1. Patchew
  2. linux
  3. [PATCH v9 0/3] Change PWM-controlled LED pin active mode and algorithm
  4. View patch

[PATCH v9 2/3] pwm: sifive: change the PWM algorithm

Nylon Chen posted 3 patches 1 year, 11 months ago
There is a newer version of this series
[PATCH v9 2/3] pwm: sifive: change the PWM algorithm
Posted by Nylon Chen 1 year, 11 months ago 
The `frac` variable represents the pulse inactive time, and the result
of this algorithm is the pulse active time.
Therefore, we must reverse the result.

The reference is SiFive FU740-C000 Manual[0]

Link: https://sifive.cdn.prismic.io/sifive/1a82e600-1f93-4f41-b2d8-86ed8b16acba_fu740-c000-manual-v1p6.pdf [0]

Co-developed-by: Zong Li <zong.li@sifive.com>
Signed-off-by: Zong Li <zong.li@sifive.com>
Co-developed-by: Vincent Chen <vincent.chen@sifive.com>
Signed-off-by: Vincent Chen <vincent.chen@sifive.com>
Signed-off-by: Nylon Chen <nylon.chen@sifive.com>
---
 drivers/pwm/pwm-sifive.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/pwm/pwm-sifive.c b/drivers/pwm/pwm-sifive.c
index eabddb7c7820..a586cfe4191b 100644
--- a/drivers/pwm/pwm-sifive.c
+++ b/drivers/pwm/pwm-sifive.c
@@ -110,9 +110,10 @@ static int pwm_sifive_get_state(struct pwm_chip *chip, struct pwm_device *pwm,
 				struct pwm_state *state)
 {
 	struct pwm_sifive_ddata *ddata = pwm_sifive_chip_to_ddata(chip);
-	u32 duty, val;
+	u32 duty, val, inactive;
 
-	duty = readl(ddata->regs + PWM_SIFIVE_PWMCMP(pwm->hwpwm));
+	inactive = readl(ddata->regs + PWM_SIFIVE_PWMCMP(pwm->hwpwm));
+	duty = (1U << PWM_SIFIVE_CMPWIDTH) - 1 - inactive;
 
 	state->enabled = duty > 0;
 
@@ -123,7 +124,7 @@ static int pwm_sifive_get_state(struct pwm_chip *chip, struct pwm_device *pwm,
 	state->period = ddata->real_period;
 	state->duty_cycle =
 		(u64)duty * ddata->real_period >> PWM_SIFIVE_CMPWIDTH;
-	state->polarity = PWM_POLARITY_INVERSED;
+	state->polarity = PWM_POLARITY_NORMAL;
 
 	return 0;
 }
@@ -139,7 +140,7 @@ static int pwm_sifive_apply(struct pwm_chip *chip, struct pwm_device *pwm,
 	int ret = 0;
 	u32 frac;
 
-	if (state->polarity != PWM_POLARITY_INVERSED)
+	if (state->polarity != PWM_POLARITY_NORMAL)
 		return -EINVAL;
 
 	cur_state = pwm->state;
@@ -159,6 +160,7 @@ static int pwm_sifive_apply(struct pwm_chip *chip, struct pwm_device *pwm,
 	frac = DIV64_U64_ROUND_CLOSEST(num, state->period);
 	/* The hardware cannot generate a 100% duty cycle */
 	frac = min(frac, (1U << PWM_SIFIVE_CMPWIDTH) - 1);
+	frac = (1U << PWM_SIFIVE_CMPWIDTH) - 1 - frac;
 
 	mutex_lock(&ddata->lock);
 	if (state->period != ddata->approx_period) {
-- 
2.43.0
Re: [PATCH v9 2/3] pwm: sifive: change the PWM algorithm
Posted by Uwe Kleine-König 1 year, 10 months ago 
On Thu, Feb 22, 2024 at 04:12:30PM +0800, Nylon Chen wrote:
> The `frac` variable represents the pulse inactive time, and the result
> of this algorithm is the pulse active time.
> Therefore, we must reverse the result.
> 
> The reference is SiFive FU740-C000 Manual[0]
> 
> Link: https://sifive.cdn.prismic.io/sifive/1a82e600-1f93-4f41-b2d8-86ed8b16acba_fu740-c000-manual-v1p6.pdf [0]
> 
> Co-developed-by: Zong Li <zong.li@sifive.com>
> Signed-off-by: Zong Li <zong.li@sifive.com>
> Co-developed-by: Vincent Chen <vincent.chen@sifive.com>
> Signed-off-by: Vincent Chen <vincent.chen@sifive.com>
> Signed-off-by: Nylon Chen <nylon.chen@sifive.com>
> ---
>  drivers/pwm/pwm-sifive.c | 10 ++++++----
>  1 file changed, 6 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/pwm/pwm-sifive.c b/drivers/pwm/pwm-sifive.c
> index eabddb7c7820..a586cfe4191b 100644
> --- a/drivers/pwm/pwm-sifive.c
> +++ b/drivers/pwm/pwm-sifive.c
> @@ -110,9 +110,10 @@ static int pwm_sifive_get_state(struct pwm_chip *chip, struct pwm_device *pwm,
>  				struct pwm_state *state)
>  {
>  	struct pwm_sifive_ddata *ddata = pwm_sifive_chip_to_ddata(chip);
> -	u32 duty, val;
> +	u32 duty, val, inactive;
>  
> -	duty = readl(ddata->regs + PWM_SIFIVE_PWMCMP(pwm->hwpwm));
> +	inactive = readl(ddata->regs + PWM_SIFIVE_PWMCMP(pwm->hwpwm));
> +	duty = (1U << PWM_SIFIVE_CMPWIDTH) - 1 - inactive;
>  
>  	state->enabled = duty > 0;
>  
> @@ -123,7 +124,7 @@ static int pwm_sifive_get_state(struct pwm_chip *chip, struct pwm_device *pwm,
>  	state->period = ddata->real_period;
>  	state->duty_cycle =
>  		(u64)duty * ddata->real_period >> PWM_SIFIVE_CMPWIDTH;
> -	state->polarity = PWM_POLARITY_INVERSED;
> +	state->polarity = PWM_POLARITY_NORMAL;
>  
>  	return 0;
>  }
> @@ -139,7 +140,7 @@ static int pwm_sifive_apply(struct pwm_chip *chip, struct pwm_device *pwm,
>  	int ret = 0;
>  	u32 frac;
>  
> -	if (state->polarity != PWM_POLARITY_INVERSED)
> +	if (state->polarity != PWM_POLARITY_NORMAL)
>  		return -EINVAL;
>  
>  	cur_state = pwm->state;
> @@ -159,6 +160,7 @@ static int pwm_sifive_apply(struct pwm_chip *chip, struct pwm_device *pwm,
>  	frac = DIV64_U64_ROUND_CLOSEST(num, state->period);
>  	/* The hardware cannot generate a 100% duty cycle */

Is this still true now that we know that PWM_SIFIVE_PWMCMP is the
inactive time in a period? If you fix that, the same claim in the header
of the driver needs adaption, too.

>  	frac = min(frac, (1U << PWM_SIFIVE_CMPWIDTH) - 1);
> +	frac = (1U << PWM_SIFIVE_CMPWIDTH) - 1 - frac;

I like the additional variable in pwm_sifive_get_state(). Can you please
add one here, too?

>  	mutex_lock(&ddata->lock);
>  	if (state->period != ddata->approx_period) {

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | https://www.pengutronix.de/ |
Re: [PATCH v9 2/3] pwm: sifive: change the PWM algorithm
Posted by Nylon Chen 1 year, 10 months ago 
Uwe Kleine-König <u.kleine-koenig@pengutronix.de> 於 2024年3月19日 週二 上午2:16寫道：
>
> On Thu, Feb 22, 2024 at 04:12:30PM +0800, Nylon Chen wrote:
> > The `frac` variable represents the pulse inactive time, and the result
> > of this algorithm is the pulse active time.
> > Therefore, we must reverse the result.
> >
> > The reference is SiFive FU740-C000 Manual[0]
> >
> > Link: https://sifive.cdn.prismic.io/sifive/1a82e600-1f93-4f41-b2d8-86ed8b16acba_fu740-c000-manual-v1p6.pdf [0]
> >
> > Co-developed-by: Zong Li <zong.li@sifive.com>
> > Signed-off-by: Zong Li <zong.li@sifive.com>
> > Co-developed-by: Vincent Chen <vincent.chen@sifive.com>
> > Signed-off-by: Vincent Chen <vincent.chen@sifive.com>
> > Signed-off-by: Nylon Chen <nylon.chen@sifive.com>
> > ---
> >  drivers/pwm/pwm-sifive.c | 10 ++++++----
> >  1 file changed, 6 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/pwm/pwm-sifive.c b/drivers/pwm/pwm-sifive.c
> > index eabddb7c7820..a586cfe4191b 100644
> > --- a/drivers/pwm/pwm-sifive.c
> > +++ b/drivers/pwm/pwm-sifive.c
> > @@ -110,9 +110,10 @@ static int pwm_sifive_get_state(struct pwm_chip *chip, struct pwm_device *pwm,
> >                               struct pwm_state *state)
> >  {
> >       struct pwm_sifive_ddata *ddata = pwm_sifive_chip_to_ddata(chip);
> > -     u32 duty, val;
> > +     u32 duty, val, inactive;
> >
> > -     duty = readl(ddata->regs + PWM_SIFIVE_PWMCMP(pwm->hwpwm));
> > +     inactive = readl(ddata->regs + PWM_SIFIVE_PWMCMP(pwm->hwpwm));
> > +     duty = (1U << PWM_SIFIVE_CMPWIDTH) - 1 - inactive;
> >
> >       state->enabled = duty > 0;
> >
> > @@ -123,7 +124,7 @@ static int pwm_sifive_get_state(struct pwm_chip *chip, struct pwm_device *pwm,
> >       state->period = ddata->real_period;
> >       state->duty_cycle =
> >               (u64)duty * ddata->real_period >> PWM_SIFIVE_CMPWIDTH;
> > -     state->polarity = PWM_POLARITY_INVERSED;
> > +     state->polarity = PWM_POLARITY_NORMAL;
> >
> >       return 0;
> >  }
> > @@ -139,7 +140,7 @@ static int pwm_sifive_apply(struct pwm_chip *chip, struct pwm_device *pwm,
> >       int ret = 0;
> >       u32 frac;
> >
> > -     if (state->polarity != PWM_POLARITY_INVERSED)
> > +     if (state->polarity != PWM_POLARITY_NORMAL)
> >               return -EINVAL;
> >
> >       cur_state = pwm->state;
> > @@ -159,6 +160,7 @@ static int pwm_sifive_apply(struct pwm_chip *chip, struct pwm_device *pwm,
> >       frac = DIV64_U64_ROUND_CLOSEST(num, state->period);
> >       /* The hardware cannot generate a 100% duty cycle */
>
> Is this still true now that we know that PWM_SIFIVE_PWMCMP is the
> inactive time in a period? If you fix that, the same claim in the header
> of the driver needs adaption, too.
I believe the statement is true, but I don't know which part the
driver header file refers to.
>
> >       frac = min(frac, (1U << PWM_SIFIVE_CMPWIDTH) - 1);
> > +     frac = (1U << PWM_SIFIVE_CMPWIDTH) - 1 - frac;
>
> I like the additional variable in pwm_sifive_get_state(). Can you please
> add one here, too?
got it
>
> >       mutex_lock(&ddata->lock);
> >       if (state->period != ddata->approx_period) {
>
Thank you for taking the time to help me review my implementation.

Nylon
> Best regards
> Uwe
>
> --
> Pengutronix e.K.                           | Uwe Kleine-König            |
> Industrial Linux Solutions                 | https://www.pengutronix.de/ |
Re: [PATCH v9 2/3] pwm: sifive: change the PWM algorithm
Posted by Nylon Chen 1 year, 10 months ago 
Nylon Chen <nylon.chen@sifive.com> 於 2024年4月2日 週二 上午10:08寫道：
>
> Uwe Kleine-König <u.kleine-koenig@pengutronix.de> 於 2024年3月19日 週二 上午2:16寫道：
> >
> > On Thu, Feb 22, 2024 at 04:12:30PM +0800, Nylon Chen wrote:
> > > The `frac` variable represents the pulse inactive time, and the result
> > > of this algorithm is the pulse active time.
> > > Therefore, we must reverse the result.
> > >
> > > The reference is SiFive FU740-C000 Manual[0]
> > >
> > > Link: https://sifive.cdn.prismic.io/sifive/1a82e600-1f93-4f41-b2d8-86ed8b16acba_fu740-c000-manual-v1p6.pdf [0]
> > >
> > > Co-developed-by: Zong Li <zong.li@sifive.com>
> > > Signed-off-by: Zong Li <zong.li@sifive.com>
> > > Co-developed-by: Vincent Chen <vincent.chen@sifive.com>
> > > Signed-off-by: Vincent Chen <vincent.chen@sifive.com>
> > > Signed-off-by: Nylon Chen <nylon.chen@sifive.com>
> > > ---
> > >  drivers/pwm/pwm-sifive.c | 10 ++++++----
> > >  1 file changed, 6 insertions(+), 4 deletions(-)
> > >
> > > diff --git a/drivers/pwm/pwm-sifive.c b/drivers/pwm/pwm-sifive.c
> > > index eabddb7c7820..a586cfe4191b 100644
> > > --- a/drivers/pwm/pwm-sifive.c
> > > +++ b/drivers/pwm/pwm-sifive.c
> > > @@ -110,9 +110,10 @@ static int pwm_sifive_get_state(struct pwm_chip *chip, struct pwm_device *pwm,
> > >                               struct pwm_state *state)
> > >  {
> > >       struct pwm_sifive_ddata *ddata = pwm_sifive_chip_to_ddata(chip);
> > > -     u32 duty, val;
> > > +     u32 duty, val, inactive;
> > >
> > > -     duty = readl(ddata->regs + PWM_SIFIVE_PWMCMP(pwm->hwpwm));
> > > +     inactive = readl(ddata->regs + PWM_SIFIVE_PWMCMP(pwm->hwpwm));
> > > +     duty = (1U << PWM_SIFIVE_CMPWIDTH) - 1 - inactive;
> > >
> > >       state->enabled = duty > 0;
> > >
> > > @@ -123,7 +124,7 @@ static int pwm_sifive_get_state(struct pwm_chip *chip, struct pwm_device *pwm,
> > >       state->period = ddata->real_period;
> > >       state->duty_cycle =
> > >               (u64)duty * ddata->real_period >> PWM_SIFIVE_CMPWIDTH;
> > > -     state->polarity = PWM_POLARITY_INVERSED;
> > > +     state->polarity = PWM_POLARITY_NORMAL;
> > >
> > >       return 0;
> > >  }
> > > @@ -139,7 +140,7 @@ static int pwm_sifive_apply(struct pwm_chip *chip, struct pwm_device *pwm,
> > >       int ret = 0;
> > >       u32 frac;
> > >
> > > -     if (state->polarity != PWM_POLARITY_INVERSED)
> > > +     if (state->polarity != PWM_POLARITY_NORMAL)
> > >               return -EINVAL;
> > >
> > >       cur_state = pwm->state;
> > > @@ -159,6 +160,7 @@ static int pwm_sifive_apply(struct pwm_chip *chip, struct pwm_device *pwm,
> > >       frac = DIV64_U64_ROUND_CLOSEST(num, state->period);
Hi Uwe,
> > >       /* The hardware cannot generate a 100% duty cycle */
Do you mean that if the inference is correct, this comment should be modified?
> >
> > Is this still true now that we know that PWM_SIFIVE_PWMCMP is the
> > inactive time in a period? If you fix that, the same claim in the header
> > of the driver needs adaption, too.
> I believe the statement is true, but I don't know which part the
> driver header file refers to.
> >
> > >       frac = min(frac, (1U << PWM_SIFIVE_CMPWIDTH) - 1);
> > > +     frac = (1U << PWM_SIFIVE_CMPWIDTH) - 1 - frac;
> >
> > I like the additional variable in pwm_sifive_get_state(). Can you please
> > add one here, too?
> got it
> >
> > >       mutex_lock(&ddata->lock);
> > >       if (state->period != ddata->approx_period) {
> >
> Thank you for taking the time to help me review my implementation.
>
> Nylon
> > Best regards
> > Uwe
> >
> > --
> > Pengutronix e.K.                           | Uwe Kleine-König            |
> > Industrial Linux Solutions                 | https://www.pengutronix.de/ |
Re: [PATCH v9 2/3] pwm: sifive: change the PWM algorithm
Posted by Uwe Kleine-König 1 year, 10 months ago 
Hello,

On Fri, Apr 12, 2024 at 10:07:12AM +0800, Nylon Chen wrote:
> Nylon Chen <nylon.chen@sifive.com> 於 2024年4月2日 週二 上午10:08寫道：
> > Uwe Kleine-König <u.kleine-koenig@pengutronix.de> 於 2024年3月19日 週二 上午2:16寫道：
> > > On Thu, Feb 22, 2024 at 04:12:30PM +0800, Nylon Chen wrote:
> > > >       /* The hardware cannot generate a 100% duty cycle */
> Do you mean that if the inference is correct, this comment should be modified?

If the polarity was wrong before and the hardware was believed to not be
able to generate a 100% relative duty cycle, then maybe it's really a 0%
relative duty cycle that's impossible?

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | https://www.pengutronix.de/ |

Patchew 2.1-devel-b67e4c2

© 2016 - 2026 Red Hat, Inc.