pwm: tiehrpwm: Enable pwmchip's parent device before setting configuration

[PATCH v4] pwm: tiehrpwm: Enable pwmchip's parent device before setting configuration

Posted by Gokul Praveen 2 weeks, 4 days ago

The period and duty cycle configurations on J7200 and J784S4 SoCs
does not get reflected after setting them using sysfs nodes.
This is because at the end of ehrpwm_pwm_config function,
the put_sync function is called which resets the hardware.

Hold the PWM controller out of low-power mode during .apply() to
make sure it accepts the writes to its registers.

This renders the calls to pm_runtime_get_sync() and
pm_runtime_put_sync() in ehrpwm_pwm_config() into no-ops, so
these can be dropped.

Fixes: 5f027d9b83db("pwm: tiehrpwm: Implement .apply() callback")
Signed-off-by: Gokul Praveen <g-praveen@ti.com>
---
v4 <==> v3
===========
* Modified subject line and commit description for better clarity.

v3 <==> v2
==========
* Simplified the logic by adding guard function for runtime power
  management.

v2 <==> v1
==========
* Removed space between Fixes and Signed-off tag

 drivers/pwm/pwm-tiehrpwm.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/pwm/pwm-tiehrpwm.c b/drivers/pwm/pwm-tiehrpwm.c
index 7a86cb090f76..2533c95b0ba9 100644
--- a/drivers/pwm/pwm-tiehrpwm.c
+++ b/drivers/pwm/pwm-tiehrpwm.c
@@ -237,8 +237,6 @@ static int ehrpwm_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm,
 	if (period_cycles < 1)
 		period_cycles = 1;
 
-	pm_runtime_get_sync(pwmchip_parent(chip));
-
 	/* Update clock prescaler values */
 	ehrpwm_modify(pc->mmio_base, TBCTL, TBCTL_CLKDIV_MASK, tb_divval);
 
@@ -290,8 +288,6 @@ static int ehrpwm_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm,
 	if (!(duty_cycles > period_cycles))
 		ehrpwm_write(pc->mmio_base, cmp_reg, duty_cycles);
 
-	pm_runtime_put_sync(pwmchip_parent(chip));
-
 	return 0;
 }
 
@@ -378,6 +374,8 @@ static int ehrpwm_pwm_apply(struct pwm_chip *chip, struct pwm_device *pwm,
 	int err;
 	bool enabled = pwm->state.enabled;
 
+	guard(pm_runtime_active)(pwmchip_parent(chip));
+
 	if (state->polarity != pwm->state.polarity) {
 		if (enabled) {
 			ehrpwm_pwm_disable(chip, pwm);
-- 
2.34.1

Re: [PATCH v4] pwm: tiehrpwm: Enable pwmchip's parent device before setting configuration

Posted by Uwe Kleine-König 2 weeks, 4 days ago

Hello Gokul,

On Wed, Jan 21, 2026 at 11:41:34AM +0530, Gokul Praveen wrote:
> The period and duty cycle configurations on J7200 and J784S4 SoCs
> does not get reflected after setting them using sysfs nodes.
> This is because at the end of ehrpwm_pwm_config function,
> the put_sync function is called which resets the hardware.
> 
> Hold the PWM controller out of low-power mode during .apply() to
> make sure it accepts the writes to its registers.
> 
> This renders the calls to pm_runtime_get_sync() and
> pm_runtime_put_sync() in ehrpwm_pwm_config() into no-ops, so
> these can be dropped.
> 
> Fixes: 5f027d9b83db("pwm: tiehrpwm: Implement .apply() callback")
> Signed-off-by: Gokul Praveen <g-praveen@ti.com>

I applied this patch to

	https://git.kernel.org/pub/scm/linux/kernel/git/ukleinek/linux.git pwm/for-next

and will send it during the next merge window. The problem is old
(v5.18-rc1~54^2~29 from 2021) and it's late in the development cycle, so
I won't bother Linus with it for 6.19.

Best regards
Uwe

Re: [PATCH v4] pwm: tiehrpwm: Enable pwmchip's parent device before setting configuration

Posted by Gokul Praveen 2 weeks, 4 days ago

Hello Uwe,

On 21/01/26 15:48, Uwe Kleine-König wrote:
> Hello Gokul,
> 
> On Wed, Jan 21, 2026 at 11:41:34AM +0530, Gokul Praveen wrote:
>> The period and duty cycle configurations on J7200 and J784S4 SoCs
>> does not get reflected after setting them using sysfs nodes.
>> This is because at the end of ehrpwm_pwm_config function,
>> the put_sync function is called which resets the hardware.
>>
>> Hold the PWM controller out of low-power mode during .apply() to
>> make sure it accepts the writes to its registers.
>>
>> This renders the calls to pm_runtime_get_sync() and
>> pm_runtime_put_sync() in ehrpwm_pwm_config() into no-ops, so
>> these can be dropped.
>>
>> Fixes: 5f027d9b83db("pwm: tiehrpwm: Implement .apply() callback")
>> Signed-off-by: Gokul Praveen <g-praveen@ti.com>
> 
> I applied this patch to
> 
> 	https://git.kernel.org/pub/scm/linux/kernel/git/ukleinek/linux.git pwm/for-next
> 
> and will send it during the next merge window. The problem is old
> (v5.18-rc1~54^2~29 from 2021) and it's late in the development cycle, so
> I won't bother Linus with it for 6.19.
> 

Thank you for your continued support on this issue as well as for 
accepting the patch.

Best Regards
Gokul Praveen
> Best regards
> Uwe

Re: [PATCH v4] pwm: tiehrpwm: Enable pwmchip's parent device before setting configuration

Posted by Kumar, Udit 2 weeks, 1 day ago

Hi Uwe/Gokul,

On 1/21/2026 4:19 PM, Gokul Praveen wrote:
> Hello Uwe,
>
> On 21/01/26 15:48, Uwe Kleine-König wrote:
>> Hello Gokul,
>>
>> On Wed, Jan 21, 2026 at 11:41:34AM +0530, Gokul Praveen wrote:
>>> The period and duty cycle configurations on J7200 and J784S4 SoCs
>>> does not get reflected after setting them using sysfs nodes.
>>> This is because at the end of ehrpwm_pwm_config function,
>>> the put_sync function is called which resets the hardware.
>>>
>>> Hold the PWM controller out of low-power mode during .apply() to
>>> make sure it accepts the writes to its registers.
>>>
>>> This renders the calls to pm_runtime_get_sync() and
>>> pm_runtime_put_sync() in ehrpwm_pwm_config() into no-ops, so
>>> these can be dropped.
>>>
>>> Fixes: 5f027d9b83db("pwm: tiehrpwm: Implement .apply() callback")
>>> Signed-off-by: Gokul Praveen <g-praveen@ti.com>
>>
>> I applied this patch to
>>
>>     https://git.kernel.org/pub/scm/linux/kernel/git/ukleinek/linux.git 
>> pwm/for-next
>>
>> and will send it during the next merge window. The problem is old
>> (v5.18-rc1~54^2~29 from 2021) and it's late in the development cycle, so
>> I won't bother Linus with it for 6.19.
>>
>
fyi,

We are hitting compilation error with this patch

https://gist.github.com/uditkumarti/d23bc9ee1886bfce70cc76e4b19fbb59


> Thank you for your continued support on this issue as well as for 
> accepting the patch.
>
> Best Regards
> Gokul Praveen
>> Best regards
>> Uwe
>

Re: [PATCH v4] pwm: tiehrpwm: Enable pwmchip's parent device before setting configuration

Posted by Uwe Kleine-König 2 weeks, 1 day ago

Hello,

On Fri, Jan 23, 2026 at 08:28:48PM +0530, Kumar, Udit wrote:
> On 1/21/2026 4:19 PM, Gokul Praveen wrote:
> > Hello Uwe,
> > 
> > On 21/01/26 15:48, Uwe Kleine-König wrote:
> > > Hello Gokul,
> > > 
> > > On Wed, Jan 21, 2026 at 11:41:34AM +0530, Gokul Praveen wrote:
> > > > The period and duty cycle configurations on J7200 and J784S4 SoCs
> > > > does not get reflected after setting them using sysfs nodes.
> > > > This is because at the end of ehrpwm_pwm_config function,
> > > > the put_sync function is called which resets the hardware.
> > > > 
> > > > Hold the PWM controller out of low-power mode during .apply() to
> > > > make sure it accepts the writes to its registers.
> > > > 
> > > > This renders the calls to pm_runtime_get_sync() and
> > > > pm_runtime_put_sync() in ehrpwm_pwm_config() into no-ops, so
> > > > these can be dropped.
> > > > 
> > > > Fixes: 5f027d9b83db("pwm: tiehrpwm: Implement .apply() callback")
> > > > Signed-off-by: Gokul Praveen <g-praveen@ti.com>
> > > 
> > > I applied this patch to
> > > 
> > >     https://git.kernel.org/pub/scm/linux/kernel/git/ukleinek/linux.git
> > > pwm/for-next
> > > 
> > > and will send it during the next merge window. The problem is old
> > > (v5.18-rc1~54^2~29 from 2021) and it's late in the development cycle, so
> > > I won't bother Linus with it for 6.19.
> > > 
> > 
> fyi,
> 
> We are hitting compilation error with this patch
> 
> https://gist.github.com/uditkumarti/d23bc9ee1886bfce70cc76e4b19fbb59

I assume you tried to compile this patch on top of a tree that doesn't
also has commit 9a0abc39450a3123fd52533a662fbd37e0d1508c?

Best regards
Uwe

Re: [PATCH v4] pwm: tiehrpwm: Enable pwmchip's parent device before setting configuration

Posted by Kumar, Udit 2 weeks, 1 day ago

On 1/23/2026 11:31 PM, Uwe Kleine-König wrote:
> Hello,
>
> On Fri, Jan 23, 2026 at 08:28:48PM +0530, Kumar, Udit wrote:
>> On 1/21/2026 4:19 PM, Gokul Praveen wrote:
>>> Hello Uwe,
>>>
>>> On 21/01/26 15:48, Uwe Kleine-König wrote:
>>>> Hello Gokul,
>>>>
>>>> On Wed, Jan 21, 2026 at 11:41:34AM +0530, Gokul Praveen wrote:
>>>>> The period and duty cycle configurations on J7200 and J784S4 SoCs
>>>>> does not get reflected after setting them using sysfs nodes.
>>>>> This is because at the end of ehrpwm_pwm_config function,
>>>>> the put_sync function is called which resets the hardware.
>>>>>
>>>>> Hold the PWM controller out of low-power mode during .apply() to
>>>>> make sure it accepts the writes to its registers.
>>>>>
>>>>> This renders the calls to pm_runtime_get_sync() and
>>>>> pm_runtime_put_sync() in ehrpwm_pwm_config() into no-ops, so
>>>>> these can be dropped.
>>>>>
>>>>> Fixes: 5f027d9b83db("pwm: tiehrpwm: Implement .apply() callback")
>>>>> Signed-off-by: Gokul Praveen <g-praveen@ti.com>
>>>> I applied this patch to
>>>>
>>>>      https://git.kernel.org/pub/scm/linux/kernel/git/ukleinek/linux.git
>>>> pwm/for-next
>>>>
>>>> and will send it during the next merge window. The problem is old
>>>> (v5.18-rc1~54^2~29 from 2021) and it's late in the development cycle, so
>>>> I won't bother Linus with it for 6.19.
>>>>
>> fyi,
>>
>> We are hitting compilation error with this patch
>>
>> https://gist.github.com/uditkumarti/d23bc9ee1886bfce70cc76e4b19fbb59
> I assume you tried to compile this patch on top of a tree that doesn't
> also has commit 9a0abc39450a3123fd52533a662fbd37e0d1508c?


Thanks for pointing to this , I just cherry-pick only patch of Gokul,

after picking 9a0abc39450a3123fd52533a662fbd37e0d1508c, compile is ok 
for me .


>
> Best regards
> Uwe