[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87bmzzr1bg.fsf@eliezer.anholt.net>
Date: Wed, 07 Sep 2016 11:50:59 -0700
From: Eric Anholt <eric@...olt.net>
To: Stefan Wahren <stefan.wahren@...e.com>,
Gerd Hoffmann <kraxel@...hat.com>,
linux-rpi-kernel@...ts.infradead.org,
linux-arm-kernel@...ts.infradead.org, devicetree@...r.kernel.org
Cc: Scott Branden <sbranden@...adcom.com>, Ray Jui <rjui@...adcom.com>,
Russell King <linux@...linux.org.uk>,
open list <linux-kernel@...r.kernel.org>,
Rob Herring <robh+dt@...nel.org>,
bcm-kernel-feedback-list@...adcom.com, swarren@...dotorg.org,
Mark Rutland <mark.rutland@....com>,
Florian Fainelli <f.fainelli@...il.com>
Subject: Re: [PATCH 1/8] ARM: dts: bcm283x: Define standard pinctrl groups in the gpio node.
Stefan Wahren <stefan.wahren@...e.com> writes:
> Hi Gerd,
>
>> Gerd Hoffmann <kraxel@...hat.com> hat am 7. September 2016 um 12:31
>> geschrieben:
>>
>>
>> From: Eric Anholt <eric@...olt.net>
>>
>> The BCM2835-ARM-Peripherals.pdf documentation specifies what the
>> function selects do for the pins, and there are a bunch of obvious
>> groupings to be made. With these created, we'll be able to replace
>> bcm2835-rpi.dtsi's main "set all of these pins to alt0" with
>> references to specific groups we want enabled.
>
> on IMX/MXS platform it's unwanted to add all possible muxes in the dtsi file. So
> i would suggest to add only the actually used muxes. That makes it easier to
> maintain.
On the other hand, I find that having to go back to the docs for
determining the fsels is worse than just verifying and defining them
here all at once. Maintaining has also gotten harder because our DTs
are split across 32 and 64-bit ARM, so rpi3 changes that require adding
one of these pinmux definitions back to the dtsi would require painful
cross-branch merges.
Download attachment "signature.asc" of type "application/pgp-signature" (801 bytes)
Powered by blists - more mailing lists