Re: [PATCH v4 2/3] ARM: dts: Update the status of RTC DT node of Exynos5250 SoC to "okay"

From: Vikas Sajjan <vikas.sajjan@linaro.org>
To: Kukjin Kim <kgene.kim@samsung.com>
Cc: Tomasz Figa <tomasz.figa@gmail.com>,
	Tushar Behera <tushar.behera@linaro.org>,
	linux-samsung-soc <linux-samsung-soc@vger.kernel.org>,
	Patch Tracking <patches@linaro.org>,
	linaro-kernel@lists.linaro.org,
	Doug Anderson <dianders@chromium.org>,
	devicetree <devicetree@vger.kernel.org>,
	sunil joshi <joshi@samsung.com>
Subject: Re: [PATCH v4 2/3] ARM: dts: Update the status of RTC DT node of Exynos5250 SoC to "okay"
Date: Mon, 19 Aug 2013 13:48:45 +0530	[thread overview]
Message-ID: <CAD025ySAYRVWxJZubwDShOtwdRXKhg2JRUoPGmhaMmH31JVdTQ@mail.gmail.com> (raw)
In-Reply-To: <52111505.9010207@samsung.com>

Hi Kukjin,

On 19 August 2013 00:10, Kukjin Kim <kgene.kim@samsung.com> wrote:
> On 08/15/13 17:59, Tomasz Figa wrote:
>>
>> Hi Tushar,
>>
> [...]
>
>
>>>> +               status = "okay";
>>>>
>>>>          };
>>>
>>>
>>> Sometime back we had a discussion on this, the decision was to enable
>>> it in respective boards.
>>
>>
>> This is not entirely true.
>>
>> According to ePAPR, chapter 2.3.4, the status property has a well defined
>> meaning and it should be set to "disabled" when "the device is not
>> presently operational, but it might become operational in the future (for
>> example, something is not plugged in, or switched off)".
>>
> So in my understanding, you mean using "okay" is wrong and only "disabled"
> is used? and in board dt file? Already there are too many "okay"...
>
>
>> This means that unless setup of the device is missing something (e.g.
>> board-specific properties, like regulators or pin config) or there is a
>> valid technical reason for disabling the device by default (e.g. it needs
>> certain SoC pins to be properly connected to something), then such device
>> should be "okay", because it is operational.
>>
>>> Also if we are going ahead with this, we would need to remove the
>>> corresponding status statements from board files.
>>
>>
>> Yes, this is true.
>>
> According to above, probably we should add "disabled" in board dt file?
>

what Tushar meant was, since we made the status as "okay" in
exynos5250.dtsi itself, its better to remove the "okay" from the board
DTS files of exynos5250 ( exynos5250-snow.dts and
exynos5250-arndale.dts). Am I right Tushar?

> If I'm wrong, correct me.
>
> (+ dt ml)
>
> Anyway, I'm not sure how to use 'okay' and 'disabled' for status...
>
> I think, every hardware information should be defined in SoC dt file and
> maybe some of them could be set disabled or okay in each board dt file...
>
> Any suggestions?
>
> Thanks,
> Kukjin

-- 
Thanks and Regards
 Vikas Sajjan