From mboxrd@z Thu Jan 1 00:00:00 1970 From: Carlo Caione Subject: Re: [PATCH 2/4] pinctrl: amlogic: Make driver independent from two-domain configuration Date: Thu, 17 Mar 2016 08:09:28 +0100 Message-ID: References: <1456869876-19320-1-git-send-email-carlo@caione.org> <1456869876-19320-3-git-send-email-carlo@caione.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Return-path: In-Reply-To: Sender: devicetree-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Kevin Hilman Cc: Carlo Caione , Linus Walleij , Beniamino Galvani , linux-arm-kernel , linux-meson , Rob Herring , devicetree , linux-6IF/jdPJHihWk0Htik3J/w@public.gmane.org, Arnd Bergmann , Carlo Caione , Tyler Baker List-Id: devicetree@vger.kernel.org On Thu, Mar 17, 2016 at 4:31 AM, Kevin Hilman wrote: > Hi Carlo > > On Tue, Mar 1, 2016 at 2:04 PM, Carlo Caione wrote: >> From: Carlo Caione >> >> In the Amlogic Meson8 / Meson8b platforms we have two different buses: >> cbus and aobus, corresponding to 2 different power domains (regular and >> always-on). On each bus a different set of registers is mapped to manage >> muxes, GPIOs and in general to control a clear subset of the pins. >> >> Considering this architecture, having two different pinctrl devices, one >> for each bus / power domain, makes much more sense than just having one >> single device. >> >> Right now we have one single pin controller driver that uses two >> different domains (represented by 'gpio' and 'gpio-ao' sub-nodes in the >> DTS) to manage the set of registers on the two buses. This dual-domain >> configuration is hardcoded into the driver that strictly requires one >> domain for each bus in the same pin controller device. >> >> With this patch we refactor the driver to allow splitting the driver in >> two parts. This change is needed to have a proper description of the HW >> in the device-tree where we want to introduce aobus and cbus. >> >> Signed-off-by: Carlo Caione > > kernelci.org detected that the meson8b-odroidc1 started failing boot > test in mainline[1] and I bisected it down to this patch, which is in > mainline in he form of commit 9dab1868ec0d (pinctrl: amlogic: Make > driver independent from two-domain configuration.) > > I confirmed that reverting this patch on top of Linus' master branch > (commit 9256d5a308c9) gets the odroid-c1 booting again. Yes, this is expected. Thank you for pointing this out. The problem is that the driver changes ended up in mainline before I could pull the DTS changes (waiting for the Rob's ACK on documentation). What's the fastest way to land the DTS changes in mainline at this point? Thanks, -- Carlo Caione -- 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 From mboxrd@z Thu Jan 1 00:00:00 1970 From: carlo@caione.org (Carlo Caione) Date: Thu, 17 Mar 2016 08:09:28 +0100 Subject: [PATCH 2/4] pinctrl: amlogic: Make driver independent from two-domain configuration In-Reply-To: References: <1456869876-19320-1-git-send-email-carlo@caione.org> <1456869876-19320-3-git-send-email-carlo@caione.org> Message-ID: To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Thu, Mar 17, 2016 at 4:31 AM, Kevin Hilman wrote: > Hi Carlo > > On Tue, Mar 1, 2016 at 2:04 PM, Carlo Caione wrote: >> From: Carlo Caione >> >> In the Amlogic Meson8 / Meson8b platforms we have two different buses: >> cbus and aobus, corresponding to 2 different power domains (regular and >> always-on). On each bus a different set of registers is mapped to manage >> muxes, GPIOs and in general to control a clear subset of the pins. >> >> Considering this architecture, having two different pinctrl devices, one >> for each bus / power domain, makes much more sense than just having one >> single device. >> >> Right now we have one single pin controller driver that uses two >> different domains (represented by 'gpio' and 'gpio-ao' sub-nodes in the >> DTS) to manage the set of registers on the two buses. This dual-domain >> configuration is hardcoded into the driver that strictly requires one >> domain for each bus in the same pin controller device. >> >> With this patch we refactor the driver to allow splitting the driver in >> two parts. This change is needed to have a proper description of the HW >> in the device-tree where we want to introduce aobus and cbus. >> >> Signed-off-by: Carlo Caione > > kernelci.org detected that the meson8b-odroidc1 started failing boot > test in mainline[1] and I bisected it down to this patch, which is in > mainline in he form of commit 9dab1868ec0d (pinctrl: amlogic: Make > driver independent from two-domain configuration.) > > I confirmed that reverting this patch on top of Linus' master branch > (commit 9256d5a308c9) gets the odroid-c1 booting again. Yes, this is expected. Thank you for pointing this out. The problem is that the driver changes ended up in mainline before I could pull the DTS changes (waiting for the Rob's ACK on documentation). What's the fastest way to land the DTS changes in mainline at this point? Thanks, -- Carlo Caione