diff mbox series

[v1,6/6] arm64: dts: rockchip: enable sfc controller on Quartz64 Model A

Message ID 20220511150117.113070-7-pgwipeout@gmail.com (mailing list archive)
State New, archived
Headers show
Series Cleanups and enablement for Quartz64-A | expand

Commit Message

Peter Geis May 11, 2022, 3:01 p.m. UTC
Add the sfc controller binding for the Quartz64 Model A. This is not
populated by default, so leave it disabled.

Signed-off-by: Peter Geis <pgwipeout@gmail.com>
---
 .../boot/dts/rockchip/rk3566-quartz64-a.dts      | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

Comments

Chris Morgan May 16, 2022, 3:26 p.m. UTC | #1
On Wed, May 11, 2022 at 11:01:17AM -0400, Peter Geis wrote:
> Add the sfc controller binding for the Quartz64 Model A. This is not
> populated by default, so leave it disabled.
> 
> Signed-off-by: Peter Geis <pgwipeout@gmail.com>
> ---
>  .../boot/dts/rockchip/rk3566-quartz64-a.dts      | 16 ++++++++++++++++
>  1 file changed, 16 insertions(+)
> 
> diff --git a/arch/arm64/boot/dts/rockchip/rk3566-quartz64-a.dts b/arch/arm64/boot/dts/rockchip/rk3566-quartz64-a.dts
> index 71df64655de5..6ec349e7e521 100644
> --- a/arch/arm64/boot/dts/rockchip/rk3566-quartz64-a.dts
> +++ b/arch/arm64/boot/dts/rockchip/rk3566-quartz64-a.dts
> @@ -603,6 +603,22 @@ &sdmmc1 {
>  	status = "okay";
>  };
>  
> +&sfc {
> +	pinctrl-0 = <&fspi_pins>;
> +	pinctrl-names = "default";
> +	#address-cells = <1>;
> +	#size-cells = <0>;
> +	status = "disabled";
> +
> +	flash@0 {
> +		compatible = "jedec,spi-nor";
> +		reg = <0>;
> +		spi-max-frequency = <24000000>;
> +		spi-rx-bus-width = <4>;
> +		spi-tx-bus-width = <1>;

This isn't really a concern, just a comment. Did you test this with the
spi-tx-bus-width of 4 by chance? While I did have to use 1 for my
implementation (the Odroid Go Advance) the Rockchip engineer I worked
with couldn't replicate the issue on his end and we ended up chalking
my issues up to an implementation specific problem. I'm only commenting
here because I don't want you to think that for this device the tx
always has to be 1, of course if your implementation does have issues
with a tx of 2 or 4 that's different...

Thank you.

> +	};
> +};
> +
>  /* spdif is exposed on con40 pin 18 */
>  &spdif {
>  	status = "okay";
> -- 
> 2.25.1
>
Peter Geis May 16, 2022, 9:51 p.m. UTC | #2
On Mon, May 16, 2022 at 11:26 AM Chris Morgan <macroalpha82@gmail.com> wrote:
>
> On Wed, May 11, 2022 at 11:01:17AM -0400, Peter Geis wrote:
> > Add the sfc controller binding for the Quartz64 Model A. This is not
> > populated by default, so leave it disabled.
> >
> > Signed-off-by: Peter Geis <pgwipeout@gmail.com>
> > ---
> >  .../boot/dts/rockchip/rk3566-quartz64-a.dts      | 16 ++++++++++++++++
> >  1 file changed, 16 insertions(+)
> >
> > diff --git a/arch/arm64/boot/dts/rockchip/rk3566-quartz64-a.dts b/arch/arm64/boot/dts/rockchip/rk3566-quartz64-a.dts
> > index 71df64655de5..6ec349e7e521 100644
> > --- a/arch/arm64/boot/dts/rockchip/rk3566-quartz64-a.dts
> > +++ b/arch/arm64/boot/dts/rockchip/rk3566-quartz64-a.dts
> > @@ -603,6 +603,22 @@ &sdmmc1 {
> >       status = "okay";
> >  };
> >
> > +&sfc {
> > +     pinctrl-0 = <&fspi_pins>;
> > +     pinctrl-names = "default";
> > +     #address-cells = <1>;
> > +     #size-cells = <0>;
> > +     status = "disabled";
> > +
> > +     flash@0 {
> > +             compatible = "jedec,spi-nor";
> > +             reg = <0>;
> > +             spi-max-frequency = <24000000>;
> > +             spi-rx-bus-width = <4>;
> > +             spi-tx-bus-width = <1>;
>
> This isn't really a concern, just a comment. Did you test this with the
> spi-tx-bus-width of 4 by chance? While I did have to use 1 for my
> implementation (the Odroid Go Advance) the Rockchip engineer I worked
> with couldn't replicate the issue on his end and we ended up chalking
> my issues up to an implementation specific problem. I'm only commenting
> here because I don't want you to think that for this device the tx
> always has to be 1, of course if your implementation does have issues
> with a tx of 2 or 4 that's different...

Yes, and it was a data disaster. Looking into it I found all of the
flash chips that I could find only support 4x RX 1x TX. Some didn't
handle 4x RX terribly well either.

>
> Thank you.
>
> > +     };
> > +};
> > +
> >  /* spdif is exposed on con40 pin 18 */
> >  &spdif {
> >       status = "okay";
> > --
> > 2.25.1
> >
diff mbox series

Patch

diff --git a/arch/arm64/boot/dts/rockchip/rk3566-quartz64-a.dts b/arch/arm64/boot/dts/rockchip/rk3566-quartz64-a.dts
index 71df64655de5..6ec349e7e521 100644
--- a/arch/arm64/boot/dts/rockchip/rk3566-quartz64-a.dts
+++ b/arch/arm64/boot/dts/rockchip/rk3566-quartz64-a.dts
@@ -603,6 +603,22 @@  &sdmmc1 {
 	status = "okay";
 };
 
+&sfc {
+	pinctrl-0 = <&fspi_pins>;
+	pinctrl-names = "default";
+	#address-cells = <1>;
+	#size-cells = <0>;
+	status = "disabled";
+
+	flash@0 {
+		compatible = "jedec,spi-nor";
+		reg = <0>;
+		spi-max-frequency = <24000000>;
+		spi-rx-bus-width = <4>;
+		spi-tx-bus-width = <1>;
+	};
+};
+
 /* spdif is exposed on con40 pin 18 */
 &spdif {
 	status = "okay";