1. Patchew
  2. linux
  3. [PATCH 0/5] arm64: dts: qcom: r0q: enable more peripherals
  4. View patch

[PATCH 4/5] arm64: dts: qcom: r0q: enable hardware clocks

Eric Gonçalves posted 5 patches 4 months, 3 weeks ago
There is a newer version of this series
[PATCH 4/5] arm64: dts: qcom: r0q: enable hardware clocks
Posted by Eric Gonçalves 4 months, 3 weeks ago 
Enable the real-time clocks found in R0Q board.

Signed-off-by: Eric Gonçalves <ghatto404@gmail.com>
---
 arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts b/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
index c1b0b21c0ec5..c088f1acf6ea 100644
--- a/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
+++ b/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
@@ -225,6 +225,21 @@ vol_up_n: vol-up-n-state {
 	};
 };
 
+&pmk8350_rtc {
+	nvmem-cells = <&rtc_offset>;
+	nvmem-cell-names = "offset";
+
+	status = "okay";
+};
+
+&pmk8350_sdam_2 {
+	status = "okay";
+
+	rtc_offset: rtc-offset@bc {
+		reg = <0xbc 0x4>;
+	};
+};
+
 &pon_pwrkey {
 	status = "okay";
 };
-- 
2.51.0
Re: [PATCH 4/5] arm64: dts: qcom: r0q: enable hardware clocks
Posted by Konrad Dybcio 4 months, 2 weeks ago 
On 9/20/25 3:46 AM, Eric Gonçalves wrote:
> Enable the real-time clocks found in R0Q board.
> 
> Signed-off-by: Eric Gonçalves <ghatto404@gmail.com>
> ---
>  arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts | 15 +++++++++++++++
>  1 file changed, 15 insertions(+)
> 
> diff --git a/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts b/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
> index c1b0b21c0ec5..c088f1acf6ea 100644
> --- a/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
> +++ b/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
> @@ -225,6 +225,21 @@ vol_up_n: vol-up-n-state {
>  	};
>  };
>  
> +&pmk8350_rtc {
> +	nvmem-cells = <&rtc_offset>;
> +	nvmem-cell-names = "offset";
> +
> +	status = "okay";
> +};
> +
> +&pmk8350_sdam_2 {
> +	status = "okay";
> +
> +	rtc_offset: rtc-offset@bc {

Is this an offset you took from somewhere downstream?

Generally you *really don't want to* poke at random SDAM cells,
as they contain a lot of important settings (incl. battery/charging)

Konrad
Re: [PATCH 4/5] arm64: dts: qcom: r0q: enable hardware clocks
Posted by Eric Gonçalves 4 months ago 

On September 25, 2025 10:09:48 AM GMT-03:00, Konrad Dybcio <konrad.dybcio@oss.qualcomm.com> wrote:
>On 9/20/25 3:46 AM, Eric Gonçalves wrote:
>> Enable the real-time clocks found in R0Q board.
>> 
>> Signed-off-by: Eric Gonçalves <ghatto404@gmail.com>
>> ---
>>  arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts | 15 +++++++++++++++
>>  1 file changed, 15 insertions(+)
>> 
>> diff --git a/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts b/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
>> index c1b0b21c0ec5..c088f1acf6ea 100644
>> --- a/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
>> +++ b/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
>> @@ -225,6 +225,21 @@ vol_up_n: vol-up-n-state {
>>  	};
>>  };
>>  
>> +&pmk8350_rtc {
>> +	nvmem-cells = <&rtc_offset>;
>> +	nvmem-cell-names = "offset";
>> +
>> +	status = "okay";
>> +};
>> +
>> +&pmk8350_sdam_2 {
>> +	status = "okay";
>> +
>> +	rtc_offset: rtc-offset@bc {
>
>Is this an offset you took from somewhere downstream?
>
>Generally you *really don't want to* poke at random SDAM cells,
>as they contain a lot of important settings (incl. battery/charging)
From another sm8450 device, I'm sure it's okay.
>
>Konrad
Re: [PATCH 4/5] arm64: dts: qcom: r0q: enable hardware clocks
Posted by Konrad Dybcio 4 months ago 
On 10/5/25 5:50 AM, Eric Gonçalves wrote:
> 
> 
> On September 25, 2025 10:09:48 AM GMT-03:00, Konrad Dybcio <konrad.dybcio@oss.qualcomm.com> wrote:
>> On 9/20/25 3:46 AM, Eric Gonçalves wrote:
>>> Enable the real-time clocks found in R0Q board.
>>>
>>> Signed-off-by: Eric Gonçalves <ghatto404@gmail.com>
>>> ---
>>>  arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts | 15 +++++++++++++++
>>>  1 file changed, 15 insertions(+)
>>>
>>> diff --git a/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts b/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
>>> index c1b0b21c0ec5..c088f1acf6ea 100644
>>> --- a/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
>>> +++ b/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
>>> @@ -225,6 +225,21 @@ vol_up_n: vol-up-n-state {
>>>  	};
>>>  };
>>>  
>>> +&pmk8350_rtc {
>>> +	nvmem-cells = <&rtc_offset>;
>>> +	nvmem-cell-names = "offset";
>>> +
>>> +	status = "okay";
>>> +};
>>> +
>>> +&pmk8350_sdam_2 {
>>> +	status = "okay";
>>> +
>>> +	rtc_offset: rtc-offset@bc {
>>
>> Is this an offset you took from somewhere downstream?
>>
>> Generally you *really don't want to* poke at random SDAM cells,
>> as they contain a lot of important settings (incl. battery/charging)
> From another sm8450 device, I'm sure it's okay.

That as an argument alone doesn't sound convincing to me, since vendors
also sometimes repurpose unused-by-Qualcomm SDAM cells

I actually found a data source internally and this cell you're trying
to use is reserved for PBS (see drivers/soc/qcom/qcom-pbs.c), meaning
you already fell into this trap..

Konrad
Re: [PATCH 4/5] arm64: dts: qcom: r0q: enable hardware clocks
Posted by Eric Gonçalves 4 months ago 

On October 6, 2025 9:31:42 AM GMT-03:00, Konrad Dybcio <konrad.dybcio@oss.qualcomm.com> wrote:
>On 10/5/25 5:50 AM, Eric Gonçalves wrote:
>> 
>> 
>> On September 25, 2025 10:09:48 AM GMT-03:00, Konrad Dybcio <konrad.dybcio@oss.qualcomm.com> wrote:
>>> On 9/20/25 3:46 AM, Eric Gonçalves wrote:
>>>> Enable the real-time clocks found in R0Q board.
>>>>
>>>> Signed-off-by: Eric Gonçalves <ghatto404@gmail.com>
>>>> ---
>>>>  arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts | 15 +++++++++++++++
>>>>  1 file changed, 15 insertions(+)
>>>>
>>>> diff --git a/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts b/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
>>>> index c1b0b21c0ec5..c088f1acf6ea 100644
>>>> --- a/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
>>>> +++ b/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
>>>> @@ -225,6 +225,21 @@ vol_up_n: vol-up-n-state {
>>>>  	};
>>>>  };
>>>>  
>>>> +&pmk8350_rtc {
>>>> +	nvmem-cells = <&rtc_offset>;
>>>> +	nvmem-cell-names = "offset";
>>>> +
>>>> +	status = "okay";
>>>> +};
>>>> +
>>>> +&pmk8350_sdam_2 {
>>>> +	status = "okay";
>>>> +
>>>> +	rtc_offset: rtc-offset@bc {
>>>
>>> Is this an offset you took from somewhere downstream?
>>>
>>> Generally you *really don't want to* poke at random SDAM cells,
>>> as they contain a lot of important settings (incl. battery/charging)
>> From another sm8450 device, I'm sure it's okay.
>
>That as an argument alone doesn't sound convincing to me, since vendors
>also sometimes repurpose unused-by-Qualcomm SDAM cells
>
>I actually found a data source internally and this cell you're trying
>to use is reserved for PBS (see drivers/soc/qcom/qcom-pbs.c), meaning
>you already fell into this trap..
Interesting, in this case then why does rtc still work? And
how can I find the real cell? It's not in downstream DT.
>
>Konrad
Re: [PATCH 4/5] arm64: dts: qcom: r0q: enable hardware clocks
Posted by Konrad Dybcio 4 months ago 
On 10/6/25 5:49 PM, Eric Gonçalves wrote:
> 
> 
> On October 6, 2025 9:31:42 AM GMT-03:00, Konrad Dybcio <konrad.dybcio@oss.qualcomm.com> wrote:
>> On 10/5/25 5:50 AM, Eric Gonçalves wrote:
>>>
>>>
>>> On September 25, 2025 10:09:48 AM GMT-03:00, Konrad Dybcio <konrad.dybcio@oss.qualcomm.com> wrote:
>>>> On 9/20/25 3:46 AM, Eric Gonçalves wrote:
>>>>> Enable the real-time clocks found in R0Q board.
>>>>>
>>>>> Signed-off-by: Eric Gonçalves <ghatto404@gmail.com>
>>>>> ---
>>>>>  arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts | 15 +++++++++++++++
>>>>>  1 file changed, 15 insertions(+)
>>>>>
>>>>> diff --git a/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts b/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
>>>>> index c1b0b21c0ec5..c088f1acf6ea 100644
>>>>> --- a/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
>>>>> +++ b/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
>>>>> @@ -225,6 +225,21 @@ vol_up_n: vol-up-n-state {
>>>>>  	};
>>>>>  };
>>>>>  
>>>>> +&pmk8350_rtc {
>>>>> +	nvmem-cells = <&rtc_offset>;
>>>>> +	nvmem-cell-names = "offset";
>>>>> +
>>>>> +	status = "okay";
>>>>> +};
>>>>> +
>>>>> +&pmk8350_sdam_2 {
>>>>> +	status = "okay";
>>>>> +
>>>>> +	rtc_offset: rtc-offset@bc {
>>>>
>>>> Is this an offset you took from somewhere downstream?
>>>>
>>>> Generally you *really don't want to* poke at random SDAM cells,
>>>> as they contain a lot of important settings (incl. battery/charging)
>>> From another sm8450 device, I'm sure it's okay.
>>
>> That as an argument alone doesn't sound convincing to me, since vendors
>> also sometimes repurpose unused-by-Qualcomm SDAM cells
>>
>> I actually found a data source internally and this cell you're trying
>> to use is reserved for PBS (see drivers/soc/qcom/qcom-pbs.c), meaning
>> you already fell into this trap..
> Interesting, in this case then why does rtc still work? And
> how can I find the real cell? It's not in downstream DT.

Well if nothing accesses that cell between your last write and your first
read, it will of course work, because SDAM is just a means of storage

As for downstream/Android, it does not use SDAM at all

Konrad
Re: [PATCH 4/5] arm64: dts: qcom: r0q: enable hardware clocks
Posted by Eric Gonçalves 4 months ago 

On October 8, 2025 5:48:53 AM GMT-03:00, Konrad Dybcio <konrad.dybcio@oss.qualcomm.com> wrote:
>On 10/6/25 5:49 PM, Eric Gonçalves wrote:
>> 
>> 
>> On October 6, 2025 9:31:42 AM GMT-03:00, Konrad Dybcio <konrad.dybcio@oss.qualcomm.com> wrote:
>>> On 10/5/25 5:50 AM, Eric Gonçalves wrote:
>>>>
>>>>
>>>> On September 25, 2025 10:09:48 AM GMT-03:00, Konrad Dybcio <konrad.dybcio@oss.qualcomm.com> wrote:
>>>>> On 9/20/25 3:46 AM, Eric Gonçalves wrote:
>>>>>> Enable the real-time clocks found in R0Q board.
>>>>>>
>>>>>> Signed-off-by: Eric Gonçalves <ghatto404@gmail.com>
>>>>>> ---
>>>>>>  arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts | 15 +++++++++++++++
>>>>>>  1 file changed, 15 insertions(+)
>>>>>>
>>>>>> diff --git a/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts b/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
>>>>>> index c1b0b21c0ec5..c088f1acf6ea 100644
>>>>>> --- a/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
>>>>>> +++ b/arch/arm64/boot/dts/qcom/sm8450-samsung-r0q.dts
>>>>>> @@ -225,6 +225,21 @@ vol_up_n: vol-up-n-state {
>>>>>>  	};
>>>>>>  };
>>>>>>  
>>>>>> +&pmk8350_rtc {
>>>>>> +	nvmem-cells = <&rtc_offset>;
>>>>>> +	nvmem-cell-names = "offset";
>>>>>> +
>>>>>> +	status = "okay";
>>>>>> +};
>>>>>> +
>>>>>> +&pmk8350_sdam_2 {
>>>>>> +	status = "okay";
>>>>>> +
>>>>>> +	rtc_offset: rtc-offset@bc {
>>>>>
>>>>> Is this an offset you took from somewhere downstream?
>>>>>
>>>>> Generally you *really don't want to* poke at random SDAM cells,
>>>>> as they contain a lot of important settings (incl. battery/charging)
>>>> From another sm8450 device, I'm sure it's okay.
>>>
>>> That as an argument alone doesn't sound convincing to me, since vendors
>>> also sometimes repurpose unused-by-Qualcomm SDAM cells
>>>
>>> I actually found a data source internally and this cell you're trying
>>> to use is reserved for PBS (see drivers/soc/qcom/qcom-pbs.c), meaning
>>> you already fell into this trap..
>> Interesting, in this case then why does rtc still work? And
>> how can I find the real cell? It's not in downstream DT.
>
>Well if nothing accesses that cell between your last write and your first
>read, it will of course work, because SDAM is just a means of storage
>
>As for downstream/Android, it does not use SDAM at all
Alright, I'll drop this patch then, thanks
>
>Konrad

Patchew 2.1-devel-b67e4c2

© 2016 - 2026 Red Hat, Inc.