[PATCH v2 0/8] Sigmastar DT warnings fixing

[PATCH v2 0/8] Sigmastar DT warnings fixing

[Daniel Palmer]

v2 heads up: I didn't ignore Krzysztof's request to squash some commits,
I wasn't sure what to squash and I thought since I've fixed everything now
it was better to send a v2 sooner than later.

I've now fixed all of the warnings except one remaining one that seems to
need a change in dt-schema itself to cleanup:

serial@221000 (ns16550a): reg-shift: 3 is not one of [0, 1, 2]
        from schema $id: http://devicetree.org/schemas/reg.yaml

The way registers are connected in these chips is weird. I think
they took their peripheral block from an 8051 microcontroller
and jammed it onto 32bit CPUs so all of the registers have
weird spacing and it really does need reg-shift == 3.

v1 blurb:
    
I saw a slide at LPC that had sigmastar in the "hall of shame"
for DT warnings. I finally got some time to look at this so
I have fixed the ones that are obvious and quick to fix.

I'll fix the rest but they might take a bit longer i.e. one
of them is about the register shift for the UART registers,
that'll need the binding to be updated I think?

FWIW this platform isn't dead and I have been rebasing all
of the out of tree stuff that make it useful (Audio support,
display, hardware video decoding etc) but its basically stuck
on rewriting the clock driver so that it is mainline-able.

Some other people have taken the limited support in mainline
and my out of tree stuff and made it run on old MStar MIPS
machines too..

change log:

v2:
 - Fix the regulator names as suggested by Krzysztof
 - Adjust the opp table name                
 - Add a commit to fix the cpupll binding and clean up the         
   warnings for that
 - Add a commit to fix the memory node in the infinity2m 
   dtsi to fix up the warnings about that.
 - Add a commit to switch the default led trigger on breadbee
   from the undocumented activity to cpu to fix up the
   warnings for the that.

Daniel Palmer (8):
  dt-bindings: clk: mstar msc313 cpupll: Correct clock-cells value
  arm: dts: mstar: breadbee: Correct regulator names
  arm: dts: mstar: breadbee: Correct LED names
  arm: dts: mstar: breadbee: Switch from "activity" led trigger to "cpu"
  arm: dts: mstar: infinity: Correct opp table name
  arm: dts: mstar: infinity2m: ssd202d: Correct memory node name
  arm: dts: mstar: v7: Move the reboot node
  arm: dts: mstar: v7: Add sizes and ranges to sram

 .../bindings/clock/mstar,msc313-cpupll.yaml     |  5 ++---
 .../mstar-infinity-breadbee-common.dtsi         | 12 ++++++------
 arch/arm/boot/dts/sigmastar/mstar-infinity.dtsi |  2 +-
 .../dts/sigmastar/mstar-infinity2m-ssd202d.dtsi |  2 +-
 arch/arm/boot/dts/sigmastar/mstar-v7.dtsi       | 17 ++++++++++-------
 5 files changed, 20 insertions(+), 18 deletions(-)

-- 
2.51.0

[PATCH v2 1/8] dt-bindings: clk: mstar msc313 cpupll: Correct clock-cells value

[Daniel Palmer]

clock-cells should have been 0 and causes warnings when
validating devicetrees.

Fix the value and fix the binding example.

Signed-off-by: Daniel Palmer <daniel@0x0f.com>
---
 .../devicetree/bindings/clock/mstar,msc313-cpupll.yaml       | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/Documentation/devicetree/bindings/clock/mstar,msc313-cpupll.yaml b/Documentation/devicetree/bindings/clock/mstar,msc313-cpupll.yaml
index a9ad7ab5230c..889419fba269 100644
--- a/Documentation/devicetree/bindings/clock/mstar,msc313-cpupll.yaml
+++ b/Documentation/devicetree/bindings/clock/mstar,msc313-cpupll.yaml
@@ -18,7 +18,7 @@ properties:
     const: mstar,msc313-cpupll
 
   "#clock-cells":
-    const: 1
+    const: 0
 
   clocks:
     maxItems: 1
@@ -36,10 +36,9 @@ additionalProperties: false
 
 examples:
   - |
-    #include <dt-bindings/clock/mstar-msc313-mpll.h>
     cpupll: cpupll@206400 {
         compatible = "mstar,msc313-cpupll";
         reg = <0x206400 0x200>;
-        #clock-cells = <1>;
+        #clock-cells = <0>;
         clocks = <&mpll MSTAR_MSC313_MPLL_DIV2>;
     };
-- 
2.51.0

[PATCH v2 2/8] arm: dts: mstar: breadbee: Correct regulator names

[Daniel Palmer]

The regulator names had a "@<n>" in them so the dt validation
stuff wanted a reg property. Fix the names to match what the regulator
is for and remove the "@<n>" bit.

Signed-off-by: Daniel Palmer <daniel@0x0f.com>
---
 .../boot/dts/sigmastar/mstar-infinity-breadbee-common.dtsi  | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/arm/boot/dts/sigmastar/mstar-infinity-breadbee-common.dtsi b/arch/arm/boot/dts/sigmastar/mstar-infinity-breadbee-common.dtsi
index 507ff2fba837..99f5a5603dc7 100644
--- a/arch/arm/boot/dts/sigmastar/mstar-infinity-breadbee-common.dtsi
+++ b/arch/arm/boot/dts/sigmastar/mstar-infinity-breadbee-common.dtsi
@@ -7,7 +7,7 @@
 #include <dt-bindings/gpio/gpio.h>
 
 / {
-	vcc_core: fixedregulator@0 {
+	vcc_core: regulator-core {
 		compatible = "regulator-fixed";
 		regulator-name = "vcc_core";
 		regulator-min-microvolt = <1000000>;
@@ -15,7 +15,7 @@ vcc_core: fixedregulator@0 {
 		regulator-boot-on;
 	};
 
-	vcc_dram: fixedregulator@1 {
+	vcc_dram: regulator-dram {
 		compatible = "regulator-fixed";
 		regulator-name = "vcc_dram";
 		regulator-min-microvolt = <1800000>;
@@ -23,7 +23,7 @@ vcc_dram: fixedregulator@1 {
 		regulator-boot-on;
 	};
 
-	vcc_io: fixedregulator@2 {
+	vcc_io: regulator-io {
 		compatible = "regulator-fixed";
 		regulator-name = "vcc_io";
 		regulator-min-microvolt = <3300000>;
-- 
2.51.0

[PATCH v2 3/8] arm: dts: mstar: breadbee: Correct LED names

[Daniel Palmer]

The dt validation stuff whats the names of the leds to match
a regexp.

Prefix the names with "led-" to make the validation happy.

Signed-off-by: Daniel Palmer <daniel@0x0f.com>
---
 .../boot/dts/sigmastar/mstar-infinity-breadbee-common.dtsi    | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arm/boot/dts/sigmastar/mstar-infinity-breadbee-common.dtsi b/arch/arm/boot/dts/sigmastar/mstar-infinity-breadbee-common.dtsi
index 99f5a5603dc7..71da10d9228f 100644
--- a/arch/arm/boot/dts/sigmastar/mstar-infinity-breadbee-common.dtsi
+++ b/arch/arm/boot/dts/sigmastar/mstar-infinity-breadbee-common.dtsi
@@ -33,11 +33,11 @@ vcc_io: regulator-io {
 
 	leds {
 		compatible = "gpio-leds";
-		red {
+		led-red {
 			gpios = <&gpio MSC313_GPIO_SR_IO16 GPIO_ACTIVE_HIGH>;
 			linux,default-trigger = "activity";
 		};
-		yellow {
+		led-yellow {
 			gpios = <&gpio MSC313_GPIO_SR_IO17 GPIO_ACTIVE_HIGH>;
 			linux,default-trigger = "heartbeat";
 		};
-- 
2.51.0

[PATCH v2 4/8] arm: dts: mstar: breadbee: Switch from "activity" led trigger to "cpu"

[Daniel Palmer]

The activity trigger exists but is not documented in the LED binding.
A patch was sent to document it but wasn't accepted.

The intention of the LED is to show if the system is busy or not
so the cpu trigger is good enough so switch to that to clean up
the warning.

Link: https://lore.kernel.org/all/20210923065500.2284347-4-daniel@0x0f.com/
Signed-off-by: Daniel Palmer <daniel@0x0f.com>
---
 arch/arm/boot/dts/sigmastar/mstar-infinity-breadbee-common.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/boot/dts/sigmastar/mstar-infinity-breadbee-common.dtsi b/arch/arm/boot/dts/sigmastar/mstar-infinity-breadbee-common.dtsi
index 71da10d9228f..0f03f471a344 100644
--- a/arch/arm/boot/dts/sigmastar/mstar-infinity-breadbee-common.dtsi
+++ b/arch/arm/boot/dts/sigmastar/mstar-infinity-breadbee-common.dtsi
@@ -35,7 +35,7 @@ leds {
 		compatible = "gpio-leds";
 		led-red {
 			gpios = <&gpio MSC313_GPIO_SR_IO16 GPIO_ACTIVE_HIGH>;
-			linux,default-trigger = "activity";
+			linux,default-trigger = "cpu";
 		};
 		led-yellow {
 			gpios = <&gpio MSC313_GPIO_SR_IO17 GPIO_ACTIVE_HIGH>;
-- 
2.51.0

[PATCH v2 5/8] arm: dts: mstar: infinity: Correct opp table name

[Daniel Palmer]

The dt validation stuff expects the name of the opp table
to match a regexp and it currently doesn't.

Update the name to match the regexp.

Signed-off-by: Daniel Palmer <daniel@0x0f.com>
---
 arch/arm/boot/dts/sigmastar/mstar-infinity.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/boot/dts/sigmastar/mstar-infinity.dtsi b/arch/arm/boot/dts/sigmastar/mstar-infinity.dtsi
index 441a917b88ba..40e38d995542 100644
--- a/arch/arm/boot/dts/sigmastar/mstar-infinity.dtsi
+++ b/arch/arm/boot/dts/sigmastar/mstar-infinity.dtsi
@@ -9,7 +9,7 @@
 #include <dt-bindings/gpio/msc313-gpio.h>
 
 / {
-	cpu0_opp_table: opp_table0 {
+	cpu0_opp_table: opp-table {
 		compatible = "operating-points-v2";
 		opp-shared;
 
-- 
2.51.0

[PATCH v2 6/8] arm: dts: mstar: infinity2m: ssd202d: Correct memory node name

[Daniel Palmer]

Memory node name didn't include the address which was causing
validation for all of the boards using the dtsi.

Signed-off-by: Daniel Palmer <daniel@0x0f.com>
---
 arch/arm/boot/dts/sigmastar/mstar-infinity2m-ssd202d.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/boot/dts/sigmastar/mstar-infinity2m-ssd202d.dtsi b/arch/arm/boot/dts/sigmastar/mstar-infinity2m-ssd202d.dtsi
index 176e10a29896..74621a7628b2 100644
--- a/arch/arm/boot/dts/sigmastar/mstar-infinity2m-ssd202d.dtsi
+++ b/arch/arm/boot/dts/sigmastar/mstar-infinity2m-ssd202d.dtsi
@@ -7,7 +7,7 @@
 #include "mstar-infinity2m-ssd20xd.dtsi"
 
 / {
-	memory {
+	memory@20000000 {
 		device_type = "memory";
 		reg = <0x20000000 0x8000000>;
 	};
-- 
2.51.0

[PATCH v2 7/8] arm: dts: mstar: v7: Move the reboot node

[Daniel Palmer]

The reboot node is currently inside of soc and that makes
the dt validation stuff complain that it wants some properties
that don't exist for reboot which is just some glue and not
a real device.

Move the node out of soc to fix the warnings.

Signed-off-by: Daniel Palmer <daniel@0x0f.com>
---
 arch/arm/boot/dts/sigmastar/mstar-v7.dtsi | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/arch/arm/boot/dts/sigmastar/mstar-v7.dtsi b/arch/arm/boot/dts/sigmastar/mstar-v7.dtsi
index 3eeafd8c7121..db661e189eec 100644
--- a/arch/arm/boot/dts/sigmastar/mstar-v7.dtsi
+++ b/arch/arm/boot/dts/sigmastar/mstar-v7.dtsi
@@ -73,6 +73,13 @@ xtal_div2: xtal_div2 {
 		};
 	};
 
+	reboot {
+		compatible = "syscon-reboot";
+		regmap = <&pmsleep>;
+		offset = <0xb8>;
+		mask = <0x79>;
+	};
+
 	soc: soc {
 		compatible = "simple-bus";
 		#address-cells = <1>;
@@ -105,13 +112,6 @@ pmsleep: syscon@1c00 {
 				reg = <0x1c00 0x100>;
 			};
 
-			reboot {
-				compatible = "syscon-reboot";
-				regmap = <&pmsleep>;
-				offset = <0xb8>;
-				mask = <0x79>;
-			};
-
 			rtc@2400 {
 				compatible = "mstar,msc313-rtc";
 				reg = <0x2400 0x40>;
-- 
2.51.0

[PATCH v2 8/8] arm: dts: mstar: v7: Add sizes and ranges to sram

[Daniel Palmer]

Add some missing properties to the SRAM node to make the dt
validation happy.

Signed-off-by: Daniel Palmer <daniel@0x0f.com>
---
 arch/arm/boot/dts/sigmastar/mstar-v7.dtsi | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/arm/boot/dts/sigmastar/mstar-v7.dtsi b/arch/arm/boot/dts/sigmastar/mstar-v7.dtsi
index db661e189eec..e50e4000d1ee 100644
--- a/arch/arm/boot/dts/sigmastar/mstar-v7.dtsi
+++ b/arch/arm/boot/dts/sigmastar/mstar-v7.dtsi
@@ -187,6 +187,9 @@ pm_uart: serial@221000 {
 		imi: sram@a0000000 {
 			compatible = "mmio-sram";
 			reg = <0xa0000000 0x10000>;
+			#address-cells = <1>;
+			#size-cells = <1>;
+			ranges = <0 0xa0000000 0x10000>;
 		};
 	};
 };
-- 
2.51.0

Re: [PATCH v2 1/8] dt-bindings: clk: mstar msc313 cpupll: Correct clock-cells value

[Rob Herring (Arm)]

On Thu, 01 Jan 2026 12:42:59 +0900, Daniel Palmer wrote:
> clock-cells should have been 0 and causes warnings when
> validating devicetrees.
> 
> Fix the value and fix the binding example.
> 
> Signed-off-by: Daniel Palmer <daniel@0x0f.com>
> ---
>  .../devicetree/bindings/clock/mstar,msc313-cpupll.yaml       | 5 ++---
>  1 file changed, 2 insertions(+), 3 deletions(-)
> 

My bot found errors running 'make dt_binding_check' on your patch:

yamllint warnings/errors:

dtschema/dtc warnings/errors:
Lexical error: Documentation/devicetree/bindings/clock/mstar,msc313-cpupll.example.dts:22.29-51 Unexpected 'MSTAR_MSC313_MPLL_DIV2'
FATAL ERROR: Syntax error parsing input tree
make[2]: *** [scripts/Makefile.dtbs:145: Documentation/devicetree/bindings/clock/mstar,msc313-cpupll.example.dtb] Error 1
make[2]: *** Waiting for unfinished jobs....
make[1]: *** [/builds/robherring/dt-review-ci/linux/Makefile:1565: dt_binding_check] Error 2
make: *** [Makefile:248: __sub-make] Error 2

doc reference errors (make refcheckdocs):

See https://patchwork.kernel.org/project/devicetree/patch/20260101034306.1233091-2-daniel@0x0f.com

The base for the series is generally the latest rc1. A different dependency
should be noted in *this* patch.

If you already ran 'make dt_binding_check' and didn't see the above
error(s), then make sure 'yamllint' is installed and dt-schema is up to
date:

pip3 install dtschema --upgrade

Please check and re-submit after running the above command yourself. Note
that DT_SCHEMA_FILES can be set to your schema file to speed up checking
your schema. However, it must be unset to test all examples with your schema.

Re: [PATCH v2 1/8] dt-bindings: clk: mstar msc313 cpupll: Correct clock-cells value

[Krzysztof Kozlowski]

On Thu, Jan 01, 2026 at 12:42:59PM +0900, Daniel Palmer wrote:
> clock-cells should have been 0 and causes warnings when

Why it should have been 0? That's what your commit msg should explain.


> validating devicetrees.
> 
> Fix the value and fix the binding example.
> 
> Signed-off-by: Daniel Palmer <daniel@0x0f.com>
> ---
>  .../devicetree/bindings/clock/mstar,msc313-cpupll.yaml       | 5 ++---
>  1 file changed, 2 insertions(+), 3 deletions(-)
> 
> diff --git a/Documentation/devicetree/bindings/clock/mstar,msc313-cpupll.yaml b/Documentation/devicetree/bindings/clock/mstar,msc313-cpupll.yaml
> index a9ad7ab5230c..889419fba269 100644
> --- a/Documentation/devicetree/bindings/clock/mstar,msc313-cpupll.yaml
> +++ b/Documentation/devicetree/bindings/clock/mstar,msc313-cpupll.yaml
> @@ -18,7 +18,7 @@ properties:
>      const: mstar,msc313-cpupll
>  
>    "#clock-cells":
> -    const: 1
> +    const: 0

That's technically ABI break and your commit msg must explain also the
impact of that break.

Best regards,
Krzysztof