From mboxrd@z Thu Jan 1 00:00:00 1970
From: Philipp Zabel
Subject: Re: [RFCv2 1/3] docs: dts: Added documentation for Xilinx Zynq
Reset Controller bindings.
Date: Tue, 28 Jul 2015 17:16:44 +0200
Message-ID: <1438096604.3193.51.camel@pengutronix.de>
References: <1437783682-13632-1-git-send-email-moritz.fischer@ettus.com>
<1437783682-13632-2-git-send-email-moritz.fischer@ettus.com>
<1438070702.3193.4.camel@pengutronix.de> <55B73C64.5050903@xilinx.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
Return-path:
In-Reply-To:
Sender: devicetree-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
To: Moritz Fischer
Cc: Michal Simek , robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org, pawel.moll-5wv7dgnIgG8@public.gmane.org, mark.rutland-5wv7dgnIgG8@public.gmane.org, ijc+devicetree-KcIKpvwj1kUDXYZnReoRVg@public.gmane.org, Kumar Gala , =?ISO-8859-1?Q?S=F6ren?= Brinkmann , linux-lFZ/pmaqli7XmaaqVzeoHQ@public.gmane.org, devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-arm-kernel , linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
List-Id: devicetree@vger.kernel.org
Hi Moritz, Michal,
Am Dienstag, den 28.07.2015, 06:57 -0700 schrieb Moritz Fischer:
[...]
> >>> +Example:
> >>> + rstc: rstc@240 {
> >>> + #reset-cells = <1>;
> >>> + compatible = "xlnx,zynq-reset-pl";
> >>> + syscon = <&slcr>;
> >>
> >> Why the syscon phandle if rstc always is the child of slcr? Why not just
> >> request the syscon for the rstc's parent node.
> >
> > We are using this description for pincntrl which was properly reviewed
> > that's why I expect Moritz just use the same style.
> > But yes also referencing parent should work.
>
> Michal is right, I tried to be consistent with the pinctrl. Either one
> is fine for me.
> We'll just have to make a decision :-)
Do you have a pointer to the pinctrl review? I'd like to know if
somebody had a good reason to use the phandle over the parent-child
relationship. I'd rather not add DT properties if they are not
necessary.
Regarding consistency, since the pinctrl node is also a child of the
slcr, you could just as well make the syscon phandle optional there and
remove it from the DT without breaking backwards compatibility.
> > TBH I don't have strong preference but having unified style is something
> > what I would prefer.
> >
> > Also I see that using parent is used by others and it looks like that
> > having something like syscon_regmap_lookup_parent will be worth to have.
That would be useful, yes.
regards
Philipp
--
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