On 23/01/2025 16:20, Johan Hovold wrote: > [Some people who received this message don't often get email from johan@xxxxxxxxxx. Learn why this is important at https://aka.ms/LearnAboutSenderIdentification ] > > This email is not from Hexagon’s Office 365 instance. Please be careful while clicking links, opening attachments, or replying to this email. > > > On Thu, Jan 23, 2025 at 02:01:00PM +0000, POPESCU Catalin wrote: >> On 23/01/2025 12:15, Johan Hovold wrote: >>> On Wed, Jan 22, 2025 at 02:47:32PM +0100, Catalin Popescu wrote: >>>> A simple usb device has a single configuration and a single interface >>>> and is considered as a "combined node" when defined in the devicetree. >>>> If available, its interface node is simply ignored which is a problem >>>> whenever the interface node has subnodes. To prevent that from happening >>>> first check for any subnode and ignore the interface node only if no >>>> subnode was found. >>>> >>>> Example: FTDI chip FT234XD that has only one UART interface which is >>>> being used as a serdev by other driver. >>>> >>>> device@1 { >>>> compatible = "usb403,6015"; >>>> reg = <1>; >>>> >>>> #address-cells = <2>; >>>> #size-cells = <0>; >>>> >>>> interface@0 { >>>> compatible = "usbif403,6015.config1.0"; >>> Your example makes no sense since if this is the only interface then the >>> interface node should not be here. >> That's the problem my patch is trying to address ... >> Why is it OK to describe multiple interfaces and it is not OK to >> describe the interface of a simple USB device ? > It's part of the spec. See commit 1a7e3948cb9f ("USB: add device-tree > support for interfaces") and its reference to "Open Firmware Recommended > Practice: Universal Serial Bus Version 1". I see, thanks a lot for the clarification. BR, > Johan
