[PATCH 0/2] arm64: dts: imx8mm-evk: Fix hdmi@3d node

[PATCH 0/2] arm64: dts: imx8mm-evk: Fix hdmi@3d node

[Liu Ying]

Hi,

This series aims to fix hdmi@3d node in imx8mm-evk.dtsi.
This should make CHECK_DTBS happy if we check imx8mm-evk.dtb against
adi,adv7533.yaml.

Patch 1 is a preparation for patch 2.
Patch 2 fixes the hdmi@3d node.

Liu Ying (2):
  arm64: dts: imx8mm-evk: Add VDDEXT_3V3 fixed regulator
  arm64: dts: imx8mm-evk: Fix hdmi@3d node

 arch/arm64/boot/dts/freescale/imx8mm-evk.dtsi | 32 ++++++++++++-------
 1 file changed, 20 insertions(+), 12 deletions(-)

-- 
2.37.1

[PATCH 1/2] arm64: dts: imx8mm-evk: Add VDDEXT_3V3 fixed regulator

[Liu Ying]

The VDDEXT_3V3 fixed regulator supplies 3.3v power for MIPI CSI/DSI
through mini-SAS connector and M.2 PCIe according to board schematic,
so add the fixed regulator support.

Signed-off-by: Liu Ying <victor.liu@nxp.com>
---
 arch/arm64/boot/dts/freescale/imx8mm-evk.dtsi | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/arch/arm64/boot/dts/freescale/imx8mm-evk.dtsi b/arch/arm64/boot/dts/freescale/imx8mm-evk.dtsi
index df8e808ac473..7cd1cedfa9e9 100644
--- a/arch/arm64/boot/dts/freescale/imx8mm-evk.dtsi
+++ b/arch/arm64/boot/dts/freescale/imx8mm-evk.dtsi
@@ -72,6 +72,13 @@ reg_usdhc2_vmmc: regulator-usdhc2 {
 		enable-active-high;
 	};
 
+	reg_vddext_3v3: regulator-vddext-3v3 {
+		compatible = "regulator-fixed";
+		regulator-name = "VDDEXT_3V3";
+		regulator-min-microvolt = <3300000>;
+		regulator-max-microvolt = <3300000>;
+	};
+
 	backlight: backlight {
 		compatible = "pwm-backlight";
 		pwms = <&pwm1 0 5000000 0>;
-- 
2.37.1

[PATCH 2/2] arm64: dts: imx8mm-evk: Fix hdmi@3d node

[Liu Ying]

The hdmi@3d node's compatible string is "adi,adv7535" instead of
"adi,adv7533" or "adi,adv751*".

Fix the hdmi@3d node by means of:
* Use default register addresses for "cec", "edid" and "packet", because
  there is no need to use a non-default address map.
* Add missing interrupt related properties.
* Drop "adi,input-*" properties which are only valid for adv751*.
* Add "*-supply" properties, since most are required.
* Fix label names - s/adv7533/adv7535/.

Fixes: a27335b3f1e0 ("arm64: dts: imx8mm-evk: Add HDMI support")
Signed-off-by: Liu Ying <victor.liu@nxp.com>
---
 arch/arm64/boot/dts/freescale/imx8mm-evk.dtsi | 25 ++++++++++---------
 1 file changed, 13 insertions(+), 12 deletions(-)

diff --git a/arch/arm64/boot/dts/freescale/imx8mm-evk.dtsi b/arch/arm64/boot/dts/freescale/imx8mm-evk.dtsi
index 7cd1cedfa9e9..6752c3027436 100644
--- a/arch/arm64/boot/dts/freescale/imx8mm-evk.dtsi
+++ b/arch/arm64/boot/dts/freescale/imx8mm-evk.dtsi
@@ -26,7 +26,7 @@ hdmi-connector {
 
 		port {
 			hdmi_connector_in: endpoint {
-				remote-endpoint = <&adv7533_out>;
+				remote-endpoint = <&adv7535_out>;
 			};
 		};
 	};
@@ -324,15 +324,16 @@ &i2c2 {
 
 	hdmi@3d {
 		compatible = "adi,adv7535";
-		reg = <0x3d>, <0x3c>, <0x3e>, <0x3f>;
-		reg-names = "main", "cec", "edid", "packet";
+		reg = <0x3d>;
+		interrupt-parent = <&gpio1>;
+		interrupts = <9 IRQ_TYPE_EDGE_FALLING>;
 		adi,dsi-lanes = <4>;
-
-		adi,input-depth = <8>;
-		adi,input-colorspace = "rgb";
-		adi,input-clock = "1x";
-		adi,input-style = <1>;
-		adi,input-justification = "evenly";
+		avdd-supply = <&buck5_reg>;
+		dvdd-supply = <&buck5_reg>;
+		pvdd-supply = <&buck5_reg>;
+		a2vdd-supply = <&buck5_reg>;
+		v3p3-supply = <&reg_vddext_3v3>;
+		v1p2-supply = <&buck5_reg>;
 
 		ports {
 			#address-cells = <1>;
@@ -341,7 +342,7 @@ ports {
 			port@0 {
 				reg = <0>;
 
-				adv7533_in: endpoint {
+				adv7535_in: endpoint {
 					remote-endpoint = <&dsi_out>;
 				};
 			};
@@ -349,7 +350,7 @@ adv7533_in: endpoint {
 			port@1 {
 				reg = <1>;
 
-				adv7533_out: endpoint {
+				adv7535_out: endpoint {
 					remote-endpoint = <&hdmi_connector_in>;
 				};
 			};
@@ -415,7 +416,7 @@ port@1 {
 			reg = <1>;
 
 			dsi_out: endpoint {
-				remote-endpoint = <&adv7533_in>;
+				remote-endpoint = <&adv7535_in>;
 				data-lanes = <1 2 3 4>;
 			};
 		};
-- 
2.37.1

Re: [PATCH 1/2] arm64: dts: imx8mm-evk: Add VDDEXT_3V3 fixed regulator

[Fabio Estevam]

Hi Liu,

On Mon, Aug 14, 2023 at 5:07 AM Liu Ying <victor.liu@nxp.com> wrote:

> +       reg_vddext_3v3: regulator-vddext-3v3 {
> +               compatible = "regulator-fixed";
> +               regulator-name = "VDDEXT_3V3";
> +               regulator-min-microvolt = <3300000>;
> +               regulator-max-microvolt = <300000>;
> +       };

Booting the dtb with this change will cause this regulator to be
disabled as there is no consumer for it.

I suggest squashing both patches:

Tested-by: Fabio Estevam <festevam@gmail.com>

RE: [PATCH 1/2] arm64: dts: imx8mm-evk: Add VDDEXT_3V3 fixed regulator

[Ying Liu]

On Friday, August 18, 2023 10:06 PM Fabio Estevam <festevam@gmail.com> wrote:
> 
> Hi Liu,

Hi Fabio,

> 
> On Mon, Aug 14, 2023 at 5:07 AM Liu Ying <victor.liu@nxp.com> wrote:
> 
> > +       reg_vddext_3v3: regulator-vddext-3v3 {
> > +               compatible = "regulator-fixed";
> > +               regulator-name = "VDDEXT_3V3";
> > +               regulator-min-microvolt = <3300000>;
> > +               regulator-max-microvolt = <300000>;
> > +       };
> 
> Booting the dtb with this change will cause this regulator to be
> disabled as there is no consumer for it.

Software cannot control this regulator by any means.  It's on once
the board is powered on.  Do you mean it will be disabled in software
level if no consumer?  If that's the case, is the disablement harmful?

> 
> I suggest squashing both patches:

The patch looks self-contained to me.  Can we keep it as-is?

> 
> Tested-by: Fabio Estevam <festevam@gmail.com>

Thanks for your test!

Regards,
Liu Ying

Re: [PATCH 1/2] arm64: dts: imx8mm-evk: Add VDDEXT_3V3 fixed regulator

[Shawn Guo]

On Mon, Aug 21, 2023 at 02:33:17AM +0000, Ying Liu wrote:
> On Friday, August 18, 2023 10:06 PM Fabio Estevam <festevam@gmail.com> wrote:
> > On Mon, Aug 14, 2023 at 5:07 AM Liu Ying <victor.liu@nxp.com> wrote:
> > 
> > > +       reg_vddext_3v3: regulator-vddext-3v3 {
> > > +               compatible = "regulator-fixed";
> > > +               regulator-name = "VDDEXT_3V3";
> > > +               regulator-min-microvolt = <3300000>;
> > > +               regulator-max-microvolt = <300000>;
> > > +       };
> > 
> > Booting the dtb with this change will cause this regulator to be
> > disabled as there is no consumer for it.
> 
> Software cannot control this regulator by any means.  It's on once
> the board is powered on.  Do you mean it will be disabled in software
> level if no consumer?  If that's the case, is the disablement harmful?
> 
> > 
> > I suggest squashing both patches:
> 
> The patch looks self-contained to me.  Can we keep it as-is?

While I agree the patch is self-contained, squashing them will make it
easier to apply the change as a fix and cherry-pick the fix to stable
kernel.

Applied with squashing, thanks!

Shawn