[PATCH] arm64: dts: qcom: ipq9574: Fix 'l2' regulator min/max voltages

Varadarajan Narayanan posted 1 patch 3 months, 2 weeks ago
arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
[PATCH] arm64: dts: qcom: ipq9574: Fix 'l2' regulator min/max voltages
Posted by Varadarajan Narayanan 3 months, 2 weeks ago
The min and max voltages on l2 regulator is 850000uV. This was
incorrectly set at 1800000uV earlier and that affected the
stability of the networking subsystem.

Fixes: d5506524d9d9 ("arm64: dts: qcom: ipq9574: Add LDO regulator node")
Signed-off-by: Varadarajan Narayanan <quic_varada@quicinc.com>
---
 arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi b/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi
index bdb396afb992..21b04a2c629d 100644
--- a/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi
+++ b/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi
@@ -106,8 +106,8 @@ ipq9574_s1: s1 {
 		};
 
 		mp5496_l2: l2 {
-			regulator-min-microvolt = <1800000>;
-			regulator-max-microvolt = <1800000>;
+			regulator-min-microvolt = <850000>;
+			regulator-max-microvolt = <850000>;
 			regulator-always-on;
 			regulator-boot-on;
 		};

base-commit: 5d4809e25903ab8e74034c1f23c787fd26d52934
-- 
2.34.1
Re: [PATCH] arm64: dts: qcom: ipq9574: Fix 'l2' regulator min/max voltages
Posted by Konrad Dybcio 3 months, 2 weeks ago
On 6/23/25 1:32 PM, Varadarajan Narayanan wrote:
> The min and max voltages on l2 regulator is 850000uV. This was
> incorrectly set at 1800000uV earlier and that affected the
> stability of the networking subsystem.
> 
> Fixes: d5506524d9d9 ("arm64: dts: qcom: ipq9574: Add LDO regulator node")
> Signed-off-by: Varadarajan Narayanan <quic_varada@quicinc.com>
> ---

Shouldn't there be a consumer for it, instead of it being always-on?

Konrad

>  arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi b/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi
> index bdb396afb992..21b04a2c629d 100644
> --- a/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi
> +++ b/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi
> @@ -106,8 +106,8 @@ ipq9574_s1: s1 {
>  		};
>  
>  		mp5496_l2: l2 {
> -			regulator-min-microvolt = <1800000>;
> -			regulator-max-microvolt = <1800000>;
> +			regulator-min-microvolt = <850000>;
> +			regulator-max-microvolt = <850000>;
>  			regulator-always-on;
>  			regulator-boot-on;
>  		};
> 
> base-commit: 5d4809e25903ab8e74034c1f23c787fd26d52934
Re: [PATCH] arm64: dts: qcom: ipq9574: Fix 'l2' regulator min/max voltages
Posted by Varadarajan Narayanan 3 months, 2 weeks ago
On Mon, Jun 23, 2025 at 01:34:22PM +0200, Konrad Dybcio wrote:
> On 6/23/25 1:32 PM, Varadarajan Narayanan wrote:
> > The min and max voltages on l2 regulator is 850000uV. This was
> > incorrectly set at 1800000uV earlier and that affected the
> > stability of the networking subsystem.
> >
> > Fixes: d5506524d9d9 ("arm64: dts: qcom: ipq9574: Add LDO regulator node")
> > Signed-off-by: Varadarajan Narayanan <quic_varada@quicinc.com>
> > ---
>
> Shouldn't there be a consumer for it, instead of it being always-on?

The uniphy block is the consumer of this voltage. The PMIC configures it
to 850000uV based on OTP settings and s/w doesn't change it as uniphy
doesn't do scaling.

-Varada

> >  arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi b/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi
> > index bdb396afb992..21b04a2c629d 100644
> > --- a/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi
> > +++ b/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi
> > @@ -106,8 +106,8 @@ ipq9574_s1: s1 {
> >  		};
> >
> >  		mp5496_l2: l2 {
> > -			regulator-min-microvolt = <1800000>;
> > -			regulator-max-microvolt = <1800000>;
> > +			regulator-min-microvolt = <850000>;
> > +			regulator-max-microvolt = <850000>;
> >  			regulator-always-on;
> >  			regulator-boot-on;
> >  		};
> >
> > base-commit: 5d4809e25903ab8e74034c1f23c787fd26d52934
Re: [PATCH] arm64: dts: qcom: ipq9574: Fix 'l2' regulator min/max voltages
Posted by Dmitry Baryshkov 3 months, 2 weeks ago
On Tue, Jun 24, 2025 at 11:38:34AM +0530, Varadarajan Narayanan wrote:
> On Mon, Jun 23, 2025 at 01:34:22PM +0200, Konrad Dybcio wrote:
> > On 6/23/25 1:32 PM, Varadarajan Narayanan wrote:
> > > The min and max voltages on l2 regulator is 850000uV. This was
> > > incorrectly set at 1800000uV earlier and that affected the
> > > stability of the networking subsystem.
> > >
> > > Fixes: d5506524d9d9 ("arm64: dts: qcom: ipq9574: Add LDO regulator node")
> > > Signed-off-by: Varadarajan Narayanan <quic_varada@quicinc.com>
> > > ---
> >
> > Shouldn't there be a consumer for it, instead of it being always-on?
> 
> The uniphy block is the consumer of this voltage. The PMIC configures it
> to 850000uV based on OTP settings and s/w doesn't change it as uniphy
> doesn't do scaling.

Please express this in DT terms, specifying the regulator as a supplier
to the uniphy and then making the uniphy driver use that supply.

> 
> -Varada
> 
> > >  arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi | 4 ++--
> > >  1 file changed, 2 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi b/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi
> > > index bdb396afb992..21b04a2c629d 100644
> > > --- a/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi
> > > +++ b/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi
> > > @@ -106,8 +106,8 @@ ipq9574_s1: s1 {
> > >  		};
> > >
> > >  		mp5496_l2: l2 {
> > > -			regulator-min-microvolt = <1800000>;
> > > -			regulator-max-microvolt = <1800000>;
> > > +			regulator-min-microvolt = <850000>;
> > > +			regulator-max-microvolt = <850000>;
> > >  			regulator-always-on;
> > >  			regulator-boot-on;
> > >  		};
> > >
> > > base-commit: 5d4809e25903ab8e74034c1f23c787fd26d52934

-- 
With best wishes
Dmitry
Re: [PATCH] arm64: dts: qcom: ipq9574: Fix 'l2' regulator min/max voltages
Posted by Varadarajan Narayanan 3 months, 2 weeks ago
On Wed, Jun 25, 2025 at 04:26:29AM +0300, Dmitry Baryshkov wrote:
> On Tue, Jun 24, 2025 at 11:38:34AM +0530, Varadarajan Narayanan wrote:
> > On Mon, Jun 23, 2025 at 01:34:22PM +0200, Konrad Dybcio wrote:
> > > On 6/23/25 1:32 PM, Varadarajan Narayanan wrote:
> > > > The min and max voltages on l2 regulator is 850000uV. This was
> > > > incorrectly set at 1800000uV earlier and that affected the
> > > > stability of the networking subsystem.
> > > >
> > > > Fixes: d5506524d9d9 ("arm64: dts: qcom: ipq9574: Add LDO regulator node")
> > > > Signed-off-by: Varadarajan Narayanan <quic_varada@quicinc.com>
> > > > ---
> > >
> > > Shouldn't there be a consumer for it, instead of it being always-on?
> >
> > The uniphy block is the consumer of this voltage. The PMIC configures it
> > to 850000uV based on OTP settings and s/w doesn't change it as uniphy
> > doesn't do scaling.
>
> Please express this in DT terms, specifying the regulator as a supplier
> to the uniphy and then making the uniphy driver use that supply.

Sure, we will update UNIPHY patch series [1] to enable this voltage supplier.

Thanks
Varada

1 - https://lore.kernel.org/linux-arm-msm/20250207-ipq_pcs_6-14_rc1-v5-0-be2ebec32921@quicinc.com/

> > > >  arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi | 4 ++--
> > > >  1 file changed, 2 insertions(+), 2 deletions(-)
> > > >
> > > > diff --git a/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi b/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi
> > > > index bdb396afb992..21b04a2c629d 100644
> > > > --- a/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi
> > > > +++ b/arch/arm64/boot/dts/qcom/ipq9574-rdp-common.dtsi
> > > > @@ -106,8 +106,8 @@ ipq9574_s1: s1 {
> > > >  		};
> > > >
> > > >  		mp5496_l2: l2 {
> > > > -			regulator-min-microvolt = <1800000>;
> > > > -			regulator-max-microvolt = <1800000>;
> > > > +			regulator-min-microvolt = <850000>;
> > > > +			regulator-max-microvolt = <850000>;
> > > >  			regulator-always-on;
> > > >  			regulator-boot-on;
> > > >  		};
> > > >
> > > > base-commit: 5d4809e25903ab8e74034c1f23c787fd26d52934
>
> --
> With best wishes
> Dmitry
Re: [PATCH] arm64: dts: qcom: ipq9574: Fix 'l2' regulator min/max voltages
Posted by Konrad Dybcio 3 months, 2 weeks ago
On 6/24/25 8:08 AM, Varadarajan Narayanan wrote:
> On Mon, Jun 23, 2025 at 01:34:22PM +0200, Konrad Dybcio wrote:
>> On 6/23/25 1:32 PM, Varadarajan Narayanan wrote:
>>> The min and max voltages on l2 regulator is 850000uV. This was
>>> incorrectly set at 1800000uV earlier and that affected the
>>> stability of the networking subsystem.
>>>
>>> Fixes: d5506524d9d9 ("arm64: dts: qcom: ipq9574: Add LDO regulator node")
>>> Signed-off-by: Varadarajan Narayanan <quic_varada@quicinc.com>
>>> ---
>>
>> Shouldn't there be a consumer for it, instead of it being always-on?
> 
> The uniphy block is the consumer of this voltage. The PMIC configures it
> to 850000uV based on OTP settings and s/w doesn't change it as uniphy
> doesn't do scaling.

Is there a chance something in the networking part can probe before
the regulator driver, resulting in a momentary undervolting?

Konrad
Re: [PATCH] arm64: dts: qcom: ipq9574: Fix 'l2' regulator min/max voltages
Posted by Varadarajan Narayanan 3 months, 2 weeks ago
On Tue, Jun 24, 2025 at 04:38:59PM +0200, Konrad Dybcio wrote:
> On 6/24/25 8:08 AM, Varadarajan Narayanan wrote:
> > On Mon, Jun 23, 2025 at 01:34:22PM +0200, Konrad Dybcio wrote:
> >> On 6/23/25 1:32 PM, Varadarajan Narayanan wrote:
> >>> The min and max voltages on l2 regulator is 850000uV. This was
> >>> incorrectly set at 1800000uV earlier and that affected the
> >>> stability of the networking subsystem.
> >>>
> >>> Fixes: d5506524d9d9 ("arm64: dts: qcom: ipq9574: Add LDO regulator node")
> >>> Signed-off-by: Varadarajan Narayanan <quic_varada@quicinc.com>
> >>> ---
> >>
> >> Shouldn't there be a consumer for it, instead of it being always-on?
> >
> > The uniphy block is the consumer of this voltage. The PMIC configures it
> > to 850000uV based on OTP settings and s/w doesn't change it as uniphy
> > doesn't do scaling.
>
> Is there a chance something in the networking part can probe before
> the regulator driver, resulting in a momentary undervolting?

No. The PMIC configures it to 850000uV based on OTP at power up itself,
before Linux driver probe.

Thanks
Varada