2023-05-19 11:56:28

by Nitin Yadav

Subject: [PATCH v2 2/2] arm64: dts: ti: k3-am62-main: add sa3_secproxy in cbass_main

Add sa3_secproxy node in k3-am62-main.dtsi to keep device tree
nodes in sync with u-boot nodes.

Signed-off-by: Nitin Yadav <[email protected]>
---
arch/arm64/boot/dts/ti/k3-am62-main.dtsi | 9 +++++++++
1 file changed, 9 insertions(+)

diff --git a/arch/arm64/boot/dts/ti/k3-am62-main.dtsi b/arch/arm64/boot/dts/ti/k3-am62-main.dtsi
index b3e4857bbbe4..7c2af5b0e022 100644
--- a/arch/arm64/boot/dts/ti/k3-am62-main.dtsi
+++ b/arch/arm64/boot/dts/ti/k3-am62-main.dtsi
@@ -42,6 +42,15 @@ gic_its: msi-controller@1820000 {
};
};

+ sa3_secproxy: secproxy@44880000 {
+ compatible = "ti,am654-secure-proxy";
+ #mbox-cells = <1>;
+ reg-names = "rt", "scfg", "target_data";
+ reg = <0x00 0x44880000 0x00 0x20000>,
+ <0x0 0x44860000 0x0 0x20000>,
+ <0x0 0x43600000 0x0 0x10000>;
+ };
+
main_conf: syscon@100000 {
compatible = "syscon", "simple-mfd";
reg = <0x00 0x00100000 0x00 0x20000>;
--
2.25.1

2023-05-19 18:26:13

by Andrew Davis

Subject: Re: [PATCH v2 2/2] arm64: dts: ti: k3-am62-main: add sa3_secproxy in cbass_main

On 5/19/23 6:34 AM, Nitin Yadav wrote:
> Add sa3_secproxy node in k3-am62-main.dtsi to keep device tree
> nodes in sync with u-boot nodes.

That is not a good reason, nodes should be added because the device
exists and should be described. Simply say that here.

>
> Signed-off-by: Nitin Yadav <[email protected]>
> ---
> arch/arm64/boot/dts/ti/k3-am62-main.dtsi | 9 +++++++++
> 1 file changed, 9 insertions(+)
>
> diff --git a/arch/arm64/boot/dts/ti/k3-am62-main.dtsi b/arch/arm64/boot/dts/ti/k3-am62-main.dtsi
> index b3e4857bbbe4..7c2af5b0e022 100644
> --- a/arch/arm64/boot/dts/ti/k3-am62-main.dtsi
> +++ b/arch/arm64/boot/dts/ti/k3-am62-main.dtsi
> @@ -42,6 +42,15 @@ gic_its: msi-controller@1820000 {
> };
> };
>
> + sa3_secproxy: secproxy@44880000 {
> + compatible = "ti,am654-secure-proxy";

The "ti,am654-secure-proxy" binding has interrupts as a required
property, so this will cause new binding check warnings.

We also already have a Secure Proxy instance in this DT, why
do we need this other one? Is this the instance that was added
for for the R5 use? I guess that would explain why there are no
interrupts to the big ARM core.. Can we actually use this
node in Linux then? If not mark it disabled/reserved.

Andrew

> + #mbox-cells = <1>;
> + reg-names = "rt", "scfg", "target_data";
> + reg = <0x00 0x44880000 0x00 0x20000>,
> + <0x0 0x44860000 0x0 0x20000>,
> + <0x0 0x43600000 0x0 0x10000>;
> + };
> +
> main_conf: syscon@100000 {
> compatible = "syscon", "simple-mfd";
> reg = <0x00 0x00100000 0x00 0x20000>;