[PATCH 0/8] ARM: dts: keystone*: Continued warnings cleanups

[PATCH 0/8] ARM: dts: keystone*: Continued warnings cleanups

[Andrew F. Davis]

Hello all,

Just a couple cleanups for DT warnings when compiling with W=1.

This clears up 51 more warnings when building keystone_defconfig.

Based on Santosh's "for_4.16/keystone-dts" branch.

Thanks,
Andrew

Andrew F. Davis (8):
  ARM: dts: keystone: Move keystone_irq to under device-state-control
  ARM: dts: keystone: Move reset-controller to under
    device-state-control
  ARM: dts: keystone-k2e: Move keystone_dsp_gpio to under
    device-state-control
  ARM: dts: keystone-k2l: Move keystone_dsp_gpio to under
    device-state-control
  ARM: dts: keystone-k2hk: Move keystone_dsp_gpio to under
    device-state-control
  ARM: dts: keystone-k2g: Move keystone_irq to under
    device-state-control
  ARM: dts: keystone-k2g: Move keystone_dsp_gpio to under
    device-state-control
  ARM: dts: keystone-k2g: Move usb-phy nodes out of soc0

 arch/arm/boot/dts/keystone-k2e.dtsi  |  13 +++--
 arch/arm/boot/dts/keystone-k2g.dtsi  |  64 ++++++++++++---------
 arch/arm/boot/dts/keystone-k2hk.dtsi | 104 +++++++++++++++++++----------------
 arch/arm/boot/dts/keystone-k2l.dtsi  |  52 ++++++++++--------
 arch/arm/boot/dts/keystone.dtsi      |  34 +++++++-----
 5 files changed, 151 insertions(+), 116 deletions(-)

-- 
2.15.1

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 1/8] ARM: dts: keystone: Move keystone_irq to under device-state-control

[Andrew F. Davis]

The keystone_irq node describes a device that is a component of the device
state control module. As such, it should not be a member of soc0 bus
but instead a sub-node of device-state-control.

This move also fixes a warning about not having a reg property. Now
that this is a sub-node of device-state-control, a syscon type node,
we add this reg property but relative to the syscon base, this way
when the dt-binding/driver are updated we can drop the non-standard
ti,syscon-dev property completely and simply use get_resource() in
the driver.

Signed-off-by: Andrew F. Davis <afd-l0cyMroinI0@public.gmane.org>
Acked-by: Nishanth Menon <nm-l0cyMroinI0@public.gmane.org>
---
 arch/arm/boot/dts/keystone.dtsi | 21 ++++++++++++---------
 1 file changed, 12 insertions(+), 9 deletions(-)

diff --git a/arch/arm/boot/dts/keystone.dtsi b/arch/arm/boot/dts/keystone.dtsi
index 93ea5c69ea77..158e0a903f7e 100644
--- a/arch/arm/boot/dts/keystone.dtsi
+++ b/arch/arm/boot/dts/keystone.dtsi
@@ -87,8 +87,19 @@
 		};
 
 		devctrl: device-state-control@2620000 {
-			compatible = "ti,keystone-devctrl", "syscon";
+			#address-cells = <1>;
+			#size-cells = <1>;
+			compatible = "ti,keystone-devctrl", "syscon", "simple-mfd";
 			reg = <0x02620000 0x1000>;
+
+			kirq0: keystone_irq@2a0 {
+				compatible = "ti,keystone-irq";
+				reg = <0x2a0 0x4>;
+				interrupts = <GIC_SPI 4 IRQ_TYPE_EDGE_RISING>;
+				interrupt-controller;
+				#interrupt-cells = <1>;
+				ti,syscon-dev = <&devctrl 0x2a0>;
+			};
 		};
 
 		rstctrl: reset-controller {
@@ -282,14 +293,6 @@
 				  1 0 0x21000A00 0x00000100>;
 		};
 
-		kirq0: keystone_irq@26202a0 {
-			compatible = "ti,keystone-irq";
-			interrupts = <GIC_SPI 4 IRQ_TYPE_EDGE_RISING>;
-			interrupt-controller;
-			#interrupt-cells = <1>;
-			ti,syscon-dev = <&devctrl 0x2a0>;
-		};
-
 		pcie0: pcie@21800000 {
 			compatible = "ti,keystone-pcie", "snps,dw-pcie";
 			clocks = <&clkpcie>;
-- 
2.15.1

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 1/8] ARM: dts: keystone: Move keystone_irq to under device-state-control

[Suman Anna]

Hi Andrew,

On 01/02/2018 11:01 AM, Andrew F. Davis wrote:
> The keystone_irq node describes a device that is a component of the device
> state control module. 

I would prefer 'address space' to be added to this statement as this
module is really not a single IP but really a collection of different
register sets providing different functionalities. Some of these
comments apply to the following patches as well.

As such, it should not be a member of soc0 bus
> but instead a sub-node of device-state-control.
> 
> This move also fixes a warning about not having a reg property. Now
> that this is a sub-node of device-state-control, a syscon type node,
> we add this reg property but relative to the syscon base, this way
> when the dt-binding/driver are updated we can drop the non-standard
> ti,syscon-dev property completely and simply use get_resource() in
> the driver.

Please add an appropriate 'ranges' property in the parent node following
the parent-child node convention, it's upto individual drivers to use
the appropriate API for whether they want to deal with the offset or the
actual bus addresses. You should not tie this into forcing to use a
get_resource() without ranges to get the offset.

> 
> Signed-off-by: Andrew F. Davis <afd-l0cyMroinI0@public.gmane.org>
> Acked-by: Nishanth Menon <nm-l0cyMroinI0@public.gmane.org>
> ---
>  arch/arm/boot/dts/keystone.dtsi | 21 ++++++++++++---------
>  1 file changed, 12 insertions(+), 9 deletions(-)
> 
> diff --git a/arch/arm/boot/dts/keystone.dtsi b/arch/arm/boot/dts/keystone.dtsi
> index 93ea5c69ea77..158e0a903f7e 100644
> --- a/arch/arm/boot/dts/keystone.dtsi
> +++ b/arch/arm/boot/dts/keystone.dtsi
> @@ -87,8 +87,19 @@
>  		};
>  
>  		devctrl: device-state-control@2620000 {
> -			compatible = "ti,keystone-devctrl", "syscon";
> +			#address-cells = <1>;
> +			#size-cells = <1>;
> +			compatible = "ti,keystone-devctrl", "syscon", "simple-mfd";

nit, can you please maintain the current order of compatible and reg,
and add the new properties after them.

regards
Suman

>  			reg = <0x02620000 0x1000>;
> +
> +			kirq0: keystone_irq@2a0 {
> +				compatible = "ti,keystone-irq";
> +				reg = <0x2a0 0x4>;
> +				interrupts = <GIC_SPI 4 IRQ_TYPE_EDGE_RISING>;
> +				interrupt-controller;
> +				#interrupt-cells = <1>;
> +				ti,syscon-dev = <&devctrl 0x2a0>;
> +			};
>  		};
>  
>  		rstctrl: reset-controller {
> @@ -282,14 +293,6 @@
>  				  1 0 0x21000A00 0x00000100>;
>  		};
>  
> -		kirq0: keystone_irq@26202a0 {
> -			compatible = "ti,keystone-irq";
> -			interrupts = <GIC_SPI 4 IRQ_TYPE_EDGE_RISING>;
> -			interrupt-controller;
> -			#interrupt-cells = <1>;
> -			ti,syscon-dev = <&devctrl 0x2a0>;
> -		};
> -
>  		pcie0: pcie@21800000 {
>  			compatible = "ti,keystone-pcie", "snps,dw-pcie";
>  			clocks = <&clkpcie>;
> 

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 1/8] ARM: dts: keystone: Move keystone_irq to under device-state-control

[Andrew F. Davis]

On 01/02/2018 05:30 PM, Suman Anna wrote:
> Hi Andrew,
> 
> On 01/02/2018 11:01 AM, Andrew F. Davis wrote:
>> The keystone_irq node describes a device that is a component of the device
>> state control module. 
> 
> I would prefer 'address space' to be added to this statement as this
> module is really not a single IP but really a collection of different
> register sets providing different functionalities. Some of these
> comments apply to the following patches as well.
> 

Works for me, will re-word.

> As such, it should not be a member of soc0 bus
>> but instead a sub-node of device-state-control.
>>
>> This move also fixes a warning about not having a reg property. Now
>> that this is a sub-node of device-state-control, a syscon type node,
>> we add this reg property but relative to the syscon base, this way
>> when the dt-binding/driver are updated we can drop the non-standard
>> ti,syscon-dev property completely and simply use get_resource() in
>> the driver.
> 
> Please add an appropriate 'ranges' property in the parent node following
> the parent-child node convention, it's upto individual drivers to use
> the appropriate API for whether they want to deal with the offset or the
> actual bus addresses. You should not tie this into forcing to use a
> get_resource() without ranges to get the offset.
> 

Will add.

>>
>> Signed-off-by: Andrew F. Davis <afd-l0cyMroinI0@public.gmane.org>
>> Acked-by: Nishanth Menon <nm-l0cyMroinI0@public.gmane.org>
>> ---
>>  arch/arm/boot/dts/keystone.dtsi | 21 ++++++++++++---------
>>  1 file changed, 12 insertions(+), 9 deletions(-)
>>
>> diff --git a/arch/arm/boot/dts/keystone.dtsi b/arch/arm/boot/dts/keystone.dtsi
>> index 93ea5c69ea77..158e0a903f7e 100644
>> --- a/arch/arm/boot/dts/keystone.dtsi
>> +++ b/arch/arm/boot/dts/keystone.dtsi
>> @@ -87,8 +87,19 @@
>>  		};
>>  
>>  		devctrl: device-state-control@2620000 {
>> -			compatible = "ti,keystone-devctrl", "syscon";
>> +			#address-cells = <1>;
>> +			#size-cells = <1>;
>> +			compatible = "ti,keystone-devctrl", "syscon", "simple-mfd";
> 
> nit, can you please maintain the current order of compatible and reg,
> and add the new properties after them.
> 

#address/size-cells are the first properties in many other nodes,
including the top level soc0, I have no real preference, so I'll change
it around if you prefer.

> regards
> Suman
> 
>>  			reg = <0x02620000 0x1000>;
>> +
>> +			kirq0: keystone_irq@2a0 {
>> +				compatible = "ti,keystone-irq";
>> +				reg = <0x2a0 0x4>;
>> +				interrupts = <GIC_SPI 4 IRQ_TYPE_EDGE_RISING>;
>> +				interrupt-controller;
>> +				#interrupt-cells = <1>;
>> +				ti,syscon-dev = <&devctrl 0x2a0>;
>> +			};
>>  		};
>>  
>>  		rstctrl: reset-controller {
>> @@ -282,14 +293,6 @@
>>  				  1 0 0x21000A00 0x00000100>;
>>  		};
>>  
>> -		kirq0: keystone_irq@26202a0 {
>> -			compatible = "ti,keystone-irq";
>> -			interrupts = <GIC_SPI 4 IRQ_TYPE_EDGE_RISING>;
>> -			interrupt-controller;
>> -			#interrupt-cells = <1>;
>> -			ti,syscon-dev = <&devctrl 0x2a0>;
>> -		};
>> -
>>  		pcie0: pcie@21800000 {
>>  			compatible = "ti,keystone-pcie", "snps,dw-pcie";
>>  			clocks = <&clkpcie>;
>>
> 
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 2/8] ARM: dts: keystone: Move reset-controller to under device-state-control

[Andrew F. Davis]

The reset-controller node describes a device that is a component of the
device state control module. As such, it should not be a member of soc0
bus but instead a sub-node of device-state-control.

This move also fixes a warning about not having a reg property. Now
that this is a sub-node of device-state-control, a syscon type node,
we add this reg property but relative to the syscon base, this way
when the dt-binding/driver are updated we can drop the non-standard
ti,syscon-dev property completely and simply use get_resource() in
the driver.

Signed-off-by: Andrew F. Davis <afd-l0cyMroinI0@public.gmane.org>
Acked-by: Nishanth Menon <nm-l0cyMroinI0@public.gmane.org>
---
 arch/arm/boot/dts/keystone.dtsi | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/arch/arm/boot/dts/keystone.dtsi b/arch/arm/boot/dts/keystone.dtsi
index 158e0a903f7e..a36502e94826 100644
--- a/arch/arm/boot/dts/keystone.dtsi
+++ b/arch/arm/boot/dts/keystone.dtsi
@@ -100,13 +100,14 @@
 				#interrupt-cells = <1>;
 				ti,syscon-dev = <&devctrl 0x2a0>;
 			};
-		};
 
-		rstctrl: reset-controller {
-			compatible = "ti,keystone-reset";
-			ti,syscon-pll = <&pllctrl 0xe4>;
-			ti,syscon-dev = <&devctrl 0x328>;
-			ti,wdt-list = <0>;
+			rstctrl: reset-controller@328 {
+				compatible = "ti,keystone-reset";
+				reg = <0x328 0x10>;
+				ti,syscon-pll = <&pllctrl 0xe4>;
+				ti,syscon-dev = <&devctrl 0x328>;
+				ti,wdt-list = <0>;
+			};
 		};
 
 		/include/ "keystone-clocks.dtsi"
-- 
2.15.1

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 3/8] ARM: dts: keystone-k2e: Move keystone_dsp_gpio to under device-state-control

[Andrew F. Davis]

The keystone_dsp_gpio node describes a device that is a component of the
device state control module. As such, it should not be a member of soc0
bus but instead a sub-node of device-state-control.

This move also fixes a warning about not having a reg property. Now
that this is a sub-node of device-state-control, a syscon type node,
we add this reg property but relative to the syscon base, this way
when the dt-binding/driver are updated we can drop the non-standard
gpio,syscon-dev property completely and simply use get_resource() in
the driver.

Signed-off-by: Andrew F. Davis <afd-l0cyMroinI0@public.gmane.org>
Acked-by: Nishanth Menon <nm-l0cyMroinI0@public.gmane.org>
---
 arch/arm/boot/dts/keystone-k2e.dtsi | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/arch/arm/boot/dts/keystone-k2e.dtsi b/arch/arm/boot/dts/keystone-k2e.dtsi
index 0bcd3f8a9c45..085e7326ea8e 100644
--- a/arch/arm/boot/dts/keystone-k2e.dtsi
+++ b/arch/arm/boot/dts/keystone-k2e.dtsi
@@ -109,11 +109,14 @@
 			};
 		};
 
-		dspgpio0: keystone_dsp_gpio@2620240 {
-			compatible = "ti,keystone-dsp-gpio";
-			gpio-controller;
-			#gpio-cells = <2>;
-			gpio,syscon-dev = <&devctrl 0x240>;
+		devctrl: device-state-control@2620000 {
+			dspgpio0: keystone_dsp_gpio@240 {
+				compatible = "ti,keystone-dsp-gpio";
+				reg = <0x240 0x4>;
+				gpio-controller;
+				#gpio-cells = <2>;
+				gpio,syscon-dev = <&devctrl 0x240>;
+			};
 		};
 
 		dsp0: dsp@10800000 {
-- 
2.15.1

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 4/8] ARM: dts: keystone-k2l: Move keystone_dsp_gpio to under device-state-control

[Andrew F. Davis]

The keystone_dsp_gpio nodes describe devices that are a component of the
device state control module. As such, they should not be a member of soc0
bus but instead sub-nodes of device-state-control.

This move also fixes warnings about not having a reg property. Now
that this is a sub-node of device-state-control, a syscon type node,
we add this reg property but relative to the syscon base, this way
when the dt-binding/driver are updated we can drop the non-standard
gpio,syscon-dev property completely and simply use get_resource() in
the driver.

Signed-off-by: Andrew F. Davis <afd-l0cyMroinI0@public.gmane.org>
Acked-by: Nishanth Menon <nm-l0cyMroinI0@public.gmane.org>
---
 arch/arm/boot/dts/keystone-k2l.dtsi | 52 +++++++++++++++++++++----------------
 1 file changed, 29 insertions(+), 23 deletions(-)

diff --git a/arch/arm/boot/dts/keystone-k2l.dtsi b/arch/arm/boot/dts/keystone-k2l.dtsi
index b61a830f4a4d..374c80124c4e 100644
--- a/arch/arm/boot/dts/keystone-k2l.dtsi
+++ b/arch/arm/boot/dts/keystone-k2l.dtsi
@@ -289,32 +289,38 @@
 			clocks = <&clkosr>;
 		};
 
-		dspgpio0: keystone_dsp_gpio@2620240 {
-			compatible = "ti,keystone-dsp-gpio";
-			gpio-controller;
-			#gpio-cells = <2>;
-			gpio,syscon-dev = <&devctrl 0x240>;
-		};
+		devctrl: device-state-control@2620000 {
+			dspgpio0: keystone_dsp_gpio@240 {
+				compatible = "ti,keystone-dsp-gpio";
+				reg = <0x240 0x4>;
+				gpio-controller;
+				#gpio-cells = <2>;
+				gpio,syscon-dev = <&devctrl 0x240>;
+			};
 
-		dspgpio1: keystone_dsp_gpio@2620244 {
-			compatible = "ti,keystone-dsp-gpio";
-			gpio-controller;
-			#gpio-cells = <2>;
-			gpio,syscon-dev = <&devctrl 0x244>;
-		};
+			dspgpio1: keystone_dsp_gpio@244 {
+				compatible = "ti,keystone-dsp-gpio";
+				reg = <0x244 0x4>;
+				gpio-controller;
+				#gpio-cells = <2>;
+				gpio,syscon-dev = <&devctrl 0x244>;
+			};
 
-		dspgpio2: keystone_dsp_gpio@2620248 {
-			compatible = "ti,keystone-dsp-gpio";
-			gpio-controller;
-			#gpio-cells = <2>;
-			gpio,syscon-dev = <&devctrl 0x248>;
-		};
+			dspgpio2: keystone_dsp_gpio@248 {
+				compatible = "ti,keystone-dsp-gpio";
+				reg = <0x248 0x4>;
+				gpio-controller;
+				#gpio-cells = <2>;
+				gpio,syscon-dev = <&devctrl 0x248>;
+			};
 
-		dspgpio3: keystone_dsp_gpio@262024c {
-			compatible = "ti,keystone-dsp-gpio";
-			gpio-controller;
-			#gpio-cells = <2>;
-			gpio,syscon-dev = <&devctrl 0x24c>;
+			dspgpio3: keystone_dsp_gpio@24c {
+				compatible = "ti,keystone-dsp-gpio";
+				reg = <0x24c 0x4>;
+				gpio-controller;
+				#gpio-cells = <2>;
+				gpio,syscon-dev = <&devctrl 0x24c>;
+			};
 		};
 
 		dsp0: dsp@10800000 {
-- 
2.15.1

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 5/8] ARM: dts: keystone-k2hk: Move keystone_dsp_gpio to under device-state-control

[Andrew F. Davis]

The keystone_dsp_gpio nodes describe devices that are a component of the
device state control module. As such, they should not be a member of soc0
bus but instead sub-nodes of device-state-control.

This move also fixes warnings about not having a reg property. Now
that this is a sub-node of device-state-control, a syscon type node,
we add this reg property but relative to the syscon base, this way
when the dt-binding/driver are updated we can drop the non-standard
gpio,syscon-dev property completely and simply use get_resource() in
the driver.

Signed-off-by: Andrew F. Davis <afd-l0cyMroinI0@public.gmane.org>
Acked-by: Nishanth Menon <nm-l0cyMroinI0@public.gmane.org>
---
 arch/arm/boot/dts/keystone-k2hk.dtsi | 104 +++++++++++++++++++----------------
 1 file changed, 57 insertions(+), 47 deletions(-)

diff --git a/arch/arm/boot/dts/keystone-k2hk.dtsi b/arch/arm/boot/dts/keystone-k2hk.dtsi
index ed59474522cb..ca0f198ba627 100644
--- a/arch/arm/boot/dts/keystone-k2hk.dtsi
+++ b/arch/arm/boot/dts/keystone-k2hk.dtsi
@@ -87,60 +87,70 @@
 			};
 		};
 
-		dspgpio0: keystone_dsp_gpio@2620240 {
-			compatible = "ti,keystone-dsp-gpio";
-			gpio-controller;
-			#gpio-cells = <2>;
-			gpio,syscon-dev = <&devctrl 0x240>;
-		};
+		devctrl: device-state-control@2620000 {
+			dspgpio0: keystone_dsp_gpio@240 {
+				compatible = "ti,keystone-dsp-gpio";
+				reg = <0x240 0x4>;
+				gpio-controller;
+				#gpio-cells = <2>;
+				gpio,syscon-dev = <&devctrl 0x240>;
+			};
 
-		dspgpio1: keystone_dsp_gpio@2620244 {
-			compatible = "ti,keystone-dsp-gpio";
-			gpio-controller;
-			#gpio-cells = <2>;
-			gpio,syscon-dev = <&devctrl 0x244>;
-		};
+			dspgpio1: keystone_dsp_gpio@244 {
+				compatible = "ti,keystone-dsp-gpio";
+				reg = <0x244 0x4>;
+				gpio-controller;
+				#gpio-cells = <2>;
+				gpio,syscon-dev = <&devctrl 0x244>;
+			};
 
-		dspgpio2: keystone_dsp_gpio@2620248 {
-			compatible = "ti,keystone-dsp-gpio";
-			gpio-controller;
-			#gpio-cells = <2>;
-			gpio,syscon-dev = <&devctrl 0x248>;
-		};
+			dspgpio2: keystone_dsp_gpio@248 {
+				compatible = "ti,keystone-dsp-gpio";
+				reg = <0x248 0x4>;
+				gpio-controller;
+				#gpio-cells = <2>;
+				gpio,syscon-dev = <&devctrl 0x248>;
+			};
 
-		dspgpio3: keystone_dsp_gpio@262024c {
-			compatible = "ti,keystone-dsp-gpio";
-			gpio-controller;
-			#gpio-cells = <2>;
-			gpio,syscon-dev = <&devctrl 0x24c>;
-		};
+			dspgpio3: keystone_dsp_gpio@24c {
+				compatible = "ti,keystone-dsp-gpio";
+				reg = <0x24c 0x4>;
+				gpio-controller;
+				#gpio-cells = <2>;
+				gpio,syscon-dev = <&devctrl 0x24c>;
+			};
 
-		dspgpio4: keystone_dsp_gpio@2620250 {
-			compatible = "ti,keystone-dsp-gpio";
-			gpio-controller;
-			#gpio-cells = <2>;
-			gpio,syscon-dev = <&devctrl 0x250>;
-		};
+			dspgpio4: keystone_dsp_gpio@250 {
+				compatible = "ti,keystone-dsp-gpio";
+				reg = <0x250 0x4>;
+				gpio-controller;
+				#gpio-cells = <2>;
+				gpio,syscon-dev = <&devctrl 0x250>;
+			};
 
-		dspgpio5: keystone_dsp_gpio@2620254 {
-			compatible = "ti,keystone-dsp-gpio";
-			gpio-controller;
-			#gpio-cells = <2>;
-			gpio,syscon-dev = <&devctrl 0x254>;
-		};
+			dspgpio5: keystone_dsp_gpio@254 {
+				compatible = "ti,keystone-dsp-gpio";
+				reg = <0x254 0x4>;
+				gpio-controller;
+				#gpio-cells = <2>;
+				gpio,syscon-dev = <&devctrl 0x254>;
+			};
 
-		dspgpio6: keystone_dsp_gpio@2620258 {
-			compatible = "ti,keystone-dsp-gpio";
-			gpio-controller;
-			#gpio-cells = <2>;
-			gpio,syscon-dev = <&devctrl 0x258>;
-		};
+			dspgpio6: keystone_dsp_gpio@258 {
+				compatible = "ti,keystone-dsp-gpio";
+				reg = <0x258 0x4>;
+				gpio-controller;
+				#gpio-cells = <2>;
+				gpio,syscon-dev = <&devctrl 0x258>;
+			};
 
-		dspgpio7: keystone_dsp_gpio@262025c {
-			compatible = "ti,keystone-dsp-gpio";
-			gpio-controller;
-			#gpio-cells = <2>;
-			gpio,syscon-dev = <&devctrl 0x25c>;
+			dspgpio7: keystone_dsp_gpio@25c {
+				compatible = "ti,keystone-dsp-gpio";
+				reg = <0x25c 0x4>;
+				gpio-controller;
+				#gpio-cells = <2>;
+				gpio,syscon-dev = <&devctrl 0x25c>;
+			};
 		};
 
 		dsp0: dsp@10800000 {
-- 
2.15.1

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 6/8] ARM: dts: keystone-k2g: Move keystone_irq to under device-state-control

[Andrew F. Davis]

The keystone_irq node describes a device that is a component of the device
state control module. As such, it should not be a member of soc0 bus
but instead a sub-node of device-state-control.

This move also fixes a warning about not having a reg property. Now
that this is a sub-node of device-state-control, a syscon type node,
we add this reg property but relative to the syscon base, this way
when the dt-binding/driver are updated we can drop the non-standard
ti,syscon-dev property completely and simply use get_resource() in
the driver.

Signed-off-by: Andrew F. Davis <afd-l0cyMroinI0@public.gmane.org>
Acked-by: Nishanth Menon <nm-l0cyMroinI0@public.gmane.org>
---
 arch/arm/boot/dts/keystone-k2g.dtsi | 21 ++++++++++++---------
 1 file changed, 12 insertions(+), 9 deletions(-)

diff --git a/arch/arm/boot/dts/keystone-k2g.dtsi b/arch/arm/boot/dts/keystone-k2g.dtsi
index fd061718dc0a..19f938e3c27e 100644
--- a/arch/arm/boot/dts/keystone-k2g.dtsi
+++ b/arch/arm/boot/dts/keystone-k2g.dtsi
@@ -97,8 +97,19 @@
 		};
 
 		devctrl: device-state-control@2620000 {
-			compatible = "ti,keystone-devctrl", "syscon";
+			#address-cells = <1>;
+			#size-cells = <1>;
+			compatible = "ti,keystone-devctrl", "syscon", "simple-mfd";
 			reg = <0x02620000 0x1000>;
+
+			kirq0: keystone_irq@2a0 {
+				compatible = "ti,keystone-irq";
+				reg = <0x2a0 0x10>;
+				interrupts = <GIC_SPI 1 IRQ_TYPE_EDGE_RISING>;
+				interrupt-controller;
+				#interrupt-cells = <1>;
+				ti,syscon-dev = <&devctrl 0x2a0>;
+			};
 		};
 
 		uart0: serial@2530c00 {
@@ -188,14 +199,6 @@
 			status = "disabled";
 		};
 
-		kirq0: keystone_irq@26202a0 {
-			compatible = "ti,keystone-irq";
-			interrupts = <GIC_SPI 1 IRQ_TYPE_EDGE_RISING>;
-			interrupt-controller;
-			#interrupt-cells = <1>;
-			ti,syscon-dev = <&devctrl 0x2a0>;
-		};
-
 		dspgpio0: keystone_dsp_gpio@2620240 {
 			compatible = "ti,keystone-dsp-gpio";
 			gpio-controller;
-- 
2.15.1

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 7/8] ARM: dts: keystone-k2g: Move keystone_dsp_gpio to under device-state-control

[Andrew F. Davis]

The keystone_dsp_gpio node describes a device that is a component of the
device state control module. As such, it should not be a member of soc0
bus but instead a sub-node of device-state-control.

This move also fixes a warning about not having a reg property. Now
that this is a sub-node of device-state-control, a syscon type node,
we add this reg property but relative to the syscon base, this way
when the dt-binding/driver are updated we can drop the non-standard
gpio,syscon-dev property completely and simply use get_resource() in
the driver.

Signed-off-by: Andrew F. Davis <afd-l0cyMroinI0@public.gmane.org>
Acked-by: Nishanth Menon <nm-l0cyMroinI0@public.gmane.org>
---
 arch/arm/boot/dts/keystone-k2g.dtsi | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/arch/arm/boot/dts/keystone-k2g.dtsi b/arch/arm/boot/dts/keystone-k2g.dtsi
index 19f938e3c27e..71a8588f42e4 100644
--- a/arch/arm/boot/dts/keystone-k2g.dtsi
+++ b/arch/arm/boot/dts/keystone-k2g.dtsi
@@ -110,6 +110,14 @@
 				#interrupt-cells = <1>;
 				ti,syscon-dev = <&devctrl 0x2a0>;
 			};
+
+			dspgpio0: keystone_dsp_gpio@240 {
+				compatible = "ti,keystone-dsp-gpio";
+				reg = <0x240 0x4>;
+				gpio-controller;
+				#gpio-cells = <2>;
+				gpio,syscon-dev = <&devctrl 0x240>;
+			};
 		};
 
 		uart0: serial@2530c00 {
@@ -199,13 +207,6 @@
 			status = "disabled";
 		};
 
-		dspgpio0: keystone_dsp_gpio@2620240 {
-			compatible = "ti,keystone-dsp-gpio";
-			gpio-controller;
-			#gpio-cells = <2>;
-			gpio,syscon-dev = <&devctrl 0x240>;
-		};
-
 		dsp0: dsp@10800000 {
 			compatible = "ti,k2g-dsp";
 			reg = <0x10800000 0x00100000>,
-- 
2.15.1

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 8/8] ARM: dts: keystone-k2g: Move usb-phy nodes out of soc0

[Andrew F. Davis]

The two usb-phy nodes are not memory-mapped devices on the
SoC but rather dummy phys for SoCs with transparent autonomous
NOP phys. Move these to outside soc0 to their own node like
done elsewhere.

Signed-off-by: Andrew F. Davis <afd-l0cyMroinI0@public.gmane.org>
Acked-by: Nishanth Menon <nm-l0cyMroinI0@public.gmane.org>
---
 arch/arm/boot/dts/keystone-k2g.dtsi | 28 ++++++++++++++++++----------
 1 file changed, 18 insertions(+), 10 deletions(-)

diff --git a/arch/arm/boot/dts/keystone-k2g.dtsi b/arch/arm/boot/dts/keystone-k2g.dtsi
index 71a8588f42e4..a8cce19952b8 100644
--- a/arch/arm/boot/dts/keystone-k2g.dtsi
+++ b/arch/arm/boot/dts/keystone-k2g.dtsi
@@ -69,6 +69,24 @@
 		interrupts = <GIC_SPI 4 IRQ_TYPE_EDGE_RISING>;
 	};
 
+	usbphy {
+		#address-cells = <1>;
+		#size-cells = <0>;
+		compatible = "simple-bus";
+
+		usb0_phy: usb-phy@0 {
+			compatible = "usb-nop-xceiv";
+			reg = <0>;
+			status = "disabled";
+		};
+
+		usb1_phy: usb-phy@1 {
+			compatible = "usb-nop-xceiv";
+			reg = <1>;
+			status = "disabled";
+		};
+	};
+
 	soc0: soc@0 {
 		#address-cells = <1>;
 		#size-cells = <1>;
@@ -464,11 +482,6 @@
 			status = "disabled";
 		};
 
-		usb0_phy: usb-phy@0 {
-			compatible = "usb-nop-xceiv";
-			status = "disabled";
-		};
-
 		keystone_usb0: keystone-dwc3@2680000 {
 			compatible = "ti,keystone-dwc3";
 			#address-cells = <1>;
@@ -492,11 +505,6 @@
 			};
 		};
 
-		usb1_phy: usb-phy@1 {
-			compatible = "usb-nop-xceiv";
-			status = "disabled";
-		};
-
 		keystone_usb1: keystone-dwc3@2580000 {
 			compatible = "ti,keystone-dwc3";
 			#address-cells = <1>;
-- 
2.15.1

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 0/8] ARM: dts: keystone*: Continued warnings cleanups

[Santosh Shilimkar]

On 1/2/2018 9:01 AM, Andrew F. Davis wrote:
> Hello all,
> 
> Just a couple cleanups for DT warnings when compiling with W=1.
> 
> This clears up 51 more warnings when building keystone_defconfig.
> 
> Based on Santosh's "for_4.16/keystone-dts" branch.
> 
Will see if this can be queued up post merge window as 'non-critical'
fixes o.w it waits for 4.17. Please repost it after addressing
Suman's comments.

Regards,
Santosh