[PATCH v4 3/3] board: toradex: add verdin am62 support

[Marcel Ziswiler]

Hi Nishanth

On Wed, 2023-08-02 at 17:29 -0500, Nishanth Menon wrote:
> On 22:54-20230728, Marcel Ziswiler wrote:
> > From: Marcel Ziswiler <marcel.ziswiler at toradex.com>
> > 
> > This adds initial support for the Toradex Verdin AM62 Quad 1GB WB IT
> > V1.0A module and subsequent V1.1 launch configuration SKUs. They are
> > strapped to boot from their on-module eMMC. U-Boot supports booting
> > from the on-module eMMC only, DFU support is disabled for now due to
> > missing AM62x USB support.
> > 
> > Boot sequence is:
> > SYSFW ---> R5 SPL (both in tiboot3.bin) ---> ATF (TF-A) ---> OP-TEE
> >   ---> A53 SPL (part of tispl.bin) ---> U-boot proper (u-boot.img)
> 
> Please state the dts is from v6.5-rc1 - this is to ensure we can track
> over time the updates needed - typically we want to get to a place where
> we can update in one shot to stay in sync.

I did, but only in the change log. Let me add it to the commit message as well.

> just a side note: we might want to see how to get those eeproms merged
> in for kernel (we'd really like to get rid of u-boot.dtsi)

Yes, one may dream...

> Also drop  #include "k3-am625-verdin-wifi-dev-binman.dtsi" from r5.dts
> (you already get it from u-boot.dtsi)

Okay.


> you also dont need the memory@, reserved-memory sa_secproxy, esm etc in
> r5.dts - please see am62-r5-sk.dts to see how it is done. all of these
> are now inherited from SoC.dtsi or board.dts.. (yaay - your r5.dts is
> going to be much tinier once you clean that up)..

I guess I missed that recent change. Thanks!

> > Signed-off-by: Marcel Ziswiler <marcel.ziswiler at toradex.com>
> > 
> > ---
> > 
> 
> [...]
> 
> Marcel, copying the files from v6.5-rc1, I see the following deltas
> 
> we try to migrate the dts files as is.. so trying to understand the
> #if 0 here. Just disabling the driver in defconfig should get the
> feature out, correct?
> 
> I suspect some debuggy stuff creeped in perhaps?

No, USB support is not there yet and this used to fail. I actually still do have a series on top adding USB
support and dropped those there, but yeah, now with the synced stuff I may drop those here already. Thanks!

> diff --git a/arch/arm/dts/k3-am62-verdin-dev.dtsi b/arch/arm/dts/k3-am62-verdin-dev.dtsi
> index e5443cd3c2d3..846caee7dfa4 100644
> --- a/arch/arm/dts/k3-am62-verdin-dev.dtsi
> +++ b/arch/arm/dts/k3-am62-verdin-dev.dtsi
> @@ -160,7 +160,6 @@
>         status = "okay";
>  };
>  
> -#if 0
>  /* Verdin USB_1 */
>  &usbss0 {
>         status = "okay";
> @@ -178,7 +177,6 @@
>  &usb1 {
>         status = "okay";
>  };
> -#endif
>  
>  /* Verdin CTRL_WAKE1_MICO# */
>  &verdin_gpio_keys {
> diff --git a/arch/arm/dts/k3-am62-verdin.dtsi b/arch/arm/dts/k3-am62-verdin.dtsi
> index 9f626c869725..57dd061911ab 100644
> --- a/arch/arm/dts/k3-am62-verdin.dtsi
> +++ b/arch/arm/dts/k3-am62-verdin.dtsi
> @@ -36,10 +36,8 @@
>                 serial2 = &main_uart0;
>                 serial3 = &mcu_uart0;
>                 serial4 = &main_uart5;
> -#if 0
>                 usb0 = &usb0;
>                 usb1 = &usb1;
> -#endif
>         };
>  
>         verdin_gpio_keys: gpio-keys {
> @@ -1369,7 +1367,6 @@
>         status = "disabled";
>  };
>  
> -#if 0
>  /* Verdin USB_1 */
>  &usbss0 {
>         ti,vbus-divider;
> @@ -1395,7 +1392,6 @@
>         dr_mode = "host";
>         status = "disabled";
>  };
> -#endif
>  
>  /* Verdin UART_2 */
>  &wkup_uart0 {

Cheers

Marcel