Re: [PATCH] arm64: dts: qcom: Add backlight support for eDP panel

[Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>]

On 11/5/25 10:58 AM, Yongxing Mou wrote:
> 
> 
> On 10/30/2025 7:27 PM, Konrad Dybcio wrote:
>> On 10/30/25 12:07 PM, Yongxing Mou wrote:
>>>
>>>
>>> On 10/30/2025 5:41 PM, Konrad Dybcio wrote:
>>>> On 10/30/25 10:08 AM, Yongxing Mou wrote:
>>>>>
>>>>>
>>>>> On 10/28/2025 5:08 PM, Konrad Dybcio wrote:
>>>>>> On 10/28/25 8:04 AM, Yongxing Mou wrote:
>>>>>>> Previously, the eDP panel backlight was enabled via UEFI. Added backlight
>>>>>>> control node in kernel DTS due to some meta may not enable the backlight.
>>>>>>>
>>>>>>> Aligned with other x1e80100-based platforms: the PWM signal is controlled
>>>>>>> by PMK8550, and the backlight enable signal is handled by PMC8380.
>>>>>>>
>>>>>>> Signed-off-by: Yongxing Mou <yongxing.mou@oss.qualcomm.com>
>>>>>>> ---
>>>>>>
>>>>>> The subject must say "hamoa-iot-evk:"
>>>>> Thanks and got it , will fix it in next version.>> arch/arm64/boot/dts/qcom/hamoa-iot-evk.dts | 55 ++++++++++++++++++++++++++++++
>>>>>>>     1 file changed, 55 insertions(+)
>>>>>>>
>>>>>>> diff --git a/arch/arm64/boot/dts/qcom/hamoa-iot-evk.dts b/arch/arm64/boot/dts/qcom/hamoa-iot-evk.dts
>>>>>>> index 36dd6599402b4650b7f8ad2c0cd22212116a25fe..fda1b3a3c7673be74832c27849231cba4bc1f25f 100644
>>>>>>> --- a/arch/arm64/boot/dts/qcom/hamoa-iot-evk.dts
>>>>>>> +++ b/arch/arm64/boot/dts/qcom/hamoa-iot-evk.dts
>>>>>>> @@ -17,6 +17,16 @@ aliases {
>>>>>>>             serial1 = &uart14;
>>>>>>>         };
>>>>>>>     +    backlight: backlight {
>>>>>>> +        compatible = "pwm-backlight";
>>>>>>> +        pwms = <&pmk8550_pwm 0 5000000>;
>>>>>>
>>>>>> Try adjusting the backlight value.. you'll find some funny behavior
>>>>>> near the max level.. which reminds me I should send some fixes for
>>>>>> some laptop DTs
>>>>>>
>>>>>    I tried sliding the backlight between the maximum and minimum values. It seems that at a certain sliding speed, when approaching the maximum value, there is some brightness flickering. Is this the phenomenon you’re referring to?> [...]
>>>>
>>>> "maybe" - for me it was the brightness actually going down beyond ~80%
>>>>
>>>> The issue is that the PWM hardware can't really do a period that's
>>>> == 5000000 ns
>>>>
>>>> But it seems like this is no longer an issue since the driver now
>>>> rounds up to the closest supported value, see last couple commits
>>>> to drivers/leds/rgb/leds-qcom-lpg.c
>>>>
>>>> Konrad
>>> Oh,thanks for sharing that. I saw the phenomenon you mentioned. Previously, my baseline had already merged this patch series: https://lore.kernel.org/all/20250305-leds-qcom-lpg-fix-max-pwm-on-hi-res-v4-0-bfe124a53a9f@linaro.org/, so I didn’t observe the ‘brightness actually going down beyond ~80%.’and just very slight flickering. After I reverted patches, it indeed appeared.  So here i want to confirm,
>>> pwms = <&pmk8550_pwm 0 5000000>
>>> can i keep the value 5000000 ?
>>
>> I think so.. perhaps make sure the period is appropriate for
>> the backlight device at the other end since you mentioned it's
>> flickering
>>
>> Konrad
> Compared to ‘brightness actually going down beyond ~80%’, this is just a very slight flicker that only occurs when sliding the brightness bar at a specific speed.It won’t occur in most cases. Also tried another value pwms = <&pmk8550_pwm 0 4266537>, and it looks the same, so can we keep it at 5000000 here?

As I said, just put whatever the backlight that the PWM controller
outputs to expects.

Konrad