From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vladimir Zapolskiy Subject: Re: [PATCH v3] ARM: dts: add gpio-ranges property to iMX GPIO controllers Date: Fri, 9 Sep 2016 11:44:10 +0300 Message-ID: <729f2fd7-df40-c631-5365-892ee25854b3@mentor.com> References: <1473386556-1934-1-git-send-email-vladimir_zapolskiy@mentor.com> <20160909072020.ieehalkgxxjfewhs@pengutronix.de> Mime-Version: 1.0 Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 8bit Return-path: In-Reply-To: <20160909072020.ieehalkgxxjfewhs-bIcnvbaLZ9MEGnE8C9+IrQ@public.gmane.org> Sender: devicetree-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: =?UTF-8?Q?Uwe_Kleine-K=c3=b6nig?= Cc: Shawn Guo , Sascha Hauer , Fabio Estevam , Mark Rutland , devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Russell King , Rob Herring , Deepak Das , linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org List-Id: devicetree@vger.kernel.org Hi Uwe, On 09/09/2016 10:20 AM, Uwe Kleine-König wrote: > On Fri, Sep 09, 2016 at 05:02:36AM +0300, Vladimir Zapolskiy wrote: >> To establish a connection between GPIO controllers and pin multiplexor >> controller add gpio-ranges properties to all GPIO controllers found >> on iMX50, iMX6Q/D, iMX6DL/S, iMX6SL, iMX6SX, iMX6UL and iMX7D/S >> SoCs. The change was done after human parsing of output from > > When looking at doing the same for i.MX35 there are a few pin pairs that > are driven by the same GPIO, e.g.: > > $ grep GPIO3_9 imx35-pinfunc.h > #define MX35_PAD_CTS1__GPIO3_9 0x194 0x5d8 0x918 0x5 0x0 > #define MX35_PAD_FEC_COL__GPIO3_9 0x2ec 0x750 0x918 0x5 0x1 > > With this it's not possible to correctly add a gpio-ranges, right? this is an interesting exceptional case, I don't think that pinctrl and gpio subsystems are ready to handle it. If both pads are muxed to the same GPIO, then I presume on output it behaves like a tee, but do you know what would be the input value, if one line is driven high and another one is low? > i.MX1, i.MX21, i.MX25, i.MX27 and i.MX53 could benefit from such a > change though. i.MX35 and i.MX51 have duplicates. > I keep my hands off it, because I don't have any boards powered by these SoCs to be able to test the change. FWIW I hold one i.MX31 powered board from old times, but I see that i.MX31 has poor DTS support in general, there is no GPIO controller device node to add the property. -- With best wishes, Vladimir -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html