On Wed, 2017-07-26 at 15:55 +0200, Yegor Yefremov wrote: > On Wed, Jul 26, 2017 at 3:35 PM, Andy Shevchenko > <andriy.shevchenko@xxxxxxxxxxxxxxx> wrote: > > On Wed, 2017-07-26 at 15:26 +0200, Yegor Yefremov wrote: > > The problen now while property _is_ there in ACPI table. > > > > > || > > > + of_property_read_bool(dev- > > > >of_node, > > > "wakeup-source")) > > > > ...and this should be device_property besides that fact that you > > need to > > check it once in _noauto() and forbid requesting GPIOs by returning > > an > > error immediately. > > Something like that: Precisely. > @@ -118,6 +119,9 @@ struct mctrl_gpios *mctrl_gpio_init_noauto(struct > device *dev, unsigned int idx) > struct mctrl_gpios *gpios; > enum mctrl_gpio_idx i; > > + if (device_property_present(dev, "wakeup-source")) > + return ERR_PTR(-EINVAL); ...and this patch should be in _before_ the one which enables mctrl for 8250. > > P.S. Take into account that use of this property should be described > > in > > the bindings. Before that it should be discussed (I pointed above > > what > > might be cons of use this name/property). > > OK. Let's wait for suggestions. > I would recommend to prepare and send a formal patch with Cc list of all stakeholders (some developers from serial, DT, ARM, x86). In a commit message would be good to explain we are trying to distinguish the purpose of GPIOs in UART description (DT or ACPI or even platform code [via built-in device properties]). -- Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx> Intel Finland Oy -- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
