Quoting Krzysztof Kozlowski (2024-02-11 05:39:36)

On 10/02/2024 08:09, Stephen Boyd wrote:

quoted

diff --git a/Documentation/devicetree/bindings/chrome/google,pogo-pin-connector.yaml b/Documentation/devicetree/bindings/chrome/google,pogo-pin-connector.yaml
new file mode 100644
index 000000000000..5ba68fd95fcd
--- /dev/null
+++ b/Documentation/devicetree/bindings/chrome/google,pogo-pin-connector.yaml

@@ -0,0 +1,61 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---

[...]

quoted

+properties:
+  compatible:
+    const: google,pogo-pin-connector
+
+  "#address-cells":
+    const: 1
+
+  "#size-cells":
+    const: 0
+
+  port:
+    $ref: /schemas/graph.yaml#/properties/port
+    description: Connection to USB2 port providing USB signals
+    required:
+      - endpoint

Drop required.

Why? I'd like to make it so you can't have the node defined without
connecting it up to the rest of the system. Is that bad?

quoted

+
+patternProperties:
+  "^keyboard@[0-9a-f]{1,2}$":
+    description: The detachable keyboard

If this is detachable why do you define it in DT? Only hard-wired USB
devices, which need some sort of special handling. are described in DT.

From the commit text:

 We expect to find a keyboard on the other side of this connector with a
 specific vid/pid, so describe that as a child device at the port of the
 usb device connected upstream.

ChromeOS userspace is checking that the connected device downstream of
this port has the expected vid/pid to quickly rule out USB keyboards
that aren't the detachable keyboard. I wanted to express this in DT so
that it didn't live in ChromeOS userspace forever.

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel