Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1318126
| From | Laxman Dewangan <ldewangan@nvidia.com> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [PATCH V4 1/5] DT: mfd: add device-tree binding doc fro PMIC max77620/max20024 |
| Date | 2016-01-26 17:40 +0100 |
| Message-ID | <qVeOT-6K1-15@gated-at.bofh.it> (permalink) |
| References | <qSBRE-6uw-3@gated-at.bofh.it> <qSBRE-6uw-9@gated-at.bofh.it> <qUNYp-3kJ-73@gated-at.bofh.it> <qURfz-5VH-1@gated-at.bofh.it> <qVdg6-5z6-23@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
On Tuesday 26 January 2016 08:28 PM, Lee Jones wrote:
> On Mon, 25 Jan 2016, Laxman Dewangan wrote:
>
>> Hmm. I describe the boolean and tristate only. Do I need to define
>> type for integer,string also?
> I wouldn't describe any of them. It's normally pretty obvious which
> properties are boolean by the lack of required cell description.
>
Rob suggested to use the type also for Boolean and tristate. I think it
is good to have for all places that what type of values are valid.
>>>> +The property for fps child nodes as:
>>>> +Required properties:
>>>> + -reg: FPS number like 0, 1, 2 for FPS0, FPS1 and FPS2 respectively.
>>> I'm surprised Rob Acked this. We don't usually do device numbers in DT.
>> What is best way to make the child node for FPS and differentiate FPS0,1, 2?
>> What is your suggestion here?
> There are lots of ways you can solve this and so many examples of
> others doing so. I suggest you have a look at some DTS files and
> figure it out. One possible solution is to use different compatible
> strings.
Here, I think I can go similar to regulators where child node name
identifies the regulators.
fps-config {
fps0 {
maxim,fps-time-period-us = <1280>;
maxim,fps-enable-input = <FPS_EN_SRC_EN0>;
};
fps1 {
maxim,fps-time-period-us = <2560>;
maxim,fps-enable-input = <FPS_EN_SRC_EN1>;
};
fps2 {
maxim,fps-time-period-us = <640>;
maxim,fps-enable-input = <FPS_EN_SRC_SW>;
};
};
So node name gives the FPS name.
>>> +Pinmux and GPIO:
>>> +===============
>>> I think this whole section needs moving to ../pinctrl and needs to be
>>> reviewed by Linus W.
>> Is this mean I need to create DT binding doc for the each subsystem
>> differently?
>> Actually during AS3722, I had different understanding to have single file.
> Yes, that way you have each of the the subsystem experts review your
> documentation. You can then link to them from this document.
OK, I will add different dt binding doc in respective driver and squash
that with respected submodule drivers.
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[PATCH V4 1/5] DT: mfd: add device-tree binding doc fro PMIC max77620/max20024 Laxman Dewangan <ldewangan@nvidia.com> - 2016-01-19 11:40 +0100
Re: [PATCH V4 1/5] DT: mfd: add device-tree binding doc fro PMIC max77620/max20024 Lee Jones <lee.jones@linaro.org> - 2016-01-25 13:00 +0100
Re: [PATCH V4 1/5] DT: mfd: add device-tree binding doc fro PMIC max77620/max20024 Laxman Dewangan <ldewangan@nvidia.com> - 2016-01-25 16:30 +0100
Re: [PATCH V4 1/5] DT: mfd: add device-tree binding doc fro PMIC max77620/max20024 Lee Jones <lee.jones@linaro.org> - 2016-01-26 16:00 +0100
Re: [PATCH V4 1/5] DT: mfd: add device-tree binding doc fro PMIC max77620/max20024 Laxman Dewangan <ldewangan@nvidia.com> - 2016-01-26 17:40 +0100
Re: [PATCH V4 1/5] DT: mfd: add device-tree binding doc fro PMIC max77620/max20024 Lee Jones <lee.jones@linaro.org> - 2016-01-27 08:30 +0100
csiph-web