Skip to contentSkip to conversation

arm64: dts: nuvoton: ma35d1: add CAN nodes

PATCH RESEND Superseded Zi-Yu Chen 5 participants 7 messages Jul 8 – Sep 8

Subsystems (3)

v2 of 4 Superseded Latest is v4 →

lore
Share & tools
lore

Zi-Yu Chen

author
#0 lore raw

PATCH RESEND v2 1/1 arm64: dts: nuvoton: ma35d1: add CAN nodes

Add controller nodes for the four Bosch M_CAN blocks found on the Nuvoton MA35D1 SoC.

Additionally, configure pinctrl and enable CAN1 and CAN3 on the MA35D1 SOM board. Also, update the APLL frequency to 200MHz to ensure the CAN controllers receive the required input clock for 50MHz operation.

---
Resend note:
- resend with the complete Cc list; no patch changes

 .../boot/dts/nuvoton/ma35d1-som-256m.dts      | 32 +++++++++++-
 arch/arm64/boot/dts/nuvoton/ma35d1.dtsi       | 52 +++++++++++++++++++
 2 files changed, 83 insertions(+), 1 deletion(-)
arch/arm64/boot/dts/nuvoton/ma35d1-som-256m.dts +31 -1
index f6f20a17e501..fb23b0573bdc 100644
@@ -37,6 +37,22 @@ clk_hxt: clock-hxt {
 	};
 };
 
+&can1 {
+	pinctrl-names = "default";
+	pinctrl-0 = <&pinctrl_can1>;
+	assigned-clocks = <&clk CAN1_DIV>;
+	assigned-clock-rates = <50000000>;
+	status = "okay";
+};
+
+&can3 {
+	pinctrl-names = "default";
+	pinctrl-0 = <&pinctrl_can3>;
+	assigned-clocks = <&clk CAN3_DIV>;
+	assigned-clock-rates = <50000000>;
+	status = "okay";
+};
+
 &clk {
 	assigned-clocks = <&clk CAPLL>,
 			  <&clk DDRPLL>,
@@ -45,7 +61,7 @@ &clk {
 			  <&clk VPLL>;
 	assigned-clock-rates = <800000000>,
 			       <266000000>,
-			       <180000000>,
+			       <200000000>,
 			       <500000000>,
 			       <102000000>;
 	nuvoton,pll-mode = "integer",
@@ -56,6 +72,20 @@ &clk {
 };
 
 &pinctrl {
+	can-grp {
+		pinctrl_can1: can1-pins {
+			nuvoton,pins = <11 14 4>,
+				       <11 15 4>;
+			bias-disable;
+		};
+
+		pinctrl_can3: can3-pins {
+			nuvoton,pins = <11 10 3>,
+				       <11 11 3>;
+			bias-disable;
+		};
+	};
+
 	uart-grp {
 		pinctrl_uart0: uart0-pins {
 			nuvoton,pins = <4 14 1>,
arch/arm64/boot/dts/nuvoton/ma35d1.dtsi +52 -0

58 lines not on this page · raw message

Signature · 1 line
2.34.1
#1 lore raw
Quoted 11 lines · Zi-Yu Chen

Add controller nodes for the four Bosch M_CAN blocks found on the Nuvoton MA35D1 SoC.

Additionally, configure pinctrl and enable CAN1 and CAN3 on the MA35D1 SOM board. Also, update the APLL frequency to 200MHz to ensure the CAN controllers receive the required input clock for 50MHz operation.

---
Resend note:
- resend with the complete Cc list; no patch changes

Hi,

I don't see anything wrong with the patch contents, but I think you still have the wrong Cc list. You have sent the patch Cc: [email protected], which is where maintainers should send the patch when it is ready for inclusion in the SoC tree, and you have sent it To Olof and me. Olof retired as SoC tree maintainer a while ago, and I'm not going to apply this because it should go through the Nuvoton tree maintained by Jacky Huang and Shan-Chun Hung.

Please send future patches to the maintainers that are in charge of applying the patches, without adding [email protected] yourself. The maintainers will then forward it when they are happy with the changes.

I just noticed that we have two conflicting maintainers entries for ma35:

- Jacky and Shan-Chun are listed for the ma35 SoC, but have not sent any updates after the initial merge in 2024 and don't currently have a git tree in linux-next

- Andrew Jeffery and his team are listed for npcm with arch/arm64/boot/dts/nuvoton/ listed as one of the file locations, though that contains both ma35 and npcm. Andrew has the bmc tree in linux-next and has recently picked up patches for both npcm and ma35.

I see there is a lot of activity at the moment working on ma35 drivers, so it would be good if the maintainers could figure this out together and provide better guidance to individual developers. I don't mind pulling from one team or the other, but I would definitely prefer the MAINTAINERS file to reflect the actual state here.

If Jacky and Shan-Chun are planning to do the merges going forward, they should add their git tree to linux-next and update the MAINTAINERS file to list the git URL.

Arnd

Reviewed-by
#2 lore raw

Hi Arnd,

Thanks for the clarification.

Shan-Chun and I are still maintaining the MA35 platform.

I have reviewed this patch and it looks good to me.

We agree that the current maintenance path for MA35 could be clearer. We will work on making the MA35 maintenance flow more transparent.

Thanks for pointing this out.

Thanks,
Jacky

Quoted 52 lines · Zi-Yu Chen
Quoted 11 lines

Add controller nodes for the four Bosch M_CAN blocks found on the Nuvoton MA35D1 SoC.

Additionally, configure pinctrl and enable CAN1 and CAN3 on the MA35D1 SOM board. Also, update the APLL frequency to 200MHz to ensure the CAN controllers receive the required input clock for 50MHz operation.

---
Resend note:
- resend with the complete Cc list; no patch changes

Hi,

I don't see anything wrong with the patch contents, but I think you still have the wrong Cc list. You have sent the patch Cc: [email protected], which is where maintainers should send the patch when it is ready for inclusion in the SoC tree, and you have sent it To Olof and me. Olof retired as SoC tree maintainer a while ago, and I'm not going to apply this because it should go through the Nuvoton tree maintained by Jacky Huang and Shan-Chun Hung.

Please send future patches to the maintainers that are in charge of applying the patches, without adding [email protected] yourself. The maintainers will then forward it when they are happy with the changes.

I just noticed that we have two conflicting maintainers entries for ma35:

- Jacky and Shan-Chun are listed for the ma35 SoC, but have not sent any updates after the initial merge in 2024 and don't currently have a git tree in linux-next

- Andrew Jeffery and his team are listed for npcm with arch/arm64/boot/dts/nuvoton/ listed as one of the file locations, though that contains both ma35 and npcm. Andrew has the bmc tree in linux-next and has recently picked up patches for both npcm and ma35.

I see there is a lot of activity at the moment working on ma35 drivers, so it would be good if the maintainers could figure this out together and provide better guidance to individual developers. I don't mind pulling from one team or the other, but I would definitely prefer the MAINTAINERS file to reflect the actual state here.

If Jacky and Shan-Chun are planning to do the merges going forward, they should add their git tree to linux-next and update the MAINTAINERS file to list the git URL.

Arnd

maintainer
#3 lore raw

Hi all,

Quoted 4 lines · Jacky Huang

Hi Arnd,

Thanks for the clarification.

Quoted 8 lines

Shan-Chun and I are still maintaining the MA35 platform.

I have reviewed this patch and it looks good to me.

We agree that the current maintenance path for MA35 could be clearer. We will work on making the MA35 maintenance flow more transparent.

I discussed this with Jacky off-list. Going forward I'll take MA35 patches with R-b tags from Jacky and Zi-Yu through the BMC tree to keep things simple.

I'll pick up v2 for the next merge window if there are no other issues with it.

Andrew

#4 lore raw
Quoted 14 lines · Andrew Jeffery
Quoted 4 lines · Jacky Huang

Hi Arnd,

Thanks for the clarification.

Quoted 8 lines

Shan-Chun and I are still maintaining the MA35 platform.

I have reviewed this patch and it looks good to me.

We agree that the current maintenance path for MA35 could be clearer. We will work on making the MA35 maintenance flow more transparent.

Quoted 6 lines

I discussed this with Jacky off-list. Going forward I'll take MA35 patches with R-b tags from Jacky and Zi-Yu through the BMC tree to keep things simple.

I'll pick up v2 for the next merge window if there are no other issues with it.

Sounds good, thanks!

Arnd

#5 lore raw
In reply to Zi-Yu Chen
Quoted 6 lines · Zi-Yu Chen

Add controller nodes for the four Bosch M_CAN blocks found on the Nuvoton MA35D1 SoC.

Additionally, configure pinctrl and enable CAN1 and CAN3 on the MA35D1 SOM board. Also, update the APLL frequency to 200MHz to ensure the CAN controllers receive the required input clock for 50MHz operation.

50 MHz is an unusual CAN clock frequency. Better use a multiple of 40 MHz. Within the limits of the m_can IP core, the higher the multiple the better.

regards,
Marc

Signature · 4 lines
Pengutronix e.K.                 | Marc Kleine-Budde          |
Embedded Linux                   | https://www.pengutronix.de |
Vertretung Nürnberg              | Phone: +49-5121-206917-129 |
Amtsgericht Hildesheim, HRA 2686 | Fax:   +49-5121-206917-9   |
author
#6 lore raw

Thanks for the review.

I tested the change and confirmed that an accurate 80 MHz CAN clock can be generated. I've uploaded v3 with this change included.

regards,
Zi-Yu

Marc Kleine-Budde 於 2026/8/24 下午 06:39 寫道:

Quoted 15 lines · Zi-Yu Chen
Quoted 6 lines

Add controller nodes for the four Bosch M_CAN blocks found on the Nuvoton MA35D1 SoC.

Additionally, configure pinctrl and enable CAN1 and CAN3 on the MA35D1 SOM board. Also, update the APLL frequency to 200MHz to ensure the CAN controllers receive the required input clock for 50MHz operation.

50 MHz is an unusual CAN clock frequency. Better use a multiple of 40 MHz. Within the limits of the m_can IP core, the higher the multiple the better.

regards,
Marc