From mboxrd@z Thu Jan 1 00:00:00 1970 From: Maxime Ripard Subject: Re: [RFC PATCH V3] ARM: dts: sun6i: Add dts file for MSI Primo81 tablet Date: Thu, 18 Jun 2015 13:58:38 +0200 Message-ID: <20150618115838.GL11732@lukather> References: <20150616094026.GG11732@lukather> <1434481869-9432-1-git-send-email-merker@debian.org> Reply-To: maxime.ripard-wi1+55ScJUtKEb57/3fJTNBPR1lH4CV8@public.gmane.org Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="yK/6QRnH3Zanb0EF" Return-path: Content-Disposition: inline In-Reply-To: <1434481869-9432-1-git-send-email-merker-8fiUuRrzOP0dnm+yROfE0A@public.gmane.org> List-Post: , List-Help: , List-Archive: , List-Unsubscribe: , To: Karsten Merker Cc: Siarhei Siamashka , linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org, devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-sunxi-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org List-Id: devicetree@vger.kernel.org --yK/6QRnH3Zanb0EF Content-Type: text/plain; charset=UTF-8 Content-Disposition: inline Hi, On Tue, Jun 16, 2015 at 09:11:08PM +0200, Karsten Merker wrote: > On Tue, Jun 16, 2015 at 11:40:27AM +0200, Maxime Ripard wrote: > > > What I meant in my previous review was to use the syntax > > > > &uart0 { > > some-property; > > }; > > > > Instead of duplicating the tree structure like you're doing > > here. We've converted all the DT to that, so you can look around and > > see how it's done in other boards (note that the nodes should be > > sorted by alphabetical order). > > Ah, sorry, I had misunderstood your original email in this regard. > Following is a reworked version of the patch. > > Changelog: > > Version 1 > ========= > - Original patch by Siarhei Siamashka > > Version 2 > ========= > - Use symbolic instead of numeric pinctrl values. > > - Change the include syntax from /include/ to #include to make > the dts build with current kernels. > > Version 3 > ========= > - Use labels for nodes with modifications in relation to the dtsi > instead of replicating the tree structure. > > - Remove the FSF address from the license header as done in > http://lists.infradead.org/pipermail/linux-arm-kernel/2015-May/340437.html > for the other dts files to remove a checkpatch warning. > > - Add msi to the vendor prefix list. > > - Remove earlyprintk from the default kernel commandline. > > - Replace the console kernel commandline parameter by a > /chosen/stdout-path node and add an alias serial0 -> uart0. Next time, pleas try to keep the changelog with your patch, it's much easier to read. And please make the vendor prefix change a different patch. > I have tagged this patch RFC as I am unsure what to do with the > /chosen/stdout-path node. For now, I have set Siarhei's original > choice (first serial port), but I am unsure whether this is the > right thing to do as the Primo81 does by default not have a > user-accessible serial port. The only way to get a serial > console is to either break the case open and find some test > points that carry the RX/TX lines (which with the Primo81 case > poses a high risk of breaking the display glass), or to use an SD > card breakout board and change the pinmuxing for the SD card pins > to the "serial" function. The latter would not work without > modifying the dts, so the SD-breakout case doesn't really count > for setting the default stdout-path in the general use case. Having that as a comment on top of whatever uart you've been using would be great, so that people know where to look at on their device to find it. Maxime -- Maxime Ripard, Free Electrons Embedded Linux, Kernel and Android engineering http://free-electrons.com --yK/6QRnH3Zanb0EF--