| Message ID | 1436173112-21397-10-git-send-email-lee.jones@linaro.org |
|---|---|
| State | New |
| Headers | show |
On Tue, 07 Jul 2015, Mark Brown wrote: > On Mon, Jul 06, 2015 at 09:58:31AM +0100, Lee Jones wrote: > > If we reverse some of the logic and change the formula used, > > we can simplify the function greatly. > > > +static int pwm_voltage_to_duty_cycle(struct regulator_dev *rdev, int req_uV) > > { > > - struct pwm_regulator_data *drvdata = rdev_get_drvdata(rdev); > > You just added this function in the previous patch? You're right, it does look a little weird contained in a single patch-set. The submission in the previous patch is the tried and tested (i.e. in real releases) method written by ST. This patch contains a simplification provided by me. IMO it looks and performs better, but doesn't have the same time-under-test that the original method does. I'm merely ensuring we keep some history in order so provide and easy way back i.e. revert. If I have any say at all, I'd really like to keep this piece of history.
On Tue, 07 Jul 2015, Mark Brown wrote: > On Tue, Jul 07, 2015 at 02:45:03PM +0100, Lee Jones wrote: > > On Tue, 07 Jul 2015, Mark Brown wrote: > > > > You just added this function in the previous patch? > > > You're right, it does look a little weird contained in a single > > patch-set. The submission in the previous patch is the tried and > > tested (i.e. in real releases) method written by ST. This patch > > contains a simplification provided by me. IMO it looks and performs > > better, but doesn't have the same time-under-test that the original > > method does. I'm merely ensuring we keep some history in order so > > provide and easy way back i.e. revert. > > > If I have any say at all, I'd really like to keep this piece of > > history. > > OK, that makes sense - can you put the above in the commit message > please? Sure.
diff --git a/drivers/regulator/pwm-regulator.c b/drivers/regulator/pwm-regulator.c index b37b616..d5cb267 100644 --- a/drivers/regulator/pwm-regulator.c +++ b/drivers/regulator/pwm-regulator.c @@ -93,26 +93,13 @@ static int pwm_regulator_list_voltage(struct regulator_dev *rdev, /** * Continuous voltage call-backs */ -static int pwm_voltage_to_duty_cycle(struct regulator_dev *rdev, - int volt_mV) +static int pwm_voltage_to_duty_cycle(struct regulator_dev *rdev, int req_uV) { - struct pwm_regulator_data *drvdata = rdev_get_drvdata(rdev); - int min_mV = rdev->constraints->min_uV / 1000; - int max_mV = rdev->constraints->max_uV / 1000; - int max_duty_cycle = drvdata->max_duty_cycle; - int vdiff = min_mV - max_mV; - int pwm_code; - int tmp; - - tmp = max_duty_cycle - min_mV * max_duty_cycle / vdiff; - pwm_code = tmp + max_duty_cycle * volt_mV / vdiff; - - if (pwm_code < 0) - pwm_code = 0; - if (pwm_code > max_duty_cycle) - pwm_code = max_duty_cycle; - - return pwm_code * 100 / max_duty_cycle; + int min_uV = rdev->constraints->min_uV; + int max_uV = rdev->constraints->max_uV; + int diff = max_uV - min_uV; + + return 100 - ((((req_uV * 100) - (min_uV * 100)) / diff)); } static int pwm_regulator_get_voltage(struct regulator_dev *rdev) @@ -131,7 +118,7 @@ static int pwm_regulator_set_voltage(struct regulator_dev *rdev, int duty_cycle; int ret; - duty_cycle = pwm_voltage_to_duty_cycle(rdev, min_uV / 1000); + duty_cycle = pwm_voltage_to_duty_cycle(rdev, min_uV); ret = pwm_config(drvdata->pwm, (drvdata->pwm->period / 100) * duty_cycle,
If we reverse some of the logic and change the formula used, we can simplify the function greatly. Signed-off-by: Lee Jones <lee.jones@linaro.org> --- drivers/regulator/pwm-regulator.c | 27 +++++++-------------------- 1 file changed, 7 insertions(+), 20 deletions(-)