[PATCH RFC 1/3] DT: bindings: mmc: Add property for 3.3V only support

[Shawn Lin]

+ Adrian

Let's queue Adrian here who now maintains SDHCI stuff.

On 2016/8/11 5:39, Rob Herring wrote:
> On Wed, Aug 10, 2016 at 3:27 PM, Stefan Wahren <stefan.wahren at i2se.com> wrote:
>> Hi,
>>
>>> Rob Herring <robh at kernel.org> hat am 10. August 2016 um 20:44 geschrieben:
>>>
>>>
>>> On Sat, Aug 06, 2016 at 12:55:38PM +0000, Stefan Wahren wrote:
>>>> Currently there is no proper way to define that a MMC host supports
>>>> only 3.3V. The property no-1-8-v is broken and has different meanings
>>>> for different sdhci variants. So add a new property for 3.3V only
>>>> support and mark no-1-8-v as deprecated.
>>>
>>> Why is it broken?
>>
>> i want to quote Ulf Hansson here [1]:
>>
>>   The problem with the "no-1-8-v" binding is that it's describing what
>>   the hardware *can't* do. It thus becomes easy to abuse it.
>
> Sounds like a quirk which is perfectly normal for a property. I'd
> agree it should be what the h/w *can* do if h/w didn't have capability
> bit that does that.
>
>>   I suggest we stop using it, we should mark it deprecated.
>>
>> [1] - http://www.spinics.net/lists/linux-mmc/msg34604.html
>>
>>> How would I override a controller saying 1.8V is
>>> supported and it is not?
>>
>> Sorry, i'm not sure that i understand your question. In case a board or a MMC
>> controller doesn't support 1.8V, it usually supports only 3.3V which is the
>> intension of this patch.
>
> Some controllers have capability bits saying what voltages they
> support, right? And those bits can be wrong (unless firmware sets them
> I'd expect that is the common case) which as I read it is what
> "no-1-8-v" was for. So with the "mmc-ddr-*v" properties, what does not
> present mean and how do they relate to controller capability bits? I
> assume they would override the controller bits, but you can only
> override the capability bit not set case. I would think the property
> not present means use the capability bit, not that that voltage is not
> supported. I think you probably need tri-state properties here where
> value 1 means supported, 0 means not supported, and not present means
> use capability bit (or other method).
>
> Rob
> --
> To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
> the body of a message to majordomo at vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>


-- 
Best Regards
Shawn Lin