection of other I2C devices on those buses.
> >>>>
> >>>> Are you able to read the HDMI EDID with such configuration? I have the
> >>>> patch ready for verdin imx8mp, I just did not have time to figure out
> >>>> this last details.
> >>>
> >>> Yes with ddc-i2c-bus in hdmi_tx{} node, no with ddc-i2c-bus in connector node. Maybe that's what you're running into ? The DW HDMI core needs the ddc-i2c-bus property in hdmi_tx{} node if you use non-native I2C bus for the DDC channel.
> >>
> >> What benefit does the hdmi-connector provide over just omitting it?
> >> Just for documentation purposes?
> > 
> > I was under the impression describing the hardware in DT in full was the best practice (TM), hence it is in full here.
> 
> Sure, I am just wondering what effect, if any, it has in how Linux interprets
> the device tree. I have an i.MX8MP board with HDMI as well, but without
> connector (yet).

AFAICT having a hdmi-connector has no effect at all. Linux drivers are
not using it.

Best regards,
Alexander
-- 
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
http://www.tq-group.com/