RE: [PATCH] ARM: dts: am335x-boneblack: add i2c1 DT entry

From: Li, Yong B
Date: Fri Jun 03 2016 - 01:46:59 EST


Hi Tony,

I can submit a V2 version about the "disabled", or just drop this patch?

Thanks,
Yong
-----Original Message-----
From: Tony Lindgren [mailto:tony@xxxxxxxxxxx]
Sent: Friday, June 3, 2016 1:57 AM
To: Robert Nelson <robertcnelson@xxxxxxxxx>
Cc: Li, Yong B <yong.b.li@xxxxxxxxx>; Benoit Cousson <bcousson@xxxxxxxxxxxx>; Rob Herring <robh+dt@xxxxxxxxxx>; Pawel Moll <pawel.moll@xxxxxxx>; Mark Rutland <mark.rutland@xxxxxxx>; Ian Campbell <ijc+devicetree@xxxxxxxxxxxxxx>; Kumar Gala <galak@xxxxxxxxxxxxxx>; Russell King <linux@xxxxxxxxxxxxxxxx>; linux-omap@xxxxxxxxxxxxxxx; devicetree <devicetree@xxxxxxxxxxxxxxx>; linux-arm-kernel@xxxxxxxxxxxxxxxxxxx; linux kernel <linux-kernel@xxxxxxxxxxxxxxx>; sdliyong@xxxxxxxxx
Subject: Re: [PATCH] ARM: dts: am335x-boneblack: add i2c1 DT entry

* Robert Nelson <robertcnelson@xxxxxxxxx> [160602 10:48]:
> On Thu, Jun 2, 2016 at 12:39 PM, Tony Lindgren <tony@xxxxxxxxxxx> wrote:
> >
> > * Robert Nelson <robertcnelson@xxxxxxxxx> [160602 10:39]:
> > > On Thu, Jun 2, 2016 at 2:41 AM, Yong Li <yong.b.li@xxxxxxxxx> wrote:
> > >
> > > > From: Yong Li <sdliyong@xxxxxxxxx>
> > > >
> > > > Without this patch, I2C-1 is missing on beaglebone black boards
> > > >
> > > > Signed-off-by: Yong Li <sdliyong@xxxxxxxxx>
> > > > ---
> > > > arch/arm/boot/dts/am335x-boneblack.dts | 15 +++++++++++++++
> > > > 1 file changed, 15 insertions(+)
> > > >
> > >
> > >
> > > This is also 'common' for the white/green, if your really going to
> > > push this, please add it to:
> > >
> > > am335x-bone-common.dtsi
> >
> > But isn't this exposed on the cape pins? So the use can be whatever
> > for these pins?
>
> That is correct, "i2c1" is on the cape connector, with no primary
> function defined.
>
> Currently on mainline, the board and cape eeprom i2c bus's are enabled.
>
> i2c0 = board eeprom
> i2c2 = 'cape' eeprom
>
> personally, nak by me to add i2c1 by default, should use overlay's to enable it.

Yes sounds like the best we can do is define it but set it with status = "disabled" for now.

Regards,

Tony