Re: [PATCH RFC 1/5] drm/msm/hdmi: deprecate non standard gpio properties.

[Rob Clark]

On Mon, Aug 10, 2015 at 6:27 PM, Bjorn Andersson
<bjorn.andersson@xxxxxxxxxxxxxx> wrote:
> On Mon 10 Aug 15:07 PDT 2015, Rob Herring wrote:
>
> [..]
>> >> -- qcom,hdmi-tx-ddc-clk-gpio: ddc clk pin
>> >> -- qcom,hdmi-tx-ddc-data-gpio: ddc data pin
>> >> -- qcom,hdmi-tx-hpd-gpio: hpd pin
>> >> +- qcom,hdmi-tx-ddc-clk-gpios: ddc clk pin
>> >> +- qcom,hdmi-tx-ddc-data-gpios: ddc data pin
>>
>> The driver doesn't appear to do anything other than set these gpios
>> high. Presumably you would ultimately do a bit-bang i2c bus for these?
>> At that point, aren't you going to want to use the i2c-gpio binding?
>>
>> I think this binding has bigger issues like why is it not using the
>> hdmi-connector binding which has a standard "hpd-gpios" property
>> already. I can't really single out this binding though. There's a lot
>> of crap when it comes to DRM related bindings.
>>
>
> Shouldn't these pins just be muxed to the hdmi block in pinctl?
>
> I know Rob had something wrt the detect pin (hdp), but the others should
> never be gpios(?) Isn't this just a reminder of the codeaurora tree
> gpio_requesting all pins "just to be safe"?

yeah, other than hdp pin (since the debounce logic in hdmi block
doesn't seem to be bulletproof), others we only request.. rest is
echo's of codeaurora..

some of it is probably not needed upstream.. and I'm ok w/ just not
documenting that in the bindings doc, or documenting as "unsupported
and going to go away so don't use", or whatever makes the most sense..
but would like to keep the driver code alive for the time being, since
it is easier for me to maintain a bit of cruft or dead code in the
upstream driver, than dealing with it on the backport (and I'm
inevitably the one who has to do both)

(Good news is that srini mentioned he was working on audio for
8064/ifc6410 upstream, so it at least seems like there is an end in
sight for backports to 3.4 device kernel..  3.10 stuff, I think we'll
have to live with a bit longer..)

BR,
-R


>> >> +- qcom,hdmi-tx-hpd-gpios: hpd pin
>> >>  - core-vdda-supply: phandle to supply regulator
>> >>  - hdmi-mux-supply: phandle to mux regulator
>> >>
>> >> +- qcom,hdmi-tx-ddc-clk-gpio: (deprecated) use
>> >> +     "qcom,hdmi-tx-ddc-clk-gpios" instead
>> >> +- qcom,hdmi-tx-ddc-data-gpio: (deprecated) use
>> >> +     "qcom,hdmi-tx-ddc-data-gpios" instead
>> >> +- qcom,hdmi-tx-hpd-gpio: (deprecated) use
>> >> +     "qcom,hdmi-tx-hpd-gpios" instead
>> >> +
>> >>  Optional properties:
>> >> -- qcom,hdmi-tx-mux-en-gpio: hdmi mux enable pin
>> >> -- qcom,hdmi-tx-mux-sel-gpio: hdmi mux select pin
>> >> +- qcom,hdmi-tx-mux-en-gpios: hdmi mux enable pin
>> >> +- qcom,hdmi-tx-mux-sel-gpios: hdmi mux select pin
>> >> +
>> >> +- qcom,hdmi-tx-mux-en-gpio: (deprecated) use "qcom,hdmi-tx-mux-en-gpios"
>> >> +     instead
>> >> +- qcom,hdmi-tx-mux-sel-gpio: (deprecated) use "qcom,hdmi-tx-mux-sel-gpio"
>> >> +     instead
>> >>  - pinctrl-names: the pin control state names; should contain "default"
>> >>  - pinctrl-0: the default pinctrl state (active)
>> >>  - pinctrl-1: the "sleep" pinctrl state
>> >
>> > I don't see much use in listing that these properties are deprecated. We
>> > already have code to catch the deprecated names, so having them in the
>> > binding will at best be distracting.
>> >
>> > Anyway, I don't know if there's been any advice on this from the device
>> > tree bindings maintainers, so adding devicetree@xxxxxxxxxxxxxxx for
>> > visibility.
>>
>> If they need to be maintained, then marking them deprecated is
>> perfectly fine IMO. Also, if the maintainers of platforms using this
>> are okay with it, you can just switch it. Given there are no in tree
>> dts files using this, it can be argued that there is no ABI.
>>
>
> Part of some dev trees floating around no-one should have used these. So
> I think we should just document the proper ones and have the drivers
> support Rob's old properties until he's comfortable dropping them.
>
> That way we have a documented way forward and Rob can run his backports
> of this code without forking it.
>
> Regards,
> Bjorn
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/