| Message ID | 20240926101804.22471-1-macpaul.lin@mediatek.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | dt-bindings: phy: mediatek: tphy: add a property for power-domains | expand |
Il 26/09/24 12:18, Macpaul Lin ha scritto: > Some platforms requires a dependency for power-domains. > So we add property 'power-domains' and set 'maxItems: 1' in the > DT Schema. > > Signed-off-by: Macpaul Lin <macpaul.lin@mediatek.com> Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
On Thu, Sep 26, 2024 at 06:18:04PM +0800, Macpaul Lin wrote: > Some platforms requires a dependency for power-domains. Some, so not all? Why isn't this restricted on a per compatible basis? > So we add property 'power-domains' and set 'maxItems: 1' in the > DT Schema. > > Signed-off-by: Macpaul Lin <macpaul.lin@mediatek.com> > --- > Documentation/devicetree/bindings/phy/mediatek,tphy.yaml | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml b/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml > index 423b7c4e62f2..c77fe43c224a 100644 > --- a/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml > +++ b/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml > @@ -125,6 +125,9 @@ properties: > $ref: /schemas/types.yaml#/definitions/uint32 > default: 28 > > + power-domains: > + maxItems: 1 > + > # Required child node: > patternProperties: > "^(usb|pcie|sata)-phy@[0-9a-f]+$": > -- > 2.45.2 >
On 9/27/24 00:25, Conor Dooley wrote: > On Thu, Sep 26, 2024 at 06:18:04PM +0800, Macpaul Lin wrote: >> Some platforms requires a dependency for power-domains. > > Some, so not all? Why isn't this restricted on a per compatible basis? After discussion with Chunfeng and double check tphy design in detail. Chunfeng commented that tphy dose not need to add mtcmos. It is not necessary to add it, if the power of the phy is turned off, it will affect other functions. From the current USB hardware design perspective, even if mtcmos is added to the phy, it is always on. >> So we add property 'power-domains' and set 'maxItems: 1' in the >> DT Schema. >> >> Signed-off-by: Macpaul Lin <macpaul.lin@mediatek.com> >> --- >> Documentation/devicetree/bindings/phy/mediatek,tphy.yaml | 3 +++ >> 1 file changed, 3 insertions(+) >> >> diff --git a/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml b/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml >> index 423b7c4e62f2..c77fe43c224a 100644 >> --- a/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml >> +++ b/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml >> @@ -125,6 +125,9 @@ properties: >> $ref: /schemas/types.yaml#/definitions/uint32 >> default: 28 >> >> + power-domains: >> + maxItems: 1 >> + >> # Required child node: >> patternProperties: >> "^(usb|pcie|sata)-phy@[0-9a-f]+$": >> -- >> 2.45.2 >> Please drop this patch and I'll send a new fix to mt8195.dtsi. Thanks Macpaul Lin
On Tue, Oct 08, 2024 at 12:03:37PM +0800, Macpaul Lin wrote: > > > On 9/27/24 00:25, Conor Dooley wrote: > > On Thu, Sep 26, 2024 at 06:18:04PM +0800, Macpaul Lin wrote: > > > Some platforms requires a dependency for power-domains. > > > > Some, so not all? Why isn't this restricted on a per compatible basis? > > After discussion with Chunfeng and double check tphy design in detail. > Chunfeng commented that tphy dose not need to add mtcmos. > It is not necessary to add it, if the power of the phy is turned off, > it will affect other functions. > > From the current USB hardware design perspective, even if mtcmos > is added to the phy, it is always on. Firstly, I have no idea what "mtcmos" means, sorry. I am a dt-bindings guy, not someone familiar with mediatek hardware. Secondly, it sounds like this /does/ have a power domain, so it should be in the binding. That it is turned on by something else and must remain on doesn't mean it shouldn't be documented here. What if those things try to turn it off while the tphy is using it? > > > > So we add property 'power-domains' and set 'maxItems: 1' in the > > > DT Schema. > > > > > > Signed-off-by: Macpaul Lin <macpaul.lin@mediatek.com> > > > --- > > > Documentation/devicetree/bindings/phy/mediatek,tphy.yaml | 3 +++ > > > 1 file changed, 3 insertions(+) > > > > > > diff --git a/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml b/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml > > > index 423b7c4e62f2..c77fe43c224a 100644 > > > --- a/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml > > > +++ b/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml > > > @@ -125,6 +125,9 @@ properties: > > > $ref: /schemas/types.yaml#/definitions/uint32 > > > default: 28 > > > + power-domains: > > > + maxItems: 1 > > > + > > > # Required child node: > > > patternProperties: > > > "^(usb|pcie|sata)-phy@[0-9a-f]+$": > > > -- > > > 2.45.2 > > > > > Please drop this patch and I'll send a new fix to mt8195.dtsi. > > Thanks > Macpaul Lin
On 10/9/24 00:31, Conor Dooley wrote: > On Tue, Oct 08, 2024 at 12:03:37PM +0800, Macpaul Lin wrote: >> >> >> On 9/27/24 00:25, Conor Dooley wrote: >>> On Thu, Sep 26, 2024 at 06:18:04PM +0800, Macpaul Lin wrote: >>>> Some platforms requires a dependency for power-domains. >>> >>> Some, so not all? Why isn't this restricted on a per compatible basis? >> >> After discussion with Chunfeng and double check tphy design in detail. >> Chunfeng commented that tphy dose not need to add mtcmos. >> It is not necessary to add it, if the power of the phy is turned off, >> it will affect other functions. >> >> From the current USB hardware design perspective, even if mtcmos >> is added to the phy, it is always on. > > Firstly, I have no idea what "mtcmos" means, sorry. I am a dt-bindings > guy, not someone familiar with mediatek hardware. > Secondly, it sounds like this /does/ have a power domain, so it should > be in the binding. That it is turned on by something else and must > remain on doesn't mean it shouldn't be documented here. What if those > things try to turn it off while the tphy is using it? After discussion in MediaTek internal. Chunfeng agreed to add 'power-domains' property to DT schema. The v2 patch will also adding the reminder in the description for developers. >> >>>> So we add property 'power-domains' and set 'maxItems: 1' in the >>>> DT Schema. >>>> >>>> Signed-off-by: Macpaul Lin <macpaul.lin@mediatek.com> >>>> --- >>>> Documentation/devicetree/bindings/phy/mediatek,tphy.yaml | 3 +++ >>>> 1 file changed, 3 insertions(+) >>>> >>>> diff --git a/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml b/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml >>>> index 423b7c4e62f2..c77fe43c224a 100644 >>>> --- a/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml >>>> +++ b/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml >>>> @@ -125,6 +125,9 @@ properties: >>>> $ref: /schemas/types.yaml#/definitions/uint32 >>>> default: 28 >>>> + power-domains: >>>> + maxItems: 1 >>>> + >>>> # Required child node: >>>> patternProperties: >>>> "^(usb|pcie|sata)-phy@[0-9a-f]+$": >>>> -- >>>> 2.45.2 >>>> >> [snip] V2 of this patch has been send. Thanks Macpaul Lin.
diff --git a/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml b/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml index 423b7c4e62f2..c77fe43c224a 100644 --- a/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml +++ b/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml @@ -125,6 +125,9 @@ properties: $ref: /schemas/types.yaml#/definitions/uint32 default: 28 + power-domains: + maxItems: 1 + # Required child node: patternProperties: "^(usb|pcie|sata)-phy@[0-9a-f]+$":
Some platforms requires a dependency for power-domains. So we add property 'power-domains' and set 'maxItems: 1' in the DT Schema. Signed-off-by: Macpaul Lin <macpaul.lin@mediatek.com> --- Documentation/devicetree/bindings/phy/mediatek,tphy.yaml | 3 +++ 1 file changed, 3 insertions(+)