From mboxrd@z Thu Jan 1 00:00:00 1970 From: ulli.kroll@googlemail.com (Hans Ulli Kroll) Date: Tue, 25 Apr 2017 21:07:47 +0200 (CEST) Subject: [PATCH 1/5 v3] usb: host: add DT bindings for faraday fotg2 In-Reply-To: References: <20170421204058.6206-1-linus.walleij@linaro.org> Message-ID: To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Hi Linus On Tue, 25 Apr 2017, Linus Walleij wrote: > On Mon, Apr 24, 2017 at 6:53 PM, Hans Ulli Kroll > wrote: > > > Got NAK'ed from Rob on some ealier round due missing "device mode" on this > > IP. I've blatantly overrided this to a host only driver. > > > > These are the needed changes in DT to support both modes > > Note the -dr at the end of fotg210, to reflect this in an dual role device > > OK I understood the discussion such that the compatible should > simply be ""faraday,fotg210" as that is the name of the hardware > IP block. This is the name of the hardware name in the Faraday > page: > http://www.faraday-tech.com/html/Product/IPProduct/InterfaceIP/USB2_0.htm > > Any other string implies how it is used, so that was what I understood > as the reason to reject it with the "-hcd" (host controller device) suffix. > > > +- dr_mode : indicates the working mode for "fotg210-dr" compatible > > + controllers. Can be "host", "peripheral". Default to > > + "host" if not defined for backward compatibility. > > This seems right, so it is part of the generic bindings, correct? > > > usb at 68000000 { > > - compatible = "cortina,gemini-usb", "faraday,fotg210"; > > + compatible = "cortina,gemini-usb", "faraday,fotg210-dr"; > > But this would be wrong, because the compatible should only > indicate what kind of hardware it is, not how it is going to be used > (whether as host only, slave only or dual-role (OTG). > for compatible I think yes. But in Rob's opinion we missed the device part of the controller. Greetings Hans Ulli Kroll