On Mon, May 07, 2018 at 11:46:59AM -0700, Bjorn Andersson wrote:

On Fri 04 May 17:10 PDT 2018, Dmitry Torokhov wrote:

quoted

Hi Tirupathi,

On Fri, Mar 23, 2018 at 11:53:12AM +0530, Tirupathi Reddy wrote:

quoted

Add resin key support to handle different types of key events
defined in different platforms.

Signed-off-by: Tirupathi Reddy <redacted>
---
 .../bindings/input/qcom,pm8941-pwrkey.txt          | 32 +++++++++
 drivers/input/misc/pm8941-pwrkey.c                 | 81 ++++++++++++++++++++++
 2 files changed, 113 insertions(+)

diff --git a/Documentation/devicetree/bindings/input/qcom,pm8941-pwrkey.txt b/Documentation/devicetree/bindings/input/qcom,pm8941-pwrkey.txt

[..]

quoted

quoted

 EXAMPLE
 
 	pwrkey@800 {

@@ -40,4 +66,10 @@ EXAMPLE
 		interrupts = <0x0 0x8 0 IRQ_TYPE_EDGE_BOTH>;
 		debounce = <15625>;
 		bias-pull-up;
+
+		resin {
+			interrupts = <0x0 0x8 1 IRQ_TYPE_EDGE_BOTH>;
+			linux,code = <KEY_VOLUMEDOWN>;
+			bias-pull-up;
+		};
 	};

The new key and power key bindings are very similar, I would prefer if
we shared the parsing code and our new DTS looked like:

		power {
			...
		};
 
		resin {
			...
		};

(we can easily keep backward compatibility with power properties being
in device node).

As discussed here https://patchwork.kernel.org/patch/9751627/ the PON
block does, in addition to providing power and resin key support also
handle the restart reason ("reboot bootloader" in Android).

My interpretation of our conclusion was to come up with a new binding
for the "pon" and a driver for this block that instantiates the power
key device.

It seems reasonable for such binding to describe the two keys, as you
propose here Dmitry, and instantiates the two input devices based on
this.

OK, I'll wait for the new driver and binding then.

Thanks.

-- 
Dmitry