On Wed, Mar 29, 2023 at 8:32 PM Maxime Ripard <maxime@xxxxxxxxxx> wrote:
>
> The commit title is wrong, it's not a rockchip device.
Thanks for the note, I will fix it in the next version.
>
> On Wed, Mar 29, 2023 at 06:46:15PM +0530, Jagan Teki wrote:
> > The DSI downstream devices are likely to be Panel, Bridge and
> > I2C-Configured Bridge.
> >
> > It is possible to connect all three devices using upstream OF-graph port
> > or ports node however only Panel and Bridge are possible to connect via
> > child node but not possible to connect I2C-Configured Bridge via child
> > node since I2C-Configure bridges are child of I2C not upstream DSI hosts
> > and it must represent them via port or ports with endpoint linking.
> >
> > Allwinner A64 DSI node already has a port so add endpoint 0 for input
> > tcon so that the downstream DSI devices can use endpoint 1 to connect
> > Panel or Bridge or I2C-Configured Bridge.
> >
> > An example of the I2C-Configured downstream bridge representation is,
> >
> > i2c1 {
> > bridge@1b {
> > compatible = "ti,dlpc3433";
> >
> > ports {
> > port@0 {
> > reg = <0>;
> >
> > bridge_in_dsi: endpoint {
> > remote-endpoint = <&dsi_out_bridge>;
> > data-lanes = <1 2 3 4>;
> > };
> > };
> >
> > port@2 {
> > reg = <2>;
> >
> > bridge_out_dmd: endpoint {
> > remote-endpoint = <&dmd_out_bridge>;
> > };
> > };
> > };
> > };
> > };
> >
> > dsi {
> > compatible = "allwinner,sun50i-a64-mipi-dsi";
> >
> > port {
> > dsi_in_tcon0: endpoint@0 {
> > reg = <0>;
> > remote-endpoint = <tcon0_out_dsi>;
> > };
> >
> > dsi_out_bridge: endpoint@1 {
> > reg = <1>;
> > remote-endpoint = <&bridge_in_dsi>;
> > };
> > };
> > };
> >
> > Note that existing device bindings are untouched and still represent
> > the downstream devices via child nodes since the sun6i-mipi-dsi host
> > will migrate to a standardized single helper to lookup for a
> > downstream device via child or OF-graph port or port node.
> >
> > Signed-off-by: Jagan Teki <jagan@xxxxxxxxxxxxxxxxxxxx>
> > ---
> > Changes for v7:
> > - new patch
> >
> > arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi | 6 +++++-
> > 1 file changed, 5 insertions(+), 1 deletion(-)
> >
> > diff --git a/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi b/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi
> > index 77b5349f6087..3ed566dc2172 100644
> > --- a/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi
> > +++ b/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi
> > @@ -1189,7 +1189,11 @@ dsi: dsi@1ca0000 {
> > #size-cells = <0>;
> >
> > port {
> > - dsi_in_tcon0: endpoint {
> > + #address-cells = <1>;
> > + #size-cells = <0>;
> > +
> > + dsi_in_tcon0: endpoint@0 {
> > + reg = <0>;
> > remote-endpoint = <&tcon0_out_dsi>;
> > };
>
> That doesn't match the DT binding anymore, and why can't we add endpoint@1 there too?
Do you mean add endpoint@1 without any remote-endpoint like this?
dsi_out_bridge: endpoint@1 {
reg = <1>;
};
I was supposed to add this, since dtbs_check doesn't give any error. I
have skipped this, as I thought it wouldn't be needed.
Thanks,
agan.
