Re: [PATCH 5/6] mfd: TPS65910: Fix tps65910_set_voltage

[Kyle Manna <kyle.manna@fuel7.com>]

Hi Mark,

I've reviewed the patch and the core regulator framework after reading 
your comments.  This patch worked around a subtle bug that I didn't 
notice and is hence not needed.  The root cause was fixed and the 
voltage selectors work as desired.

Will update the patch series and resend.

- Kyle

On 10/19/2011 08:32 AM, Mark Brown wrote:
> On Tue, Oct 18, 2011 at 01:26:27PM -0500, Kyle Manna wrote:
>
> *Always* CC maintainers on patches.
>
>> Previously tps65910_set_voltage() only selected from a fixed number of
>> voltages.  Rename that function to tps65910_set_voltage_sel(). Do the
>> same for tps65911_set_voltage().
> What is the issue being fixed here?  This looks like a stylistic change
> rateher than a bug fix.
>
>> Also add a tps65910_set_voltage that works with the regulator framework
>> and applies the correct voltage with apply_uv is set in the regulator's
>> constraints.
> So this is adding support for a new chip?  Whatever the answer it's
> clearly a distinct change from the above and should therefore be a
> separate patch.
>
>> +	/* Pick the nearest selector */
>> +	for (i = 0; i<  tps65910_regs[id].table_len; i++) {
>> +		new_uV = tps65910_regs[id].table[i] * 1000;
>> +
>> +		if (new_uV>= min_uV&&  new_uV<= max_uV&&
>> +		   (abs(new_uV - midpoint)<  abs(selected_uV - midpoint))) {
>> +			*selector = i;
>> +			selected_uV = tps65910_regs[id].table[i] * 1000;
>> +		}
>> +	}
> This looks wrong, the expected behaviour for the regulator API is that
> the driver will pick the minimum voltage within the range.  Why is this
> being done?