On Tue, Jan 9, 2018 at 12:36 PM, Karthik Ramasubramanian <kramasub@xxxxxxxxxxxxxx> wrote: > > > On 1/2/2018 8:55 AM, Rob Herring wrote: >> >> On Wed, Dec 27, 2017 at 09:27:25AM -0700, Karthikeyan Ramasubramanian >> wrote: >>> >>> Add device tree binding support for GENI based UART Controller in the >>> QUP Wrapper. >>> >>> Signed-off-by: Karthikeyan Ramasubramanian <kramasub@xxxxxxxxxxxxxx> >>> Signed-off-by: Girish Mahadevan <girishm@xxxxxxxxxxxxxx> >>> --- >>> .../devicetree/bindings/serial/qcom,geni-uart.txt | 31 >>> ++++++++++++++++++++++ >>> 1 file changed, 31 insertions(+) >>> create mode 100644 >>> Documentation/devicetree/bindings/serial/qcom,geni-uart.txt >>> >>> diff --git a/Documentation/devicetree/bindings/serial/qcom,geni-uart.txt >>> b/Documentation/devicetree/bindings/serial/qcom,geni-uart.txt >>> new file mode 100644 >>> index 0000000..e60ec6a >>> --- /dev/null >>> +++ b/Documentation/devicetree/bindings/serial/qcom,geni-uart.txt >>> @@ -0,0 +1,31 @@ >>> +Qualcomm Technologies Inc. GENI based Serial UART Controller driver >>> + >>> +This serial UART driver supports console use-cases. This driver is meant >>> +only for Generic Interface (GENI) based Qualcomm Universal Peripheral >>> (QUP) >>> +cores and isn't backwards compatible. >>> + >>> +Required properties: >>> +- compatible: should contain "qcom,geni-uart, qcom,geni-console" >> >> >> Is console different programming model or just how you are using the >> h/w? for the latter, drop it as we have stdout-path to select a console. > > The console programming model is different from a regular UART port and > hence the compatible field contains console in it. And "console" is what the h/w reference manual calls it? If so, then it is fine. If not, sounds like a Linuxism. Rob
