Re: [PATCH v4 2/4] mmc: core: Add mmc_regulator_set_vqmmc()

[Mark Brown]

On Fri, Mar 20, 2015 at 11:55:50AM +0100, Ulf Hansson wrote:
> On 19 March 2015 at 12:36, Mark Brown <broonie@xxxxxxxxxx> wrote:

> > The implementation *should* do that anyway, it's just not trivial to
> > implement in an efficient fashion with the current information we have
> > from drivers.

> The APIs regulator_count_voltages() and regulator_list_voltage(), are
> currently used from the mmc core to find out which voltages that is
> supported (with 0.1V granularity). Then that information can be used
> when trying to set a new voltage.

> But I guess such a wrapper API is out of the question?

> Anyway, I get the feeling that we will need to do the same for this case.

As previously discussed the problem is that there can be a *lot* of
voltages on a modern regulator with fine grained voltage steps and
tolerances are also used for things like cpufreq where we care about
performance.  We need something that doesn't require a linear scan of
possible values.

> >> would be good to allow both upper and lower limits to be zero.

> > The lower limit can be zero already though it isn't clear to me that
> > this is useful.  Setting an upper limit of zero seems nonsensical, an
> > upper limit that is lower than the lower limit isn't terribly obvious
> > and removing the upper limit isn't safe - it means that we'll happily
> > oversupply things which is a road to physical damage.

> I am not sure I follow here. In the regulator_set_voltage_tol() you
> can only specifiy one limit (tolerance?). What Dough proposed was to
> add a new API which can have both a low tolerance value and a high
> tolerance value.

Tolerances are not limits - when you say "limit" that sounds like a hard
value.  I can't see any reason why the code would prevent anyone setting
a tolerance of zero, though it should be rare that this is actually the
best thing to do.

Attachment: signature.asc
Description: Digital signature