Re: [PATCH v2 1/2] mfd: intel_quark_i2c_gpio: Use dmi_system_id table for retrieving frequency

[Jan Kiszka]

On 2017-05-22 19:26, Andy Shevchenko wrote:
> On Mon, May 22, 2017 at 8:25 PM, Jan Kiszka <jan.kiszka@xxxxxxxxxxx> wrote:
>> On 2017-05-22 19:20, Andy Shevchenko wrote:
>>> On Mon, May 22, 2017 at 8:18 PM, Jan Kiszka <jan.kiszka@xxxxxxxxxxx> wrote:
>>>> On 2017-05-22 19:12, Andy Shevchenko wrote:
>>>>> On Mon, May 22, 2017 at 1:53 PM, Jan Kiszka <jan.kiszka@xxxxxxxxxxx> wrote:
>>>>>> Avoids reimplementation of DMI matching in intel_quark_i2c_setup.
>>>>>
>>>>> What's wrong with current approach? I suppose this will make sense
>>>>> when we will have an issue / impediment. Right now it looks like
>>>>> over-engineering.
>>>>>
>>>>> (Yes, I know what I said to you about this earlier, OTOH see above)
>>>>>
>>>>
>>>> It allows matching on additional DMI tags without additional code -
>>>> patch 2 depends on it.
>>>
>>> And since there is no difference to the frequency the name is enough.
>>> So, I wouldn't go with this series as is. See above.
>>
>> Nope: Just like for the stmmac, we need to include the asset tags to
>> avoid matching variations of the devices which may carry the same board
>> name. While I will try to avoid that this happens, we are better safe
>> than sorry here.
> 
> Do we have an issue right now?
> Yes / No

Andy, we are trying to design a robust upstream driver here, no ad-hoc
BSP that will not survive the hardware anyway.

Jan

-- 
Siemens AG, Corporate Technology, CT RDA ITP SES-DE
Corporate Competence Center Embedded Linux