Silvio, Marc, On Sat, Dec 13, 2014 at 09:21:23AM +0000, Marc Zyngier wrote: > On Sat, Dec 13 2014 at 1:46:45 am GMT, Silvio Fricke <silvio.fricke@xxxxxxxxx> wrote: > > Hi Silvio, > > > I have found some dts entries which are not evaluated by the drivers. This > > patch remove this entries from the dts files. > > Jason has mentioned I should CC: Thomas, Marc and him self to this > > mails. > > As far as I can tell, this looks correct. A few key things though: > > - Please write decent commit logs. Indicate *why* you think these > properties can be removed (hint: not finding corresponding in the > drivers is not enough a reason, the binding itself matters). Right. The real issue is the state of these properties within the binding documentation. Linux isn't the only consumer of these dts files (*BSD, barebox, etc), so us not using them doesn't necessarily mean the property should be removed. The most important lesson (recently learned ;-) ) is to make sure all properties are accurately describing the *hardware*. Not Linux's implementation of talking to the hardware. That'll go a long way if the changelog discusses how property X is an implementation detail vice a hardware characteristic or feature. > - Cc the relevant platform maintainers. You're changing things that they > care about, basic courtesy is to keep them in the loop. True. In Silvio's defense, I just told him to Cc us to get the ball rolling. (I tend to dislike long discussions on chat mediums, I'd much rather they be in email. So I was definitely too succinct). For the record, you can use ./scripts/get_maintainer.pl on your patches to make sure you get most of the right people. In this case, that'll include the DT maintainers. You may want to run a get_maintainer.pl -f arch/arm/mach-$ARCH for each of the dts file you touch. Those maintainers usually wrangle the patches for their dts files. Unfortunately, not all of them have updated MAINTAINERS (<cough>mvebu<cough>) to include the dts file patterns. thx, Jason. -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
