Hi Quentin,

On 1/12/2026 10:00 PM, Quentin Schulz wrote:

Hi Chaoyi,

On 1/12/26 3:28 AM, Chaoyi Chen wrote:

quoted

From: Chaoyi Chen <redacted>

The WCH CH334/CH335[0] are USB2.0 protocol compliant 4-port USB HUB
controller chips, supporting USB2.0 high-speed and full-speed for
upstream ports, and USB2.0 high-speed 480Mbps, full-speed 12Mbps and
low-speed 1.5Mbps for downstream ports, supporting not only low-cost STT
mode (single TT schedules 4 downstream ports in time share), but also
supports high performance MTT mode (4 TTs each corresponding to 1 port,
concurrent processing).

Add a device tree binding for it.

[0]: https://www.wch-ic.com/downloads/CH334DS1_PDF.html

Signed-off-by: Chaoyi Chen <redacted>
---
  .../devicetree/bindings/usb/wch,ch334.yaml    | 65 +++++++++++++++++++
  1 file changed, 65 insertions(+)
  create mode 100644 Documentation/devicetree/bindings/usb/wch,ch334.yaml

diff --git a/Documentation/devicetree/bindings/usb/wch,ch334.yaml b/Documentation/devicetree/bindings/usb/wch,ch334.yaml
new file mode 100644
index 000000000000..2eeb92f25b4c
--- /dev/null
+++ b/Documentation/devicetree/bindings/usb/wch,ch334.yaml

@@ -0,0 +1,65 @@
+# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/usb/wch,ch334.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: WCH CH334/CH335 USB 2.0 Hub Controller
+
+maintainers:
+  - Chaoyi Chen <kernel@airkyi.com>
+
+allOf:
+  - $ref: usb-hub.yaml#
+
+properties:
+  compatible:
+    enum:
+      - usb1a86,8091
+

Which driver does this node bind to? I couldn't quickly find a driver which would match this compatible?

Oh, I missed that part. It will be fixed in the next version.

quoted

+  reg: true
+
+  reset-gpios:
+    description: GPIO controlling the RESET# pin.
+
+  vdd-supply:
+    description:
+      The regulator that provides 3.3V core power to the hub.
+
+  vdd2-supply:
+    description:
+      The regulator that provides 3.3V or 5V power to the hub.
+

There's v5 and vdd33 as power input, shouldn't we maybe use those names instead? How did you come up with vdd/vdd2?

That make sense. Will fix in next version.

There's also a timing that needs to be respected after a power-on event so that the reset has enough time to be performed, c.f. 3.2.1 Power-on Reset in the datasheet you linked to in the commit log. How are you making sure we wait those (apparently, the wording in the datasheet is confusing) 14ms?

This part should be described in the driver. 
I'll fix it in the next version.

-- 
Best, 
Chaoyi