Re: [PATCH 1/4] arm: dts: add support for Laird WB45N cpu module and DVK

From: Nicolas Ferre
Date: Fri Jun 15 2018 - 08:07:17 EST

On 15/06/2018 at 12:01, Ben Whitten wrote:
On 14/06/2018 at 10:51, Ben Whitten wrote:
Signed-off-by: Ben Whitten <ben.whitten@xxxxxxxxxxxxx>
---
arch/arm/boot/dts/Makefile | 3 +-
arch/arm/boot/dts/at91-wb45n.dts | 66 +++++++++++++++
arch/arm/boot/dts/at91-wb45n.dtsi | 169
++++++++++++++++++++++++++++++++++++++
3 files changed, 237 insertions(+), 1 deletion(-)
create mode 100644 arch/arm/boot/dts/at91-wb45n.dts
create mode 100644 arch/arm/boot/dts/at91-wb45n.dtsi

diff --git a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile
index 7e24249..1ee94ee 100644
--- a/arch/arm/boot/dts/Makefile
+++ b/arch/arm/boot/dts/Makefile
@@ -42,7 +42,8 @@ dtb-$(CONFIG_SOC_AT91SAM9) += \
at91sam9g25ek.dtb \
at91sam9g35ek.dtb \
at91sam9x25ek.dtb \
- at91sam9x35ek.dtb
+ at91sam9x35ek.dtb \
+ at91-wb45n.dtb
dtb-$(CONFIG_SOC_SAM_V7) += \
at91-kizbox2.dtb \
at91-nattis-2-natte-2.dtb \
diff --git a/arch/arm/boot/dts/at91-wb45n.dts b/arch/arm/boot/dts/at91-
wb45n.dts
new file mode 100644
index 0000000..4e88815
--- /dev/null
+++ b/arch/arm/boot/dts/at91-wb45n.dts
@@ -0,0 +1,66 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * at91-wb45n.dts - Device Tree file for WB45NBT board
+ *
+ * Copyright (C) 2018 Laird
+ *
+*/
+/dts-v1/;
+#include "at91-wb45n.dtsi"
+
+/ {
+ model = "Laird Workgroup Bridge 45N - Atmel AT91SAM (dt)";
+ compatible = "laird,wb45n", "laird,wbxx", "atmel,at91sam9x5",
"atmel,at91sam9";

"laird" prefix must be added to
Documentation/devicetree/bindings/vendor-prefixes.txt before using it:
you can do a little patch as a first patch of this series.
Otherwise it will trigger a warning message while running
scripts/checkpatch.pl on top of your patch.


+
+ ahb {
+ apb {
+ watchdog@fffffe40 {
+ status = "okay";
+ };
+ };
+ };
+
+ gpio_keys {
+ compatible = "gpio-keys";
+ #address-cells = <1>;
+ #size-cells = <0>;
+ irqbtn@pb18 {

I'm not sure that the @pb18 can be used like this. This address
extension must be used in a "reg" property in the node. dtc used with
warning switch on might trigger an error for this.

+ label = "IRQBTN";
+ linux,code = <99>;
+ gpios = <&pioB 18 GPIO_ACTIVE_LOW>;
+ gpio-key,wakeup = <1>;
+ };
+ };
+};
+
+&usb0 {
+ status = "okay";
+};
+
+&mmc0 {
+ status = "okay";
+};
+
+&spi0 {
+ status = "okay";
+};
+
+&macb0 {
+ status = "okay";
+};
+
+&dbgu {
+ status = "okay";
+};
+
+&usart0 {
+ status = "okay";
+};
+
+&usart3 {
+ status = "okay";
+};
+
+&i2c1 {
+ status = "okay";
+};
diff --git a/arch/arm/boot/dts/at91-wb45n.dtsi b/arch/arm/boot/dts/at91-
wb45n.dtsi
new file mode 100644
index 0000000..2fa58e2
--- /dev/null
+++ b/arch/arm/boot/dts/at91-wb45n.dtsi
@@ -0,0 +1,169 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * at91-wb45n.dtsi - Device Tree file for WB45NBT board
+ *
+ * Copyright (C) 2018 Laird
+ *
+ */
+
+#include "at91sam9g25.dtsi"
+
+/ {
+ model = "Laird Workgroup Bridge 45N - Atmel AT91SAM (dt)";
+ compatible = "laird,wb45n", "laird,wbxx", "atmel,at91sam9x5",
"atmel,at91sam9";
+
+ chosen {
+ bootargs = "ubi.mtd=6 root=ubi0:rootfs rootfstype=ubifs
rw";
+ stdout-path = "serial0:115200n8";
+ };
+
+ memory {
+ reg = <0x20000000 0x4000000>;
+ };
+
+ ahb {
+ apb {
+ shdwc@fffffe10 {

I would advice you to take exactly the node name:
"shutdown-controller@fffffe10"; Anyway, it will go away after you use
the label notation as advised by Alexandre.

+ atmel,wakeup-mode = "low";
+ };
+
+ pinctrl@fffff400 {
+ usb2 {
+ pinctrl_board_usb2: usb2-board {
+ atmel,pins =
+ <AT91_PIOB 11
AT91_PERIPH_GPIO AT91_PINCTRL_DEGLITCH>; /* PB11 gpio
vbus sense, deglitch */
+ };
+ };
+ };
+
+ rstc@fffffe00 {
+ compatible = "atmel,sama5d3-rstc";
+ };

I don't think this node is needed.

I dug through our old code reviews and found this message relating to testing
reboot over several thousand times in our testbed:
After the slow clock has been enabled on the reset controller via upstream
changes, the dram disable access and power down code is causing the SAM9G25
to hang occasionally on reboot. Using the simple reset function provided
for SAMA5D3 instead.

So it appears to be a workaround for a bug that existed ~2 years ago, may still be
relevant as there haven't been many changes to the reset code in that time.

All right, I read too quickly and thought it was sama5d3... Your feedback is interesting anyway. I'll store this for future reference and investigation.

+
+ };
+ };
+
+ atheros {
+ compatible = "atheros,ath6kl";
+ atheros,board-id = "SD32";
+ };
+};
+
+&slow_xtal {
+ clock-frequency = <32768>;
+};
+
+&main_xtal {
+ clock-frequency = <12000000>;
+};
+
+&ebi {
+ status = "okay";
+ nand_controller: nand-controller {
+ pinctrl-0 = <&pinctrl_nand_cs &pinctrl_nand_rb
&pinctrl_nand_oe_we>;
+ pinctrl-names = "default";
+ status = "okay";
+
+ nand@3 {
+ reg = <0x3 0x0 0x800000>;
+ rb-gpios = <&pioD 5 GPIO_ACTIVE_HIGH>;
+ cs-gpios = <&pioD 4 GPIO_ACTIVE_HIGH>;
+ nand-bus-width = <8>;
+ nand-ecc-mode = "hw";
+ nand-ecc-strength = <4>;
+ nand-ecc-step-size = <512>;
+ nand-on-flash-bbt;
+ label = "atmel_nand";
+
+ partitions {
+ compatible = "fixed-partitions";
+ #address-cells = <1>;
+ #size-cells = <1>;
+
+ at91bootstrap@0 {
+ label = "at91bs";
+ reg = <0x0 0x20000>;
+ };
+
+ uboot@20000 {
+ label = "u-boot";
+ reg = <0x20000 0x80000>;
+ };
+
+ ubootenv@a0000 {
+ label = "u-boot-env";
+ reg = <0xa0000 0x20000>;
+ };
+
+ ubootenv@c0000 {
+ label = "redund-env";
+ reg = <0xc0000 0x20000>;
+ };
+
+ kernel-a@e0000 {
+ label = "kernel-a";
+ reg = <0xe0000 0x280000>;
+ };
+
+ kernel-b@360000 {
+ label = "kernel-b";
+ reg = <0x360000 0x280000>;
+ };
+
+ rootfs-a@5e0000 {
+ label = "rootfs-a";
+ reg = <0x5e0000 0x2600000>;
+ };
+
+ rootfs-b@2be0000 {
+ label = "rootfs-b";
+ reg = <0x2be0000 0x2600000>;
+ };
+
+ user@51e0000 {
+ label = "user";
+ reg = <0x51e0000 0x2dc0000>;
+ };
+
+ logs@7fa0000 {
+ label = "logs";
+ reg = <0x7fa0000 0x60000>;
+ };
+
+ };
+ };
+ };
+};
+
+&usb0 {

This must be &usb1 label, isn't it?
Because you are referring to ohci binding I suspect (found by having a
look at: atmel,oc-gpio property...).

I believe usb0 is correct, as this is a at91sam9x5 part, the node in dtsi is -ohci.
sama5d3 is usb1 for -ohci.

All right, like previous comment, I thought it was sama5d3: sorry for the noise.

Best regards,
Nicolas
--
Nicolas Ferre