> -----Original Message----- > From: Guenter Roeck [mailto:groeck7@xxxxxxxxx] On Behalf Of Guenter > Roeck > Sent: 18 December 2013 17:24 > To: Opensource [Anthony Olech] > Cc: Jean Delvare; open list:HARDWARE MONITORING; Anton Vorontsov; > David Woodhouse; open list; David Dajun Chen > Subject: Re: [PATCH V1] fix adc to voltage calculation in da9052 > power driver > On Wed, Dec 18, 2013 at 03:45:32PM +0000, Opensource [Anthony Olech] > wrote: > > > -----Original Message----- > > > From: Jean Delvare [mailto:khali@xxxxxxxxxxxx] > > > Sent: 18 December 2013 15:33 > > > To: Opensource [Anthony Olech] > > > Cc: Anton Vorontsov; David Woodhouse; David Dajun Chen; open list; > > > open list:HARDWARE MONITORING > > > Subject: Re: [PATCH V1] fix adc to voltage calculation > > > in da9052 power driver On Wed, 18 Dec 2013 15:21:13 +0000, Anthony > > > Olech wrote: > > > > The ADC resolution of the PMIC is 10-bits, this means that the > > > > maximum possible value is 1023 and not the 1024 as in the code. > > > The conversion from register value to mV depends on the ADC's LSB, > > > not its range. So the maximum value which can be represented is > irrelevant. > > thanks for the speedy response, but the converted value returned by > > the function does depend on the scaling. > > For example take the two cases where value in 0x1F0, then the > > erroneous previous calculation yields 3469 and the new corrected > > calculation yields 3470 > > So please apply the patch as it truly fixes an error. > AFAICS the previous calculation should return 3468. Replacing the division by > 512 with DIV_ROUND_CLOSEST would return 3469, and the new calculation > would return 3470. > Still, an ADC would typically have an LSB. Question here is what that LSB is (in > mV), or in other words what the maximim voltage represented by 0x3ff is. > Current assumption is that it is 1.9531 mV or 2000/1024, which admittedly is > odd. It might be 2000/1023 = 1.9550 mV, or it might be 2 mV. the maximum voltage represented by 0x3ff is 2000 mV, so the existing code is incorrect and the patch is to correct the error. many thanks for your comment, Tony Olech > Thanks, > Guenter > > many thanks > > Tony Olech > > > > Signed-off-by: Anthony Olech > > > > <anthony.olech.opensource@xxxxxxxxxxx> > > > > This patch is relative to linux-next repository tag next-20131218 > > > > drivers/power/da9052-battery.c | 2 +- > > > > 1 file changed, 1 insertion(+), 1 deletion(-) diff --git > > > > a/drivers/power/da9052-battery.c b/drivers/power/da9052-battery.c > > > > index f8f4c0f..8f0f259 100644 > > > > --- a/drivers/power/da9052-battery.c > > > > +++ b/drivers/power/da9052-battery.c > > > > @@ -178,7 +178,7 @@ struct da9052_battery { static inline int > > > > volt_reg_to_mV(int value) { > > > > - return ((value * 1000) / 512) + 2500; > > > > + return DIV_ROUND_CLOSEST(value * 2000, 1023) + 2500; > > > > } > > > > static inline int ichg_reg_to_mA(int value) > > > Jean Delvare _______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
