On Wednesday 23 November 2016 09:54 PM, David Lechner wrote:

On 11/23/2016 05:12 AM, Sekhar Nori wrote:

quoted

On Wednesday 23 November 2016 08:59 AM, David Lechner wrote:

quoted

This SoC has a separate pin controller for configuring pullup/pulldown
bias on groups of pins.

Signed-off-by: David Lechner <david@lechnology.com>
---
 arch/arm/boot/dts/da850.dtsi | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/arch/arm/boot/dts/da850.dtsi b/arch/arm/boot/dts/da850.dtsi
index 8945815..1c0224c 100644
--- a/arch/arm/boot/dts/da850.dtsi
+++ b/arch/arm/boot/dts/da850.dtsi

@@ -210,6 +210,11 @@
             };

         };
+        pinconf: pin-controller@22c00c {
+            compatible = "ti,da850-pupd";
+            reg = <0x22c00c 0x8>;
+            status = "disabled";
+        };

Can you please place this below the i2c1 node. I am trying to keep the
nodes sorted by unit address. I know thats broken in many places today,
but lets add the new ones where they should eventually end up.

I can do this, but it seems that the predominant sorting pattern here is
to keep subsystems together (e.g. all i2c are together, all uart are
together, etc.)

Yeah, but that quickly gives away as there are many singleton devices
and everyone tries to add theirs at the end of the list resulting in
merge conflicts.

Would a separate patch to sort everything by unit address to get this
cleaned up be acceptable?

Agree with Kevin that it would be churn. If done, it should be last
thing that gets done in a merge window. I would not attempt it in
relatively busy merge windows like this one.

Thanks,
Sekhar